pdfix-cli 9.1.1__py3-none-win_amd64.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- pdfix_cli/__init__.py +23 -0
- pdfix_cli/__main__.py +30 -0
- pdfix_cli/_paths.py +103 -0
- pdfix_cli/_runner.py +64 -0
- pdfix_cli/runtime/LICENSE.txt +4 -0
- pdfix_cli/runtime/README.txt +1 -0
- pdfix_cli/runtime/bin/x86_64/LicenseSpringVMD.dll +0 -0
- pdfix_cli/runtime/bin/x86_64/pdf.dll +0 -0
- pdfix_cli/runtime/bin/x86_64/pdfix_app.exe +0 -0
- pdfix_cli/runtime/resources/fix_metadata.json +98 -0
- pdfix_cli/runtime/resources/make_accessible.json +739 -0
- pdfix_cli/runtime/resources/pdf_action.json +3959 -0
- pdfix_cli/runtime/resources/pdf_config.json +4034 -0
- pdfix_cli/runtime/resources/pdfix_settings.json +38 -0
- pdfix_cli/runtime/resources/quick_fix.json +125 -0
- pdfix_cli/runtime/version.txt +1 -0
- pdfix_cli-9.1.1.dist-info/METADATA +121 -0
- pdfix_cli-9.1.1.dist-info/RECORD +22 -0
- pdfix_cli-9.1.1.dist-info/WHEEL +5 -0
- pdfix_cli-9.1.1.dist-info/entry_points.txt +4 -0
- pdfix_cli-9.1.1.dist-info/licenses/LICENSE +4 -0
- pdfix_cli-9.1.1.dist-info/top_level.txt +1 -0
|
@@ -0,0 +1,4034 @@
|
|
|
1
|
+
{
|
|
2
|
+
"@metadata<desc>": "Document-level fields describing this template (author, version, dates, SDK, free-form notes).",
|
|
3
|
+
"metadata": {
|
|
4
|
+
"author": "name",
|
|
5
|
+
"version": "2.0",
|
|
6
|
+
"created": "yyyy/mm/dd",
|
|
7
|
+
"modified": "yyyy/mm/dd",
|
|
8
|
+
"sdk_version": "major.minor.patch",
|
|
9
|
+
"notes": ""
|
|
10
|
+
},
|
|
11
|
+
"@settings<desc>": "Global switches that apply across tagging and pagemap (RTL, Form XObject handling, optional page struct type, debug breakpoints).",
|
|
12
|
+
"settings": {
|
|
13
|
+
"@rtl<type>": "bool",
|
|
14
|
+
"@rtl<desc>": "If true, layout and structure output follow right-to-left reading conventions where applicable (e.g. table column order).",
|
|
15
|
+
"@rtl<use>": "Enable for Arabic, Hebrew, or mixed RTL documents; leave false for standard LTR.",
|
|
16
|
+
"@rtl<value>": [
|
|
17
|
+
"true",
|
|
18
|
+
"false"
|
|
19
|
+
],
|
|
20
|
+
"rtl": false,
|
|
21
|
+
"@substructure_form_xobject<type>": "bool",
|
|
22
|
+
"@substructure_form_xobject<desc>": "Default for Form XObjects: false keeps the form as one object; true expands and tags inner content as separate structure elements.",
|
|
23
|
+
"@substructure_form_xobject<use>": "Override per-object in object_update when some forms should stay atomic and others should expose inner text and graphics.",
|
|
24
|
+
"@substructure_form_xobject<value>": [
|
|
25
|
+
"true",
|
|
26
|
+
"false"
|
|
27
|
+
],
|
|
28
|
+
"substructure_form_xobject": true,
|
|
29
|
+
"@page_tag<type>": "string",
|
|
30
|
+
"@page_tag<desc>": "Structure type name emitted for the page node when the engine wraps page content (e.g. NonStruct). If empty, children attach without an explicit page wrapper tag.",
|
|
31
|
+
"@page_tag<use>": "Change only when your role map or downstream consumer expects a specific page-level standard type.",
|
|
32
|
+
"page_tag": "NonStruct",
|
|
33
|
+
"@debug_pagemap_stop<type>": "string",
|
|
34
|
+
"@debug_pagemap_stop<desc>": "Developer-only: set to a pipeline stage name to halt pagemap immediately after that step. Values match the stage list below (for example, word_neighbours after neighbour-based line grouping, word_connect after connecting text runs into lines).",
|
|
35
|
+
"@debug_pagemap_stop<use>": "Use to bisect layout bugs; leave empty in production. Each value corresponds to one early-return checkpoint in pagemap or text-line processing.",
|
|
36
|
+
"@debug_pagemap_stop<value>": [
|
|
37
|
+
"element_create",
|
|
38
|
+
"object_update",
|
|
39
|
+
"annot_update",
|
|
40
|
+
"line_update",
|
|
41
|
+
"rect_update",
|
|
42
|
+
"object_update_text",
|
|
43
|
+
"word_update",
|
|
44
|
+
"word_neighbours",
|
|
45
|
+
"word_connect",
|
|
46
|
+
"text_line_update",
|
|
47
|
+
"text_create",
|
|
48
|
+
"text_split",
|
|
49
|
+
"text_update",
|
|
50
|
+
"table_graphic_detect",
|
|
51
|
+
"image_detect",
|
|
52
|
+
"table_graphic_update",
|
|
53
|
+
"form_field_update",
|
|
54
|
+
"element_update",
|
|
55
|
+
"label_join",
|
|
56
|
+
"element_artifact",
|
|
57
|
+
"container_nest",
|
|
58
|
+
"image_nest",
|
|
59
|
+
"table_text_detect",
|
|
60
|
+
"image_join",
|
|
61
|
+
"element_nest",
|
|
62
|
+
"element_parent_update",
|
|
63
|
+
"table_update",
|
|
64
|
+
"element_finalize",
|
|
65
|
+
"table_form_update",
|
|
66
|
+
"artifact_update",
|
|
67
|
+
"label_update",
|
|
68
|
+
"list_detect",
|
|
69
|
+
"list_update"
|
|
70
|
+
],
|
|
71
|
+
"debug_pagemap_stop": ""
|
|
72
|
+
},
|
|
73
|
+
"@template<desc>": "Root object: pagemap thresholds, regex tables, element templates, and optional tagging rules that drive layout detection and PDF structure output.",
|
|
74
|
+
"template": {
|
|
75
|
+
"@pagemap<type>": "unknown",
|
|
76
|
+
"@pagemap<desc>": "Numeric and boolean thresholds for pagemap (tables, labels, text layout, artifacts, reading order, etc.), usually under a $if query.",
|
|
77
|
+
"@pagemap<use>": "Tune one pagemap block per document profile; pair with pagemap_regex for pattern-driven classification before word_update and line building.",
|
|
78
|
+
"@pagemap<insert>": [
|
|
79
|
+
{}
|
|
80
|
+
],
|
|
81
|
+
"pagemap": [
|
|
82
|
+
{
|
|
83
|
+
"statement": "$if",
|
|
84
|
+
"@query": {},
|
|
85
|
+
"@preflight_artifact_w1<desc>": "Weight of horizontal bbox alignment differences when comparing two header/footer samples across pages.",
|
|
86
|
+
"@preflight_artifact_w1<use>": "Raise when x-alignment should dominate the distance; lower if horizontal jitter should not break repeat detection.",
|
|
87
|
+
"preflight_artifact_w1": 1,
|
|
88
|
+
"@preflight_artifact_w2<desc>": "Weight of vertical bbox alignment differences in the same preflight comparison.",
|
|
89
|
+
"@preflight_artifact_w2<use>": "Raise when band height must match tightly; lower if vertical drift across pages is acceptable.",
|
|
90
|
+
"preflight_artifact_w2": 1,
|
|
91
|
+
"@preflight_artifact_w3<desc>": "For non-text objects, weight of width mismatch; for text pairs this term is unused (font checks apply instead).",
|
|
92
|
+
"@preflight_artifact_w3<use>": "Tune graphic repeat matching; ignored for text-to-text pairs.",
|
|
93
|
+
"preflight_artifact_w3": 1,
|
|
94
|
+
"@preflight_artifact_w4<desc>": "For text, weight of font-size difference; for graphics, weight of height mismatch between samples.",
|
|
95
|
+
"@preflight_artifact_w4<use>": "Raise to penalize mixed font sizes or mismatched graphic heights in the same cluster.",
|
|
96
|
+
"preflight_artifact_w4": 1,
|
|
97
|
+
"@preflight_artifact_w5<desc>": "Weight of numeric token / page-index stride consistency when matching running heads that include page numbers.",
|
|
98
|
+
"@preflight_artifact_w5<use>": "Raise when page-number progression should strongly decide whether two headers belong together.",
|
|
99
|
+
"preflight_artifact_w5": 1,
|
|
100
|
+
"@preflight_artifact_distance<desc>": "DBSCAN epsilon on the weighted preflight distance between header/footer or artifact samples across pages (same units as multivariate_distance output from alignment, size, and page-number terms).",
|
|
101
|
+
"@preflight_artifact_distance<use>": "Larger values merge more loosely aligned repeats into one band; smaller values require tighter match and yield more clusters.",
|
|
102
|
+
"preflight_artifact_distance": 0.7,
|
|
103
|
+
"@preflight_artifact_cluster_points<desc>": "Minimum DBSCAN cluster size (number of page samples) to accept a repeating header, footer, or artifact stripe. If 0, the engine uses about num_pages/10 as the default minimum.",
|
|
104
|
+
"@preflight_artifact_cluster_points<use>": "Higher values demand evidence on more pages before a band is confirmed; lower values catch short documents’ repeats but risk noise.",
|
|
105
|
+
"preflight_artifact_cluster_points": 2,
|
|
106
|
+
"@concurrent_threads<desc>": "Worker thread count for parallel pagemap steps (e.g. distance-matrix fills in image/label/section clustering). 0 lets the runtime pick an implementation default; 1 disables parallelism.",
|
|
107
|
+
"@concurrent_threads<use>": "Use 0 on multi-core machines for speed; set 1 to debug determinism or reduce memory spikes; use a small cap (2–4) to limit load.",
|
|
108
|
+
"concurrent_threads": 0,
|
|
109
|
+
"@text_only<desc>": "If 1, pagemap skips vector line/rect extension and graphic-object layout passes; only text extraction and text-driven structure run.",
|
|
110
|
+
"@text_only<use>": "Enable for fast text-only pipelines; keep 0 when tables, figures, or path-based labels must be detected.",
|
|
111
|
+
"text_only": 0,
|
|
112
|
+
"@rotation_detect<desc>": "If 1 (or during preflight), logical page rotation is detected from content and the working page rotation may be adjusted for reading order.",
|
|
113
|
+
"@rotation_detect<use>": "Disable if pages with intentional 90° artwork must not be auto-straightened; keep enabled for scanned or wrongly rotated office PDFs.",
|
|
114
|
+
"rotation_detect": 1,
|
|
115
|
+
"@background_color_red<desc>": "Assumed page background RGB red component (0–255) for matching filled rectangles to the paper color.",
|
|
116
|
+
"@background_color_red<use>": "Set together with green/blue to the dominant paper tint so large same-color panels classify as background, not content.",
|
|
117
|
+
"background_color_red": 255,
|
|
118
|
+
"@background_color_green<desc>": "Assumed page background RGB green component (0–255).",
|
|
119
|
+
"@background_color_green<use>": "See background_color_red; adjust per channel for tinted stock.",
|
|
120
|
+
"background_color_green": 255,
|
|
121
|
+
"@background_color_blue<desc>": "Assumed page background RGB blue component (0–255).",
|
|
122
|
+
"@background_color_blue<use>": "See background_color_red; critical for blue-gray office PDFs.",
|
|
123
|
+
"background_color_blue": 255,
|
|
124
|
+
"@background_color_diff<desc>": "Maximum per-channel difference from the assumed background RGB still treated as background when classifying rects (tolerance for off-white).",
|
|
125
|
+
"@background_color_diff<use>": "Increase if subtle tinted panels are missed as background; decrease if non-background fills are mistaken for the page color.",
|
|
126
|
+
"background_color_diff": 2,
|
|
127
|
+
"@bbox_expansion<desc>": "Extra padding in points (half-width inflate/deflate) for overlap tests: reading-order bands (e.g. TOC), nesting content into table cells, multi-line text split across cells, and table vs. image overlap.",
|
|
128
|
+
"@bbox_expansion<use>": "Increase slightly if geometry that should connect (lines, cells, images) fails hit tests; decrease if unrelated elements merge into one cell or band.",
|
|
129
|
+
"bbox_expansion": 2,
|
|
130
|
+
"@angle_deviation<desc>": "Maximum angular deviation (radians) allowed when treating two elements (e.g. text runs) as parallel on the same line, and for stroke/alignment classification.",
|
|
131
|
+
"@angle_deviation<use>": "Slightly increase for scanned or skewed pages; decrease if nearly-horizontal noise is treated as structure.",
|
|
132
|
+
"angle_deviation": 0.015707963267949,
|
|
133
|
+
"@header_ratio<desc>": "Upper bound as a fraction of page height [0,1] for the band considered when detecting header regions.",
|
|
134
|
+
"@header_ratio<use>": "Increase for tall running heads or letterheads; decrease if body text near the top is mistaken for header.",
|
|
135
|
+
"header_ratio": 0.15,
|
|
136
|
+
"@footer_ratio<desc>": "Upper bound as a fraction of page height [0,1] for the band considered when detecting footer regions.",
|
|
137
|
+
"@footer_ratio<use>": "Increase for multi-line footers; decrease if low page content is pulled into footer heuristics.",
|
|
138
|
+
"footer_ratio": 0.15,
|
|
139
|
+
"@artifact_w1<desc>": "Weight of the border-distance term in automatic artifact similarity (how much proximity to the page edge counts).",
|
|
140
|
+
"@artifact_w1<use>": "Set to 0 to ignore border distance; increase to flag edge-near content as artifact more strongly.",
|
|
141
|
+
"artifact_w1": 1,
|
|
142
|
+
"@artifact_w2<desc>": "Weight of image area relative to the page when scoring images as artifacts.",
|
|
143
|
+
"@artifact_w2<use>": "Set to 0 to skip area-based artifact bias for images; increase to treat large page-filling images as decorative more often.",
|
|
144
|
+
"artifact_w2": 1,
|
|
145
|
+
"@artifact_font_size<desc>": "Preflight only (preflight_elements): font size in points. Horizontal text with size less than or equal to this value is not added to the artifact sample set; any non-horizontal text is always added, and horizontal text strictly larger than this value is added.",
|
|
146
|
+
"@artifact_font_size<use>": "Raise to keep body-sized horizontal lines out of artifact correlation; lower so large headings and display type are more often sampled as repeats.",
|
|
147
|
+
"artifact_font_size": 40,
|
|
148
|
+
"@artifact_border_distance_max<desc>": "Distance in points from the page edge within which border proximity still contributes to artifact similarity.",
|
|
149
|
+
"@artifact_border_distance_max<use>": "Increase (e.g. 10–20) if legitimate body text near margins is marked artifact; decrease for stricter edge-only detection.",
|
|
150
|
+
"artifact_border_distance_max": 2,
|
|
151
|
+
"@artifact_similarity<desc>": "Similarity threshold [0,1] above which an element is treated as an artifact in automatic artifact detection passes.",
|
|
152
|
+
"@artifact_similarity<use>": "Raise to require stronger evidence before classifying as artifact; lower if decorative repeats (headers, watermarks) are missed.",
|
|
153
|
+
"artifact_similarity": 0.7,
|
|
154
|
+
"@artifact_untagged<desc>": "If set to 1, elements without an initial parent are automatically classified as artifacts.",
|
|
155
|
+
"@artifact_untagged<use>": "Enable only in word_neighbours-style template workflows where all meaningful content is expected to sit under an initial element. When enabled, anything outside those anchors is treated as artifact. Keep disabled for general layout or untagged PDFs so stray body content is not dropped.",
|
|
156
|
+
"artifact_untagged": 0,
|
|
157
|
+
"@path_object_max<desc>": "Maximum number of sub-paths in a compound path before it may be split into separate paths. If exceeded, the path is kept intact (processed as a complex/non-simple path).",
|
|
158
|
+
"@path_object_max<use>": "Performance & complexity guard for vector graphics. Lower value → fewer splits (faster), more paths stay “complex” (risk: worse detection of simple lines/rectangles inside complex drawings). Higher value → more splits (more precise line/rect detection from vectors), but can slow down processing on detailed graphics (logos, charts, CAD-like drawings).",
|
|
159
|
+
"path_object_max": 2000,
|
|
160
|
+
"@path_object_min<desc>": "Sub-path count threshold that enables an “isolation” check before splitting. If a path object has more than path_object_min sub-paths, it is split only when it is not considered isolated (based on bbox size vs isolated_text).",
|
|
161
|
+
"@path_object_min<use>": "Helps avoid splitting dense small vector clusters (often noise), while still allowing splitting of larger vector regions. When the sub-path count is above path_object_min, splitting happens only if the path’s bbox is larger than isolated_text in both width and height. Increase to make splitting more permissive (fewer cases fall into the isolation-gated region). Decrease to apply the isolation gate earlier (useful if small dense vector blobs are incorrectly expanded into many tiny lines or rectangles).",
|
|
162
|
+
"path_object_min": 100,
|
|
163
|
+
"@initial_element_only<desc>": "If set to 1, pagemap processes only elements explicitly marked/detected as “initial elements”; all other detected elements are ignored.",
|
|
164
|
+
"@initial_element_only<use>": "Template-driven “strict mode”. Enable (1) when you want pagemap to respect only your initial anchors/initial elements and avoid extra auto-detected structure. Keep 0 for normal layout recognition (recommended), otherwise you may lose tables/paragraphs that are not covered by initial elements.",
|
|
165
|
+
"initial_element_only": 0,
|
|
166
|
+
"@initial_elements_keep_empty<desc>": "Specifies whether to preserve initial elements that remain empty (i.e., contain no detected objects) after processing.",
|
|
167
|
+
"@initial_elements_keep_empty<use>": "Set to 1 to retain empty initial elements in the output. Set to 0 (default) to automatically remove any initial elements that failed to capture a valid object, ensuring the final structure contains only populated elements.",
|
|
168
|
+
"initial_elements_keep_empty": 0,
|
|
169
|
+
"@initial_element_expansion<desc>": "Bounding box expansion (in points) applied to an initial element when searching for its children. Used when element flag contains no_expand.",
|
|
170
|
+
"@initial_element_expansion<use>": "Controls how far around an initial element the engine reaches to capture child elements. Increase if children slightly overflow the initial bbox (e.g. tables, multi-line headers, clipped PDFs). Decrease if it pulls in content from adjacent columns or blocks.",
|
|
171
|
+
"initial_element_expansion": 2,
|
|
172
|
+
"@initial_element_overlap<desc>": "Minimum required overlap ratio (0–1) between a candidate bbox and the initial element bbox to accept the candidate as a child. Used when the initial element uses **no expand** (fixed bounds). If set to 0, `initial_element_expansion` is used instead (inflate parent bbox, then require full containment).",
|
|
173
|
+
"@initial_element_overlap<use>": "On **no expand** flag: 0 → strict inside expanded box (tune `initial_element_expansion`). \u003e0 → ratio of intersection area to **candidate** area must exceed this; lower ≈ more tolerant, higher ≈ stricter against neighbors.",
|
|
174
|
+
"initial_element_overlap": 0,
|
|
175
|
+
"@initial_element_expand_overlap<desc>": "Minimum required overlap ratio (0–1): intersection area divided by the **candidate** bbox area must exceed this value. Used when the initial element **may expand** (flag doesn't contain no_expand).",
|
|
176
|
+
"@initial_element_expand_overlap<use>": "Controls child inclusion using overlap ratio. Higher value → stricter (prevents leakage to neighbors). Lower value → more tolerant (captures partially overlapping elements but may include nearby content).",
|
|
177
|
+
"initial_element_expand_overlap": 0.5,
|
|
178
|
+
"@annot_char_overlap<desc>": "Minimum fraction (0–1) of a character’s area that must lie under an annotation to associate that annotation with the character during tagging.",
|
|
179
|
+
"@annot_char_overlap<use>": "Raise to link only clearly covered glyphs; lower if small highlights or stamps should attach to nearby text.",
|
|
180
|
+
"annot_char_overlap": 0.05,
|
|
181
|
+
"@isolated_text_ratio<desc>": "Upper bound on text bbox width as a multiple of typical font size when classifying a region as “isolated” for path splitting and graphic guards.",
|
|
182
|
+
"@isolated_text_ratio<use>": "Lower to treat fewer wide text blocks as isolated; raise when path splitting should ignore wide body columns.",
|
|
183
|
+
"isolated_text_ratio": 10,
|
|
184
|
+
"@isolated_text<desc>": "Upper bound in points on text bbox width for the same isolated-region tests when the ratio-based path is not used.",
|
|
185
|
+
"@isolated_text<use>": "Absolute cap companion to isolated_text_ratio; tune when font-size scaling is unreliable.",
|
|
186
|
+
"isolated_text": 80,
|
|
187
|
+
"@isolated_element_ratio<desc>": "Upper bound on element bbox width and height as a multiple of font size when deciding if a graphic element is isolated.",
|
|
188
|
+
"@isolated_element_ratio<use>": "Lower to flag fewer large graphics as isolated; raise when small floating icons should score as islands.",
|
|
189
|
+
"isolated_element_ratio": 6,
|
|
190
|
+
"@element_isolated_w1<desc>": "Weight of paragraph-like layout evidence when scoring whether a floating element is isolated.",
|
|
191
|
+
"@element_isolated_w1<use>": "Raise when paragraph context should strongly veto isolation; set to 0 to ignore that term.",
|
|
192
|
+
"element_isolated_w1": 1,
|
|
193
|
+
"@element_isolated_w2<desc>": "Weight of horizontal span / width cues in the same isolated-element score.",
|
|
194
|
+
"@element_isolated_w2<use>": "Raise when narrow horizontal floats should read as isolated; lower when width alone is misleading.",
|
|
195
|
+
"element_isolated_w2": 1,
|
|
196
|
+
"@element_isolated_caption<desc>": "If 1, elements that already carry a detected caption (table, figure, chart, note) are not treated as isolated layout islands.",
|
|
197
|
+
"@element_isolated_caption<use>": "Keep enabled so captioned figures stay grouped; disable only if false captions block isolation heuristics.",
|
|
198
|
+
"element_isolated_caption": 1,
|
|
199
|
+
"@element_isolated_width_min<desc>": "Minimal value of bbox width for isolated element. If zero, element_isolated_width_min_ratio is used. Size in points.",
|
|
200
|
+
"@element_isolated_width_min<use>": "Fixed minimum width in points when ratio is disabled; use for uniform point-size documents.",
|
|
201
|
+
"element_isolated_width_min": 0,
|
|
202
|
+
"@element_isolated_width_min_ratio<desc>": "Minimal value of bbox width for isolated element multiply with average page font size.",
|
|
203
|
+
"@element_isolated_width_min_ratio<use>": "Raise to drop very narrow noise; lower so slim markers still participate in isolation scoring.",
|
|
204
|
+
"element_isolated_width_min_ratio": 4,
|
|
205
|
+
"@element_isolated_width_max<desc>": "Maximal value of bbox width for isolated element. If zero, element_isolated_width_max_ratio is used. Size in points.",
|
|
206
|
+
"@element_isolated_width_max<use>": "Fixed maximum width cap when ratio is disabled.",
|
|
207
|
+
"element_isolated_width_max": 0,
|
|
208
|
+
"@element_isolated_width_max_ratio<desc>": "Maximal value of bbox width for isolated element multiply with average page font size.",
|
|
209
|
+
"@element_isolated_width_max_ratio<use>": "Lower to stop full-width panels from being treated as isolated floats; raise for wide sidebars.",
|
|
210
|
+
"element_isolated_width_max_ratio": 10,
|
|
211
|
+
"@element_isolated_similarity<desc>": "Isolation score threshold [0,1]: in preflight, text is only collected as a column candidate when its isolation score is below this value; debug builds also use it to highlight highly isolated elements.",
|
|
212
|
+
"@element_isolated_similarity<use>": "Increase if body text is skipped for column preflight; decrease if narrow columns are ignored because they look too isolated.",
|
|
213
|
+
"element_isolated_similarity": 0.7,
|
|
214
|
+
"@element_line_w1<desc>": "Weight for stroke-thickness similarity when classifying paths as lines; if 0, width-based line detection is disabled.",
|
|
215
|
+
"@element_line_w1<use>": "Set to 0 to turn off line detection from width; otherwise tune together with element_line_width_max / _ratio.",
|
|
216
|
+
"element_line_w1": 1,
|
|
217
|
+
"@element_line_width_max<desc>": "Maximal value of line width. If zero, element_line_width_max_ratio is used. Size in points.",
|
|
218
|
+
"@element_line_width_max<use>": "Absolute stroke cap; use when page font scaling should not widen the line threshold.",
|
|
219
|
+
"element_line_width_max": 8,
|
|
220
|
+
"@element_line_width_max_ratio<desc>": "When element_line_width_max is 0, maximum stroke thickness is element_line_width_max_ratio times the page’s reference font size (the same scale used when comparing strokes for line detection).",
|
|
221
|
+
"@element_line_width_max_ratio<use>": "Raise when the page’s reference font size is small but thick rules should still score as lines; lower to reject fat strokes as lines.",
|
|
222
|
+
"element_line_width_max_ratio": 1,
|
|
223
|
+
"@element_line_similarity<desc>": "Line-likeness score threshold [0,1] (width vs. thickness style metric). Used when classifying thin rectangles and images as stroke-like lines, and when rejecting a table candidate as corrupted if the region looks too much like a single rule instead of a grid.",
|
|
224
|
+
"@element_line_similarity<use>": "Higher values require a stronger line-like score before the condition triggers. Lower slightly if valid tables are rejected as corrupted or thin rules are missed; raise if decorative strokes are kept as tables or overly large fills are classified as lines.",
|
|
225
|
+
"element_line_similarity": 0.6,
|
|
226
|
+
"@element_alignment_ratio<desc>": "Blend between baseline-based and bbox-based alignment checks. Bbox alignment is weighted by element_alignment_ratio relative to baseline alignment.",
|
|
227
|
+
"@element_alignment_ratio<use>": "Toward 1 favors box edges; toward 0 favors typographic baseline alignment for line grouping.",
|
|
228
|
+
"element_alignment_ratio": 0.5,
|
|
229
|
+
"@rect_image_similarity<desc>": "Minimum image similarity [0,1] for converting a plain rectangle into an image element (when line classification does not win).",
|
|
230
|
+
"@rect_image_similarity<use>": "Increase to reduce false image rects; decrease when filled boxes should become images more often.",
|
|
231
|
+
"rect_image_similarity": 0.7,
|
|
232
|
+
"@rect_line_similarity<desc>": "Minimum line similarity [0,1] for converting a rectangle into a line element (evaluated before the image branch).",
|
|
233
|
+
"@rect_line_similarity<use>": "Increase to require a higher line similarity before a rect converts to pde_line (stricter); decrease when thin rule rects are missed.",
|
|
234
|
+
"rect_line_similarity": 0.5,
|
|
235
|
+
"@image_background_text<desc>": "Expands text bounding boxes when testing whether a graphic is a text background highlight.",
|
|
236
|
+
"@image_background_text<use>": "When enabled, image regions that closely match expanded text boxes (typical highlights or shading) are treated as background and moved to artifacts instead of standalone figures.",
|
|
237
|
+
"image_background_text": 1,
|
|
238
|
+
"@image_overlap_distance<desc>": "Maximum distance value when graphic page objects can be joined. Distance in points.",
|
|
239
|
+
"@image_overlap_distance<use>": "Graphic merging: lower values keep nearby images and icons separate (helps preserve small labels); higher values join moderately separated fragments into one image object.",
|
|
240
|
+
"image_overlap_distance": 1,
|
|
241
|
+
"@image_join_distance<desc>": "Defines the maximum allowed distance (in points) between small images for them to be considered joinable. These parameters help fine-tune the grouping of small image elements into a cohesive larger visual block based on their spatial proximity.",
|
|
242
|
+
"@image_join_distance<use>": "High value: Allows merging of images that are spaced apart moderately. Low value: Requires images to be nearly adjacent to be joined.",
|
|
243
|
+
"image_join_distance": 8,
|
|
244
|
+
"@char_clip_ratio<desc>": "Minimum ratio of visible (unclipped) character area to the character’s full glyph box.",
|
|
245
|
+
"@char_clip_ratio<use>": "Lower to keep partially clipped glyphs in words; raise to drop characters mostly outside the clip path.",
|
|
246
|
+
"char_clip_ratio": 0.5,
|
|
247
|
+
"@word_space_width_ratio<desc>": "The word_space_width_ratio is a multiplier that determines the threshold for identifying inter-word spaces by comparing the gap between characters to the typical width of a space character. It scales the space width so that small variations in spacing can be interpreted as either a valid word separator or a mere character gap.",
|
|
248
|
+
"@word_space_width_ratio<use>": "For example, if the space character in a font is 5 units wide and word_space_width_ratio is set to 1.5, the allowed gap between words would be 7.5 units. If the actual gap is less than this threshold, the algorithm may treat the characters as part of the same word, which is especially useful for fine-tuning text extraction from PDFs with variable spacing.",
|
|
249
|
+
"word_space_width_ratio": 0.6,
|
|
250
|
+
"@word_space_width_min_ratio<desc>": "The word_space_width_min_ratio is an additional multiplier that sets a minimum threshold for the allowed space between words. It ensures that, even when minimal character spacing is detected, the computed gap used to determine word boundaries does not fall below a baseline value relative to the font size.",
|
|
251
|
+
"@word_space_width_min_ratio<use>": "For instance, if the minimal character spacing in a text run is 2 units and the font size is 10, setting word_space_width_min_ratio to 0.5 would enforce a minimum allowed space of 2+(10x0.5) = 7 units. This helps prevent the algorithm from erroneously merging words when the spacing is very tight.",
|
|
252
|
+
"word_space_width_min_ratio": 0.1,
|
|
253
|
+
"@word_space_distance_max<desc>": "The maximum fixed distance (in points) allowed between two words to consider them part of the same text line.",
|
|
254
|
+
"@word_space_distance_max<use>": "Use this to set a hard limit. If the gap between the current line end and the next word is larger than this value, the word is rejected (starts a new line or chunk). If set to 0, the ratio-based calculation is used instead.",
|
|
255
|
+
"word_space_distance_max": 0,
|
|
256
|
+
"@word_space_distance_max_ratio<desc>": "The maximum distance allowed between words to join them into a line, calculated as a multiple of the font size.",
|
|
257
|
+
"@word_space_distance_max_ratio<use>": "Used only if 'word_space_distance_max' is 0. The engine calculates 'Threshold = MaxFontSize * Ratio'. If the gap is smaller than this threshold, the word joins the line. Increase this value if words in the same line are being split (e.g., justified text with large gaps). Decrease it if separate columns are being merged.",
|
|
258
|
+
"word_space_distance_max_ratio": 0,
|
|
259
|
+
"@word_space_ratio<desc>": "Scale factor applied when estimating default word spacing from the font’s space character (simple vs justified-like lines).",
|
|
260
|
+
"@word_space_ratio<use>": "Tune if words merge or split incorrectly: increase slightly when tight fonts leave gaps too small to split words; decrease when extra spaces are mistaken for word boundaries.",
|
|
261
|
+
"word_space_ratio": 1,
|
|
262
|
+
"@word_space_update_min<desc>": "Minimum ratio of detected word spacing.",
|
|
263
|
+
"@word_space_update_min<use>": "Floor for learned space width after line clustering; raise if spacing collapses too small.",
|
|
264
|
+
"word_space_update_min": 0.2,
|
|
265
|
+
"@word_space_update_max<desc>": "Maximum ratio of detected word spacing. If set to 0, automatic word-spacing updates from lines are not applied.",
|
|
266
|
+
"@word_space_update_max<use>": "Ceiling for learned spacing; set 0 to freeze defaults from the font only.",
|
|
267
|
+
"word_space_update_max": 4,
|
|
268
|
+
"@word_space_update_distance<desc>": "DBSCAN epsilon scale for clustering measured word gaps: effective eps = font size × this factor.",
|
|
269
|
+
"@word_space_update_distance<use>": "Increase slightly (e.g. 0.05–0.08) if spacing clusters fragment; decrease to separate distinct spacing modes more strictly.",
|
|
270
|
+
"word_space_update_distance": 0.04,
|
|
271
|
+
"@word_overlap<desc>": "Minimum overlap percentage (0-1) required between bounding boxes to consider words as duplicates. A word must cover at least this percentage of another word's area to be considered overlapping.",
|
|
272
|
+
"@word_overlap<use>": "Lower to deduplicate near-touching OCR layers; raise to keep legitimately stacked text separate.",
|
|
273
|
+
"word_overlap": 0.9,
|
|
274
|
+
"@text_line_baseline_ratio<desc>": "Maximum baseline shift. Value multiplies minimal font. Baseline shift moves individual characters up or down in relation to other text on the same line.",
|
|
275
|
+
"@text_line_baseline_ratio<use>": "Baseline detection. The baseline is the line upon which most letters ”sit” and below which descenders extend. Use higher value to smooth baseline shift.",
|
|
276
|
+
"text_line_baseline_ratio": 0.1,
|
|
277
|
+
"@text_line_insert_elem_ratio<desc>": "Maximum vertical shift/overlap ratio allowed when inserting an element into a text line. The value is multiplied by the minimum font size.",
|
|
278
|
+
"@text_line_insert_elem_ratio<use>": "Use this to control how strictly inline elements (like images, icons, or formulas) must align with the text baseline. Lower values enforce strict alignment (preventing merging of nearby lines), while higher values allow 'floating' elements to be accepted into the line.",
|
|
279
|
+
"text_line_insert_elem_ratio": 0.5,
|
|
280
|
+
"@text_line_underline_distance<desc>": "Distance of the underline line and text baseline. Size in points.",
|
|
281
|
+
"@text_line_underline_distance<use>": "Vertical offset between detected underline geometry and text baseline (points); adjust if underlines are missed or false positives appear.",
|
|
282
|
+
"text_line_underline_distance": 2.6,
|
|
283
|
+
"@text_line_underline_char_distance_ratio<desc>": "Distance of the underline line start/end point and character bounding box. The value is multiplied by line font size. Size in points.",
|
|
284
|
+
"@text_line_underline_char_distance_ratio<use>": "Horizontal tolerance for aligning underline endpoints to character edges, as a ratio of line font size; tighten if underline segments attach to wrong words.",
|
|
285
|
+
"text_line_underline_char_distance_ratio": 0.1,
|
|
286
|
+
"@text_line_superscript_detect<desc>": "If 1, the engine attempts to detect and merge superscript words into the same line when they are close enough to the baseline and horizontally near their main word. If 0, no special handling for superscripts is applied; all words are treated as part of the same line based on regular spacing rules.",
|
|
287
|
+
"@text_line_superscript_detect<use>": "Enable to capture superscripts (e.g. footnote markers, mathematical exponents) as part of the main text line when they are positioned appropriately. Keep disabled if superscripts should be treated as separate lines or if the document has many false positives in that position.",
|
|
288
|
+
"text_line_superscript_detect": 1,
|
|
289
|
+
"@text_line_superscript_shift_ratio<desc>": "Maximum vertical baseline shift allowed for a word to be considered a superscript, defined as a ratio of the maximum font size.",
|
|
290
|
+
"@text_line_superscript_shift_ratio<use>": "Controls vertical tolerance. The algorithm calculates 'Threshold = MaxFontSize * Ratio'. If the vertical distance between baselines is LESS than this threshold, the word is accepted as a superscript (part of the line). If greater, it is treated as a separate line.",
|
|
291
|
+
"text_line_superscript_shift_ratio": 0.5,
|
|
292
|
+
"@text_line_subscript_shift_ratio<desc>": "Maximum vertical baseline shift allowed for a word to be considered a subscript, defined as a ratio of the maximum font size.",
|
|
293
|
+
"@text_line_subscript_shift_ratio<use>": "Similar to superscript logic but for subscripts (words shifted downwards). If the baseline difference is within this ratio, it is merged into the text line; otherwise, it may form a new line.",
|
|
294
|
+
"text_line_subscript_shift_ratio": 0.4,
|
|
295
|
+
"@text_line_superscript_dist_ratio<desc>": "Maximum horizontal distance (gap or overlap) allowed between a word and its superscript, defined as a ratio of the maximum font size.",
|
|
296
|
+
"@text_line_superscript_dist_ratio<use>": "Controls horizontal proximity. The algorithm calculates 'Threshold = MaxFontSize * Ratio'. The superscript must be horizontally closer to the main word than this value to be joined to the line.",
|
|
297
|
+
"text_line_superscript_dist_ratio": 0.3,
|
|
298
|
+
"@text_line_subscript_dist_ratio<desc>": "Maximum horizontal distance (gap or overlap) allowed between a word and its subscript, defined as a ratio of the maximum font size.",
|
|
299
|
+
"@text_line_subscript_dist_ratio<use>": "Controls horizontal proximity for subscripts. Ensures that a subscript is visually attached to the preceding word.",
|
|
300
|
+
"text_line_subscript_dist_ratio": 0.3,
|
|
301
|
+
"@text_line_join_font_size_distance<desc>": "Maximum font-size difference (points) allowed when merging two lines that use different fonts into one paragraph line.",
|
|
302
|
+
"@text_line_join_font_size_distance<use>": "Raise to glue headings and body when sizes differ slightly; lower to split mixed-font blocks.",
|
|
303
|
+
"text_line_join_font_size_distance": 0.1,
|
|
304
|
+
"@text_line_distance_max<desc>": "Maximum distance between lines. If zero, text_line_distance_max_ratio is used. Size in points.",
|
|
305
|
+
"@text_line_distance_max<use>": "Vertical gap between lines: any spacing wider than this threshold (points) starts a new paragraph block. If zero, the ratio-based limit is used instead.",
|
|
306
|
+
"text_line_distance_max": 0,
|
|
307
|
+
"@text_line_distance_max_ratio<desc>": "Maximum distance between lines. The value is multiplied by line font size.",
|
|
308
|
+
"@text_line_distance_max_ratio<use>": "Used when text_line_distance_max is zero: threshold is ratio × line font size. Increase if one paragraph is split too often; decrease if distinct paragraphs merge.",
|
|
309
|
+
"text_line_distance_max_ratio": 2,
|
|
310
|
+
"@text_line_join_distance<desc>": "Maximum threshold value in line spacing detection for specific font size. The higher value allows creating paragraph with variable line spacings. The value is multiplied by font size.",
|
|
311
|
+
"@text_line_join_distance<use>": "Used in paragraph detection, this parameter sets the acceptable vertical gap between lines. A higher value permits more variability in line spacing; if set to zero, the algorithm defaults to using the minimum font size.",
|
|
312
|
+
"text_line_join_distance": 2,
|
|
313
|
+
"@text_line_chunk_distance_max<desc>": "Maximum word-to-word gap inside a line chunk before forcing a split. If zero, text_line_chunk_distance_max_ratio scales simple word spacing instead. Size in points.",
|
|
314
|
+
"@text_line_chunk_distance_max<use>": "Within a line, gaps between words wider than this limit (points) start a new chunk. Zero defers to text_line_chunk_distance_max_ratio × simple word spacing.",
|
|
315
|
+
"text_line_chunk_distance_max": 0,
|
|
316
|
+
"@text_line_chunk_distance_max_ratio<desc>": "Maximum distance between chunks. The value is multiplied by simple word spacing between words.",
|
|
317
|
+
"@text_line_chunk_distance_max_ratio<use>": "Used when text_line_chunk_distance_max is zero: scales the chunking threshold from word spacing. Raise to tolerate uneven justified lines; lower to separate tight columns.",
|
|
318
|
+
"text_line_chunk_distance_max_ratio": 6,
|
|
319
|
+
"@text_line_chunk_distance<desc>": "DBSCAN epsilon (points) for clustering measured word gaps within a chunk. If nonzero, it is used directly; if zero, text_line_chunk_distance_ratio times line font size is used instead.",
|
|
320
|
+
"@text_line_chunk_distance<use>": "Raise for more tolerant single-cluster spacing (fewer chunk splits); lower to separate spacing modes more aggressively. Ignored when zero in favor of the ratio-based epsilon.",
|
|
321
|
+
"text_line_chunk_distance": 0,
|
|
322
|
+
"@text_line_chunk_distance_ratio<desc>": "When text_line_chunk_distance is zero, DBSCAN epsilon is line font size × this ratio.",
|
|
323
|
+
"@text_line_chunk_distance_ratio<use>": "Raise to merge more spacing variation into one cluster; lower to split mixed or justified spacing into separate chunks.",
|
|
324
|
+
"text_line_chunk_distance_ratio": 0.4,
|
|
325
|
+
"@text_chunk_distance<desc>": "Maximum distance value when text chunks are vertically aligned. If zero, text_chunk_distance_ratio is used. Size in points.",
|
|
326
|
+
"@text_chunk_distance<use>": "Vertical alignment of text chunks across the page: higher values tolerate looser column grids so borderless tables are easier to find; lower values favor simple paragraphs.",
|
|
327
|
+
"text_chunk_distance": 0,
|
|
328
|
+
"@text_chunk_distance_ratio<desc>": "Maximum distance value when text chunks are vertically aligned. The value is multiplied by page font width.",
|
|
329
|
+
"@text_chunk_distance_ratio<use>": "Same as text_chunk_distance when the fixed distance is zero, scaled by typical font width. Increase for ragged columns; decrease if unrelated blocks align by chance.",
|
|
330
|
+
"text_chunk_distance_ratio": 0.42,
|
|
331
|
+
"@text_chunk_simple_distance<desc>": "Maximum distance value when text chunks create simple line. Normalized to interval [0,1].",
|
|
332
|
+
"@text_chunk_simple_distance<use>": "Similarity threshold [0,1] for treating chunk columns as a simple text line: lower values demand straighter alignment; higher values allow slightly wavy or noisy lines.",
|
|
333
|
+
"text_chunk_simple_distance": 0.4,
|
|
334
|
+
"@text_chunk_word_distance<desc>": "Similarity threshold [0,1] for splitting a single-line text chunk into words when internal spacing is ambiguous.",
|
|
335
|
+
"@text_chunk_word_distance<use>": "Lower values split glued tokens more aggressively; raise the value when legitimate single words are being chopped apart.",
|
|
336
|
+
"text_chunk_word_distance": 0.1,
|
|
337
|
+
"@text_height<desc>": "Typical text height in points used as a scale reference: minimum table cell sizing, graphic-table span caps (with table span ratios), line spacing, and other distance heuristics.",
|
|
338
|
+
"@text_height<use>": "Raise for large-type PDFs so spacing and table minima scale up; lower slightly if graphic-table rowspan/colspan merges are too aggressive on fine print.",
|
|
339
|
+
"text_height": 8,
|
|
340
|
+
"@text_simple_similarity<desc>": "Minimum similarity value when text lines create a simple paragraph normalized to interval [0,1].",
|
|
341
|
+
"@text_simple_similarity<use>": "Left-aligned paragraph glue: higher values require stronger similarity between adjacent lines before they merge; lower values form longer paragraphs more aggressively.",
|
|
342
|
+
"text_simple_similarity": 0.96,
|
|
343
|
+
"@text_justify_similarity<desc>": "Minimum similarity value when text lines create a justify paragraph normalized to interval [0,1].",
|
|
344
|
+
"@text_justify_similarity<use>": "Justified paragraph glue: higher values keep only tightly matching line spacing patterns; lower values join more lines into one justified block.",
|
|
345
|
+
"text_justify_similarity": 0.96,
|
|
346
|
+
"@text_table_similarity<desc>": "Minimum similarity value when text lines create a table normalized to interval [0,1].",
|
|
347
|
+
"@text_table_similarity<use>": "Text-table hypothesis: lower values accept looser column alignment as a table; higher values demand tighter grid-like similarity before a text table is formed.",
|
|
348
|
+
"text_table_similarity": 0.65,
|
|
349
|
+
"@text_split_distance<desc>": "Dissimilarity cutoff when adjacent text lines are evaluated for splitting one paragraph into two.",
|
|
350
|
+
"@text_split_distance<use>": "Increase when unrelated lines are glued into one paragraph; decrease when one logical paragraph is being broken across lines.",
|
|
351
|
+
"text_split_distance": 0.2,
|
|
352
|
+
"@label_image_detect<desc>": "If 1, list/label markers drawn as vector paths can be detected; if 0, path-based label candidates are skipped.",
|
|
353
|
+
"@label_image_detect<use>": "Disable when vector bullets are noise; keep enabled for Wingdings/Symbol-style markers.",
|
|
354
|
+
"label_image_detect": 1,
|
|
355
|
+
"@label_word_detect<desc>": "If 1, list/label markers from extracted words are detected; if 0, word-based label detection is disabled.",
|
|
356
|
+
"@label_word_detect<use>": "Disable for pure vector lists; keep enabled for typed “1.” or “a)” prefixes.",
|
|
357
|
+
"label_word_detect": 1,
|
|
358
|
+
"@label_distance_ratio<desc>": "Maximum label-to-text gap as a multiple of the dominant page font size for pairing markers with body text.",
|
|
359
|
+
"@label_distance_ratio<use>": "Raise for wide indents or scanned skew; lower when distant numbers are wrongly attached to lines.",
|
|
360
|
+
"label_distance_ratio": 10,
|
|
361
|
+
"@label_image_w1<desc>": "Controls how much vertical alignment matters when clustering labels. A higher value enforces stricter alignment, while a lower value allows more variation.",
|
|
362
|
+
"@label_image_w1<use>": "High value: For well-aligned labels in structured documents like tables and forms. Low or zero: For misaligned labels, scanned text, or inconsistent formatting.",
|
|
363
|
+
"label_image_w1": "1",
|
|
364
|
+
"@label_image_w2<desc>": "Weight for how closely two vector markers agree on horizontal offset from their paired sibling words (start, end, and center gaps to the sibling edge). This compares candidate A vs B, not a single absolute gap.",
|
|
365
|
+
"@label_image_w2<use>": "Raise when list markers must share the same indent relative to body text before they cluster; lower when similar bullets sit at slightly different offsets and should still be one list style.",
|
|
366
|
+
"label_image_w2": "1",
|
|
367
|
+
"@label_image_w3<desc>": "This weight controls how much the label's width consistency matters in clustering. A higher value enforces that labels should have the same width, while a lower value allows more variation in width between labels.",
|
|
368
|
+
"@label_image_w3<use>": "Higher value: When it's important for labels to have similar widths, such as in structured documents like forms. Lower value: When label width can vary, such as in less structured or more creative layouts.",
|
|
369
|
+
"label_image_w3": "1",
|
|
370
|
+
"@label_image_w4<desc>": "This weight determines how important the height consistency of labels is when clustering. A higher value enforces that labels should have the same height, while a lower value allows more flexibility in height differences.",
|
|
371
|
+
"@label_image_w4<use>": "Higher value: When height uniformity is important for the document structure, like in neatly aligned forms. Lower value: When height variations between labels are acceptable.",
|
|
372
|
+
"label_image_w4": "1",
|
|
373
|
+
"@label_image_w5<desc>": "This weight adjusts how important the height relationship is between the image label and its associated text. A higher value means the height alignment between the label and the text is more significant in clustering decisions.",
|
|
374
|
+
"@label_image_w5<use>": "Higher value: When it's crucial that the label and associated text are aligned in height, such as in forms where labels and content need to be closely related. Lower value: When label and text height alignment is less critical for grouping.",
|
|
375
|
+
"label_image_w5": "0.5",
|
|
376
|
+
"@label_image_width_min<desc>": "Specifies a fixed minimum width in points. If set to zero, label_image_width_min_ratio is used instead.",
|
|
377
|
+
"@label_image_width_min<use>": "Fixed Minimum Width (label_image_width_min > 0): Use this setting when you expect label images to have a consistent size across the document. For example, if you set label_image_width_min to 10, any image with a width of less than 10 points will not be processed as a label.",
|
|
378
|
+
"label_image_width_min": 0,
|
|
379
|
+
"@label_image_width_min_ratio<desc>": "When label_image_width_min is 0: minimum marker width as a multiple of the paired sibling word’s font size (the body word next to the marker).",
|
|
380
|
+
"@label_image_width_min_ratio<use>": "Example: ratio 1 and sibling text at 12 pt rejects markers narrower than 12 pt. Raise to drop very thin noise; lower so small bullets still pair when a sibling exists.",
|
|
381
|
+
"label_image_width_min_ratio": 0,
|
|
382
|
+
"@label_image_width_max<desc>": "Specifies a fixed maximum width in points. If set to zero, the label_image_width_max_ratio is used instead.",
|
|
383
|
+
"@label_image_width_max<use>": "Fixed Maximum Width (label_image_width_max > 0): Use this setting when you expect label images to have a consistent maximum size across the document. For example, if you set label_image_width_max to 100, any image with a width greater than 100 points will not be processed as a label.",
|
|
384
|
+
"label_image_width_max": 0,
|
|
385
|
+
"@label_image_width_max_ratio<desc>": "When label_image_width_max is 0: maximum marker width as a multiple of the paired sibling word’s font size.",
|
|
386
|
+
"@label_image_width_max_ratio<use>": "Example: ratio 6 and sibling text at 12 pt rejects markers wider than 72 pt. Lower to exclude wide icons; raise for oversized custom bullets.",
|
|
387
|
+
"label_image_width_max_ratio": 6,
|
|
388
|
+
"@label_image_distance<desc>": "Clustering threshold in points that decides when labels should be grouped together. A higher value makes clustering more flexible, allowing distant labels to merge, while a lower value keeps clusters tight and separate.",
|
|
389
|
+
"@label_image_distance<use>": "Higher values merge image-label clusters across larger gaps (scanned or uneven layouts); lower values keep clusters tight and reduce false merges.",
|
|
390
|
+
"label_image_distance": "4",
|
|
391
|
+
"@label_word_w1<desc>": "Controls how much vertical alignment matters when clustering labels. A higher value enforces stricter alignment, while a lower value allows more variation.",
|
|
392
|
+
"@label_word_w1<use>": "High value: For well-aligned labels in structured documents like tables and forms. Low or zero: For misaligned labels, scanned text, or inconsistent formatting.",
|
|
393
|
+
"label_word_w1": "1",
|
|
394
|
+
"@label_word_w2<desc>": "Weight for agreement between two word-based markers on horizontal offset from their sibling text (same start/end/center gaps as for image labels; LTR adds an extra horizontal position consistency term between the two candidates).",
|
|
395
|
+
"@label_word_w2<use>": "Raise so only markers with matching indent-to-text geometry cluster; lower to allow the same list style when offsets differ slightly.",
|
|
396
|
+
"label_word_w2": "1",
|
|
397
|
+
"@label_word_dist_sibling_ratio<desc>": "This threshold, defined as a ratio multiplied by a siblings font size, sets the maximum gap allowed between a label and its sibling element to be joined together. If the distance exceeds this value, the label and its sibling remain separate.",
|
|
398
|
+
"@label_word_dist_sibling_ratio<use>": "For example, with a ratio of 2.0 and a sibling font size of 12, a gap greater than 24 points prevents the label from being joined with its sibling.",
|
|
399
|
+
"label_word_dist_sibling_ratio": "4",
|
|
400
|
+
"@label_sibling_distance_ratio<desc>": "Multiplier applied to the label–true-sibling gap when a competing “false” sibling word exists: if the false sibling sits within end_dist × this ratio of the label, the singleton geometry is rejected (label_singleton_geometry_rejected in label and image pairing).",
|
|
401
|
+
"@label_sibling_distance_ratio<use>": "Raise to reject more label candidates when another word sits moderately past the true sibling; lower if valid markers are dropped when a nearby word looks like a false sibling.",
|
|
402
|
+
"label_sibling_distance_ratio": 1.2,
|
|
403
|
+
"@label_word_distance<desc>": "Clustering threshold in points that decides when labels should be grouped together. A higher value makes clustering more flexible, allowing distant labels to merge, while a lower value keeps clusters tight and separate.",
|
|
404
|
+
"@label_word_distance<use>": "Higher values allow word-based list markers to cluster across larger horizontal gaps; lower values require markers to sit closer to their body text.",
|
|
405
|
+
"label_word_distance": "0",
|
|
406
|
+
"@label_word_distance_ratio<desc>": "Used only when label_word_distance is 0: hierarchical clustering cutoff equals this ratio times the page font width.",
|
|
407
|
+
"@label_word_distance_ratio<use>": "Increase when markers are far from text in body-relative units; decrease when unrelated markers are incorrectly grouped.",
|
|
408
|
+
"label_word_distance_ratio": "1",
|
|
409
|
+
"@label_order_detect<desc>": "If 1, the engine ignores list that do not have a consistent label order (e.g. 1., 2., 3. or a), b), c)); if 0, no label-order checks are applied and any cluster of markers can form a list style.",
|
|
410
|
+
"@label_order_detect<use>": "Enable to enforce consistent numbering or lettering in lists, which can help filter out noise; disable if the document has irregular list markers or if false positives are being rejected due to strict ordering.",
|
|
411
|
+
"label_order_detect": "1",
|
|
412
|
+
"@toc_detect<desc>": "TOC detection. Possible values: 0 | 1. If 0 is set, it prevents generating TOC from words.",
|
|
413
|
+
"@toc_detect<use>": "Disable on pages that look like TOCs but are body content; enable for outline-style front matter.",
|
|
414
|
+
"toc_detect": 1,
|
|
415
|
+
"@toc_word_distance<desc>": "Clustering cutoff in points when splitting the TOC candidate tree into clusters (epsilon passed to the hierarchical cut). If set to 0, the cutoff is toc_word_distance_ratio × page font width instead.",
|
|
416
|
+
"@toc_word_distance<use>": "If the same TOC is broken into too many clusters, increase this value (or the ratio when distance is 0). If unrelated lines or numbers merge into one TOC, decrease it. Use 0 with a tuned ratio when you prefer body-relative spacing.",
|
|
417
|
+
"toc_word_distance": 0,
|
|
418
|
+
"@toc_word_distance_ratio<desc>": "Used only when toc_word_distance is 0: multiplied by page font width to obtain the same tree-cut epsilon as a non-zero toc_word_distance.",
|
|
419
|
+
"@toc_word_distance_ratio<use>": "Tune together with toc_word_distance: higher values merge more aggressively (larger effective cutoff in font-width units); lower values keep clusters tighter.",
|
|
420
|
+
"toc_word_distance_ratio": "1",
|
|
421
|
+
"@graphic_table_detect<desc>": "Master switch for path-based (graphic) table detection. 0 disables building tables from vector paths; 1 enables the graphic-table pipeline.",
|
|
422
|
+
"@graphic_table_detect<use>": "Disable for illustration-heavy pages where rules are decorative; keep on for form-like vector grids.",
|
|
423
|
+
"graphic_table_detect": 1,
|
|
424
|
+
"@graphic_table_detect_row<desc>": "If 0, y-coordinates from path lines and rectangle edges are not used to build row bands for path-based (graphic) tables; if 1, horizontal rules contribute row structure.",
|
|
425
|
+
"@graphic_table_detect_row<use>": "Turn off when only vertical guides define the grid (ledger lines without row strokes) to avoid phantom rows; keep on when horizontal rules define row boundaries.",
|
|
426
|
+
"graphic_table_detect_row": 1,
|
|
427
|
+
"@graphic_table_detect_col<desc>": "If 0, x-coordinates from path lines and rectangle edges are not used to build column guides for path-based (graphic) tables; if 1, vertical rules contribute column structure.",
|
|
428
|
+
"@graphic_table_detect_col<use>": "Turn off for row-only layouts or horizontal rule stacks; keep on when vertical strokes define columns.",
|
|
429
|
+
"graphic_table_detect_col": 1,
|
|
430
|
+
"@graphic_table_alignment_distance<desc>": "Maximum alignment slack in points when path-derived table elements are scored as belonging to the same row or column grid.",
|
|
431
|
+
"@graphic_table_alignment_distance<use>": "Raise for hand-drawn or slightly misaligned rules; lower for CAD-tight grids.",
|
|
432
|
+
"graphic_table_alignment_distance": 0.8,
|
|
433
|
+
"@graphic_table_split_w1<desc>": "Weight of paragraph-like text layout when deciding whether a graphic region should stay one table or be split.",
|
|
434
|
+
"@graphic_table_split_w1<use>": "Raise when prose-like flow inside a box should trigger a split.",
|
|
435
|
+
"graphic_table_split_w1": 1,
|
|
436
|
+
"@graphic_table_split_w2<desc>": "Weight of horizontal alignment consistency in the same split-or-keep score.",
|
|
437
|
+
"@graphic_table_split_w2<use>": "Raise when ragged x-alignment should favor breaking the table.",
|
|
438
|
+
"graphic_table_split_w2": 1,
|
|
439
|
+
"@graphic_table_split_w3<desc>": "Weight of column width regularity in the split-or-keep score.",
|
|
440
|
+
"@graphic_table_split_w3<use>": "Raise when uneven column widths indicate multiple unrelated blocks.",
|
|
441
|
+
"graphic_table_split_w3": 1,
|
|
442
|
+
"@graphic_table_split_w4<desc>": "Weight of the inferred column count in the split-or-keep score.",
|
|
443
|
+
"@graphic_table_split_w4<use>": "Higher weight makes column-count mismatches more influential.",
|
|
444
|
+
"graphic_table_split_w4": 0.5,
|
|
445
|
+
"@graphic_table_split_w5<desc>": "Weight of the inferred row count in the split-or-keep score.",
|
|
446
|
+
"@graphic_table_split_w5<use>": "Higher weight makes row-count mismatches more influential.",
|
|
447
|
+
"graphic_table_split_w5": 0.5,
|
|
448
|
+
"@graphic_table_split_w6<desc>": "Weight of how much page area the candidate occupies when balancing table vs. non-table layout.",
|
|
449
|
+
"@graphic_table_split_w6<use>": "Raise when full-page path art should resist being one giant table.",
|
|
450
|
+
"graphic_table_split_w6": 1,
|
|
451
|
+
"@graphic_table_split_col_max<desc>": "Upper limit on column count considered when evaluating a graphic split.",
|
|
452
|
+
"@graphic_table_split_col_max<use>": "Lower to cap cost on dense grids; raise only for wide tables you still want evaluated as one region.",
|
|
453
|
+
"graphic_table_split_col_max": 5,
|
|
454
|
+
"@graphic_table_split_row_max<desc>": "Upper limit on row count considered when evaluating a graphic split.",
|
|
455
|
+
"@graphic_table_split_row_max<use>": "Same as column cap but for row dimension.",
|
|
456
|
+
"graphic_table_split_row_max": 5,
|
|
457
|
+
"@graphic_table_split_similarity<desc>": "Blended split score [0,1] (paragraph-likeness, horizontal alignment, optional width boost via multivariate_distance). If this score is strictly greater than the threshold, the graphic table candidate is split—either into layout containers or into flat elements depending on graphic_table_split_layout_similarity.",
|
|
458
|
+
"@graphic_table_split_similarity<use>": "Raise the threshold so only stronger paragraph-like / misaligned scores trigger a split (valid tables stay tables longer). Lower it to split sooner when content looks like prose in a box.",
|
|
459
|
+
"graphic_table_split_similarity": 0.7,
|
|
460
|
+
"@graphic_table_split_layout_similarity<desc>": "When a split is triggered, split_to_containers_similarity (column/row count vs caps plus table area vs container) is compared with >= to this threshold: if true, split_table_to_containers; else split_table_to_elements. Also used when nested graphic tables are merged or split.",
|
|
461
|
+
"@graphic_table_split_layout_similarity<use>": "Raise to require a stronger layout-like score before using split_table_to_containers (more outcomes use split_table_to_elements). Lower to route borderline cases into containers more often.",
|
|
462
|
+
"graphic_table_split_layout_similarity": 0.7,
|
|
463
|
+
"@graphic_table_chart_similarity<desc>": "Similarity threshold [0,1] used when classifying a graphic region as chart-like versus tabular.",
|
|
464
|
+
"@graphic_table_chart_similarity<use>": "Lower values favor treating dense path artwork as charts or figures; higher values keep classifying similar regions as tables.",
|
|
465
|
+
"graphic_table_chart_similarity": 0.3,
|
|
466
|
+
"@graphic_table_image_w1<desc>": "Images area weight. If -1, number of images is used.",
|
|
467
|
+
"@graphic_table_image_w1<use>": "Use -1 to let image count drive the term; non-negative values weight total image area instead.",
|
|
468
|
+
"graphic_table_image_w1": -1,
|
|
469
|
+
"@graphic_table_image_w2<desc>": "Image-count weight in the chart-or-table score. If -1, raw image count is used instead of a fixed weight.",
|
|
470
|
+
"@graphic_table_image_w2<use>": "-1 substitutes count for a weighted blend; tune when many small icons skew charts.",
|
|
471
|
+
"graphic_table_image_w2": -1,
|
|
472
|
+
"@graphic_table_image_w3<desc>": "Chart similarity weight. If -1, number of paths is used.",
|
|
473
|
+
"@graphic_table_image_w3<use>": "-1 ties the term to path density; non-negative weights the explicit chart-similarity branch.",
|
|
474
|
+
"graphic_table_image_w3": -1,
|
|
475
|
+
"@graphic_table_image_w4<desc>": "Texts vertical alignment weight.",
|
|
476
|
+
"@graphic_table_image_w4<use>": "Raise when vertically stacked labels should push classification toward charts/figures.",
|
|
477
|
+
"graphic_table_image_w4": 1,
|
|
478
|
+
"@graphic_table_image_w5<desc>": "Table size weight.",
|
|
479
|
+
"@graphic_table_image_w5<use>": "Raise when large gridded regions should resist chart classification.",
|
|
480
|
+
"graphic_table_image_w5": 1,
|
|
481
|
+
"@graphic_table_image_similarity<desc>": "Minimum similarity value when graphic table has an image.",
|
|
482
|
+
"@graphic_table_image_similarity<use>": "Graphic table detection. Use lower value to recognize charts(images) instead of tables.",
|
|
483
|
+
"graphic_table_image_similarity": 0.7,
|
|
484
|
+
"@text_table_detect<desc>": "Master switch for text-based section tables (columns built from text/image chunks, not from vector grid paths).",
|
|
485
|
+
"@text_table_detect<use>": "Disable for narrative-only pages to avoid false borderless tables; keep on for financial or two-column grids.",
|
|
486
|
+
"text_table_detect": 1,
|
|
487
|
+
"@text_table_detect_row<desc>": "If 0, isolated text and images are not grouped into horizontal row bands for borderless (text-based) table hypotheses; if 1, row clustering runs.",
|
|
488
|
+
"@text_table_detect_row<use>": "Turn off on narrative pages or when only column structure matters. Also gates the default for split-by-cell unless a table_update rule overrides it.",
|
|
489
|
+
"text_table_detect_row": 1,
|
|
490
|
+
"@text_table_detect_col<desc>": "If 0, isolated elements are not clustered into vertical column bands for text-only table hypotheses; if 1, column alignment is evaluated.",
|
|
491
|
+
"@text_table_detect_col<use>": "Turn off for single-column numeric stacks that should stay paragraphs; keep on for multi-column borderless grids.",
|
|
492
|
+
"text_table_detect_col": 1,
|
|
493
|
+
"@text_table_row_alignment_type<desc>": "Table row alignment type [0 - strong, 1 - average, 2 - weak].",
|
|
494
|
+
"@text_table_row_alignment_type<use>": "Table detection. Select strong for strictly aligned rows.",
|
|
495
|
+
"text_table_row_alignment_type": 1,
|
|
496
|
+
"@text_table_col_alignment_type<desc>": "Table column alignment type [0 - strong, 1 - average, 2 - weak].",
|
|
497
|
+
"@text_table_col_alignment_type<use>": "Table detection. Select strong for strictly aligned columns.",
|
|
498
|
+
"text_table_col_alignment_type": 1,
|
|
499
|
+
"@text_table_col_similarity_type<desc>": "Table column similarity type [0 - column alignment distance, 1 - element distance, 2 - element size, 3 - max].",
|
|
500
|
+
"@text_table_col_similarity_type<use>": "Chooses how column similarity is measured (alignment distance, element distance, size, or max). Switch when columns align visually but differ in font or density.",
|
|
501
|
+
"text_table_col_similarity_type": 0,
|
|
502
|
+
"@text_table_col_distance<desc>": "Maximum deviation (points) when pairing nearest-neighbor elements into the same text-table column.",
|
|
503
|
+
"@text_table_col_distance<use>": "Column clustering for text tables: higher values allow slightly misaligned columns; lower values require tighter horizontal alignment.",
|
|
504
|
+
"text_table_col_distance": 0.8,
|
|
505
|
+
"@text_table_col_similarity<desc>": "Minimum weighted column table-likeness score [0,1]; below this the candidate region is rejected as not a text table.",
|
|
506
|
+
"@text_table_col_similarity<use>": "Lower values suit tables with wrapped paragraphs or uneven cells; higher values suit compact numeric grids.",
|
|
507
|
+
"text_table_col_similarity": 0.36,
|
|
508
|
+
"@text_table_alignment_type<desc>": "Table column alignment type [0 - strong, 1 - average, 2 - weak]. Select strong for strictly aligned table elements.",
|
|
509
|
+
"@text_table_alignment_type<use>": "Table detection. Use strong for strictly aligned tables.",
|
|
510
|
+
"text_table_alignment_type": 2,
|
|
511
|
+
"@text_table_alignment_distance<desc>": "Maximum alignment deviation (points) when scoring text regions as a single table.",
|
|
512
|
+
"@text_table_alignment_distance<use>": "Higher values favor recognizing a region as one table; lower values keep misaligned blocks as separate paragraphs.",
|
|
513
|
+
"text_table_alignment_distance": 0.4,
|
|
514
|
+
"@text_table_text_col_w1<desc>": "Weight of the text-vs-paragraph similarity term when scoring a candidate text-table column.",
|
|
515
|
+
"@text_table_text_col_w1<use>": "Raise to insist on paragraph-like consistency inside a column; lower to let mixed typography still form a column.",
|
|
516
|
+
"text_table_text_col_w1": 1,
|
|
517
|
+
"@text_table_text_col_w2<desc>": "Weight of the column-width similarity term in the same text-column distance blend.",
|
|
518
|
+
"@text_table_text_col_w2<use>": "Tune relative to text_table_text_col_w1 when width regularity matters more or less than text coherence.",
|
|
519
|
+
"text_table_text_col_w2": 1,
|
|
520
|
+
"@text_table_text_col_width_min<desc>": "Minimum text-column width in points; if 0, effective min is text_table_text_col_width_min_ratio × page font size.",
|
|
521
|
+
"@text_table_text_col_width_min<use>": "Filters out spurious narrow columns; set a fixed min in points when font scaling is wrong.",
|
|
522
|
+
"text_table_text_col_width_min": 0,
|
|
523
|
+
"@text_table_text_col_width_min_ratio<desc>": "Used only when text_table_text_col_width_min is 0: scales the minimum column width from page font size.",
|
|
524
|
+
"@text_table_text_col_width_min_ratio<use>": "Raise to require wider columns in large-type PDFs.",
|
|
525
|
+
"text_table_text_col_width_min_ratio": 1,
|
|
526
|
+
"@text_table_text_col_width_max<desc>": "Maximum text-column width in points; if 0, effective max is text_table_text_col_width_max_ratio × page font size.",
|
|
527
|
+
"@text_table_text_col_width_max<use>": "Caps single-column prose masquerading as a table column.",
|
|
528
|
+
"text_table_text_col_width_max": 0,
|
|
529
|
+
"@text_table_text_col_width_max_ratio<desc>": "Used only when text_table_text_col_width_max is 0: scales the maximum column width from page font size.",
|
|
530
|
+
"@text_table_text_col_width_max_ratio<use>": "Lower to reject full-page-width “columns”; raise for landscape tables.",
|
|
531
|
+
"text_table_text_col_width_max_ratio": 8,
|
|
532
|
+
"@text_table_image_col_w1<desc>": "Weight for the blended image-column distance (size spread and optional graphics-state check).",
|
|
533
|
+
"@text_table_image_col_w1<use>": "Increase to hard-reject image columns with uneven cell sizes; decrease when icons of different sizes should still align as one column.",
|
|
534
|
+
"text_table_image_col_w1": 1,
|
|
535
|
+
"@text_table_image_col_gs<desc>": "If non-zero, images in the same column candidate must share compatible graphics state (line width, fill/stroke type and opacity).",
|
|
536
|
+
"@text_table_image_col_gs<use>": "Disable (0) when mixed drawing styles still belong to one column; keep enabled for strict visual uniformity.",
|
|
537
|
+
"text_table_image_col_gs": 1,
|
|
538
|
+
"@text_table_image_col_width_min<desc>": "Minimum image-column width in points; if 0, use text_table_image_col_width_min_ratio × page font size.",
|
|
539
|
+
"@text_table_image_col_width_min<use>": "Drop tiny icon columns from text-table hypotheses.",
|
|
540
|
+
"text_table_image_col_width_min": 0,
|
|
541
|
+
"@text_table_image_col_width_min_ratio<desc>": "Used when text_table_image_col_width_min is 0.",
|
|
542
|
+
"@text_table_image_col_width_min_ratio<use>": "Scales minimum image-column width with body font size.",
|
|
543
|
+
"text_table_image_col_width_min_ratio": 1,
|
|
544
|
+
"@text_table_image_col_width_max<desc>": "Maximum image-column width in points; if 0, use text_table_image_col_width_max_ratio × page font size.",
|
|
545
|
+
"@text_table_image_col_width_max<use>": "Prevent a single wide figure column from defining the whole table width.",
|
|
546
|
+
"text_table_image_col_width_max": 0,
|
|
547
|
+
"@text_table_image_col_width_max_ratio<desc>": "Used when text_table_image_col_width_max is 0.",
|
|
548
|
+
"@text_table_image_col_width_max_ratio<use>": "Upper bound on image column width as a multiple of font size.",
|
|
549
|
+
"text_table_image_col_width_max_ratio": 4,
|
|
550
|
+
"@text_table_image_col_height_min<desc>": "Minimum image-column height in points; if 0, use text_table_image_col_height_min_ratio × page font size.",
|
|
551
|
+
"@text_table_image_col_height_min<use>": "Filters out one-pixel-tall image strips as column anchors.",
|
|
552
|
+
"text_table_image_col_height_min": 0,
|
|
553
|
+
"@text_table_image_col_height_min_ratio<desc>": "Used when text_table_image_col_height_min is 0.",
|
|
554
|
+
"@text_table_image_col_height_min_ratio<use>": "Scales minimum column height with font size.",
|
|
555
|
+
"text_table_image_col_height_min_ratio": 1,
|
|
556
|
+
"@text_table_image_col_height_max<desc>": "Maximum image-column height in points; if 0, use text_table_image_col_height_max_ratio × page font size.",
|
|
557
|
+
"@text_table_image_col_height_max<use>": "Caps columns dominated by a full-height image.",
|
|
558
|
+
"text_table_image_col_height_max": 0,
|
|
559
|
+
"@text_table_image_col_height_max_ratio<desc>": "Used when text_table_image_col_height_max is 0.",
|
|
560
|
+
"@text_table_image_col_height_max_ratio<use>": "Maximum image column height as a multiple of font size.",
|
|
561
|
+
"text_table_image_col_height_max_ratio": 2,
|
|
562
|
+
"@text_table_image_similarity_w1<desc>": "Weight for image-area similarity in the section image score.",
|
|
563
|
+
"@text_table_image_similarity_w1<use>": "Raise when image footprint should dominate the text-table vs figure decision.",
|
|
564
|
+
"text_table_image_similarity_w1": 1,
|
|
565
|
+
"@text_table_image_similarity_w2<desc>": "Weight applied when chart similarity overrides the area term in that blend.",
|
|
566
|
+
"@text_table_image_similarity_w2<use>": "Raise when chart-like path density should outweigh raw image area.",
|
|
567
|
+
"text_table_image_similarity_w2": 1,
|
|
568
|
+
"@text_table_image_similarity<desc>": "If section image similarity is at or above this value [0,1], the overlap is rejected as not a text table (too figure-like).",
|
|
569
|
+
"@text_table_image_similarity<use>": "Lower to reject table joins sooner near figures; raise to keep evaluating grid structure when images sit inside the candidate region.",
|
|
570
|
+
"text_table_image_similarity": 0.7,
|
|
571
|
+
"@text_table_paragraph_similarity<desc>": "If section paragraph similarity is at or above this value [0,1], the overlap is rejected as not a text table (too prose-like).",
|
|
572
|
+
"@text_table_paragraph_similarity<use>": "Lower to stop false tables when text still reads like body copy; raise only when dense paragraph-like grids must remain table candidates.",
|
|
573
|
+
"text_table_paragraph_similarity": 0.7,
|
|
574
|
+
"@table_update_delete_empty<desc>": "Default preference for post-detection cleanup: when enabled in templates, empty rows and columns are dropped so hollow grids do not linger.",
|
|
575
|
+
"@table_update_delete_empty<use>": "Keep on for clean tagging; disable when empty cells carry semantic meaning you must preserve.",
|
|
576
|
+
"table_update_delete_empty": 1,
|
|
577
|
+
"@table_update_split_by_cell<desc>": "When enabled, oversized merged cells are subdivided—e.g. a paragraph mistaken for one cell, or bullets and body text glued in one cell.",
|
|
578
|
+
"@table_update_split_by_cell<use>": "Enable when one TD holds multiple logical rows or list+paragraph mixes.",
|
|
579
|
+
"table_update_split_by_cell": 0,
|
|
580
|
+
"@table_update_split_by_row<desc>": "When enabled, multi-line cell content is broken into one line per row for downstream table semantics.",
|
|
581
|
+
"@table_update_split_by_row<use>": "Use for forms where each visual row inside a cell should become its own structure row.",
|
|
582
|
+
"table_update_split_by_row": 0,
|
|
583
|
+
"@table_update_split_label<desc>": "When enabled, label markers inside table cells are separated from body text where the detector fused them.",
|
|
584
|
+
"@table_update_split_label<use>": "Helps list markers in table cells become separate structure from cell body text.",
|
|
585
|
+
"table_update_split_label": 0,
|
|
586
|
+
"@table_update_span_empty<desc>": "When enabled, visually empty cells are merged using row/column span metadata instead of leaving many blank boxes.",
|
|
587
|
+
"@table_update_span_empty<use>": "Typical for financial tables with intentional blank spans; disable if every empty cell must remain explicit.",
|
|
588
|
+
"table_update_span_empty": 1,
|
|
589
|
+
"@table_update_span_row<desc>": "When enabled, rows are coalesced according to the largest consistent row span detected in the grid.",
|
|
590
|
+
"@table_update_span_row<use>": "Enable to collapse subrows that belong to one logical rowspan.",
|
|
591
|
+
"table_update_span_row": 0,
|
|
592
|
+
"@table_update_span_row_first<desc>": "Default order for span repair: when enabled, row-span style merging is preferred before column-span merging; when disabled, column direction runs first. Applies to post-process span_empty-style passes and to the graphic-table isolated-element span pass (same meaning as template span_row_first on table_update).",
|
|
593
|
+
"@table_update_span_row_first<use>": "Enable if row-oriented blank spans are wrong after the first pass; disable if column merges should run first on complex grids.",
|
|
594
|
+
"table_update_span_row_first": 0,
|
|
595
|
+
"@table_update_join<desc>": "When enabled, all text fragments inside one cell are concatenated into a single paragraph element.",
|
|
596
|
+
"@table_update_join<use>": "Enable for export pipelines that expect one text node per cell; keep off when fragments must stay separate.",
|
|
597
|
+
"table_update_join": 0,
|
|
598
|
+
"@table_update_cell_header<desc>": "When enabled, header rows/cells are inferred during table post-processing for accessibility and layout roles.",
|
|
599
|
+
"@table_update_cell_header<use>": "Keep on for TH/scope inference; disable only when headers are tagged manually elsewhere.",
|
|
600
|
+
"table_update_cell_header": 1,
|
|
601
|
+
"@table_span_col_ratio<desc>": "Fraction [0,1] of cell span width: isolated content intersecting adjacent graphic cells must cover at least this much of each cell’s span bbox horizontally before colspan-style merging is allowed (path-based tables with floating text).",
|
|
602
|
+
"@table_span_col_ratio<use>": "Raise to stop weak horizontal overlaps from merging columns; lower if true colspan is missed because overlap looks small.",
|
|
603
|
+
"table_span_col_ratio": 0.1,
|
|
604
|
+
"@table_span_row_ratio<desc>": "Fraction [0,1] of cell span height: same idea as table_span_col_ratio but for rowspan decisions; combined with text_height caps when measuring how much of an isolated element intersects stacked cells.",
|
|
605
|
+
"@table_span_row_ratio<use>": "Raise to require stronger vertical overlap before rowspan; lower if rowspan is missed. Tune together with text_height on fine or large type.",
|
|
606
|
+
"table_span_row_ratio": 0.2,
|
|
607
|
+
"@table_alignment_h<desc>": "Maximum slack in points when comparing vertical positions (top, bottom, or vertical center) of content in adjacent cells on the same row to infer row-wise alignment.",
|
|
608
|
+
"@table_alignment_h<use>": "Increase if baselines or row edges jitter between columns; decrease if alignment is classified into the wrong bucket (top vs bottom vs center).",
|
|
609
|
+
"table_alignment_h": 1,
|
|
610
|
+
"@table_alignment_v<desc>": "Maximum slack in points when comparing horizontal positions (left, right, or horizontal center) of content in adjacent cells in the same column to infer column-wise alignment.",
|
|
611
|
+
"@table_alignment_v<use>": "Increase if left/right edges drift between rows; decrease if neighbors are bucketed into the wrong horizontal alignment.",
|
|
612
|
+
"table_alignment_v": 4,
|
|
613
|
+
"@table_line_intersection<desc>": "Padding in points around strokes when testing intersections: graphic table construction from paths, and assigning text lines to the correct graphic cell during nesting.",
|
|
614
|
+
"@table_line_intersection<use>": "Increase slightly if rules almost touch content or cells but associations fail; decrease if text or geometry attaches across a visible rule incorrectly.",
|
|
615
|
+
"table_line_intersection": 1,
|
|
616
|
+
"@form_table_detect<desc>": "If 1, widget/form field layouts may be interpreted as tabular structure where applicable.",
|
|
617
|
+
"@form_table_detect<use>": "Disable on pure reading-order PDFs if form chrome should not influence table detection.",
|
|
618
|
+
"form_table_detect": 1,
|
|
619
|
+
"@element_prop_detect<desc>": "Update elements alternate description, actual text, expansion and annotation Contents key. Default is 1; set to 0 to skip for performance when alternate text is not required.",
|
|
620
|
+
"@element_prop_detect<use>": "The default algorithm automatically detects Alt for images, and Contents or TU for annotations. If you need different behavior, use the 'element_prop_update' function.",
|
|
621
|
+
"element_prop_detect": 1,
|
|
622
|
+
"@rd_sort<desc>": "Reading-order sort mode: 0 built-in heuristic, 1 PDF content stream order, 2 geometric (xy), 3 explicit rd_index from template.",
|
|
623
|
+
"@rd_sort<use>": "Use 2 for simply laid-out pages; use 1 when stream order matches author intent; use 3 only when you assign rd_index in rules.",
|
|
624
|
+
"rd_sort": 0,
|
|
625
|
+
"@rd_sort_direction<desc>": "When using built-in or xy geometric sort: 0 default, 1 favor column-major traversal, 2 favor row-major traversal.",
|
|
626
|
+
"@rd_sort_direction<use>": "Switch to column-first for multi-column newspapers; row-first for mostly linear layouts.",
|
|
627
|
+
"rd_sort_direction": 0,
|
|
628
|
+
"@rd_column_distance<desc>": "Maximum horizontal gap in points between elements tentatively placed in the same column during reading-order grouping.",
|
|
629
|
+
"@rd_column_distance<use>": "Increase for wide column gutters; decrease if adjacent columns are being read in the wrong order.",
|
|
630
|
+
"rd_column_distance": 0.8,
|
|
631
|
+
"@rd_overlap_min_ratio<desc>": "For reading-order overlap tests, require the shared span on the overlap axis to be at least this fraction of the smaller box width (horizontal) or smaller height (vertical). 0 keeps the legacy rule where any positive overlap counts.",
|
|
632
|
+
"@rd_overlap_min_ratio<use>": "Raise (e.g. 0.02–0.08) when two columns barely graze in X or Y and should not be treated as one column for sorting or grouping; keep 0 unless you see wrong order from sub-pixel or hairline touches.",
|
|
633
|
+
"rd_overlap_min_ratio": 0.02
|
|
634
|
+
}
|
|
635
|
+
],
|
|
636
|
+
"@pagemap_regex<type>": "string",
|
|
637
|
+
"@pagemap_regex<desc>": "The pagemap_regex node contains patterns for the recognition algorithm. Keys prefixed with regex use regular expressions; others define acceptable characters.",
|
|
638
|
+
"@pagemap_regex<value>": [
|
|
639
|
+
"regex_hyphen",
|
|
640
|
+
"regex_hyphen_rtl",
|
|
641
|
+
"regex_bullet",
|
|
642
|
+
"regex_bullet_rtl",
|
|
643
|
+
"regex_bullet_font",
|
|
644
|
+
"regex_label",
|
|
645
|
+
"regex_label_rtl",
|
|
646
|
+
"regex_decimal_numbering",
|
|
647
|
+
"regex_roman_numbering",
|
|
648
|
+
"regex_roman_numbering_rtl",
|
|
649
|
+
"regex_letter_numbering",
|
|
650
|
+
"regex_letter_numbering_rtl",
|
|
651
|
+
"regex_filling",
|
|
652
|
+
"regex_filling_char",
|
|
653
|
+
"regex_page_number",
|
|
654
|
+
"regex_page_number_rtl",
|
|
655
|
+
"regex_first_cap",
|
|
656
|
+
"regex_first_cap_rtl",
|
|
657
|
+
"regex_terminal",
|
|
658
|
+
"regex_terminal_rtl",
|
|
659
|
+
"regex_table_caption",
|
|
660
|
+
"regex_image_caption",
|
|
661
|
+
"regex_chart_caption",
|
|
662
|
+
"regex_note_caption",
|
|
663
|
+
"regex_toc_caption",
|
|
664
|
+
"regex_colon",
|
|
665
|
+
"regex_colon_rtl",
|
|
666
|
+
"regex_comma",
|
|
667
|
+
"regex_letter",
|
|
668
|
+
"regex_letter_rtl",
|
|
669
|
+
"number_chars",
|
|
670
|
+
"numbering_splitter_chars"
|
|
671
|
+
],
|
|
672
|
+
"@pagemap_regex<insert>": [
|
|
673
|
+
{}
|
|
674
|
+
],
|
|
675
|
+
"pagemap_regex": [
|
|
676
|
+
{
|
|
677
|
+
"statement": "$if",
|
|
678
|
+
"@query": {},
|
|
679
|
+
"regex_hyphen": "\\w+-$",
|
|
680
|
+
"regex_bullet": "^[\\u2010\\u2011\\u2212\\u005E\\u005B\\uF0A7\\uF097\\uF0BB\\u25CF\\u2022\\u25D8\\u25CB\\u25D9\\u2023\\u2043\\uF0B7\\u2212\\u204C\\u204D\\u25E6\\u29BE\\u29BF\\u21E8\\u25BA\\u25C4\\u2219\\u25A0\\uF06C\\u25A1\\u005D\\u25C6\\u006F]$",
|
|
681
|
+
"regex_bullet_font": "(Wingdings)|(Symbol)",
|
|
682
|
+
"regex_label": "^[\\[\\(]?((M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3}))|(\\d+)|([a-zA-Z]))[\\)\\]\\.]$",
|
|
683
|
+
"label_chars": ".()[]",
|
|
684
|
+
"regex_decimal_numbering": "^[\\[\\(]?(?:\\d{1,4}\\.){0,5}\\d{0,4}\\s?[\\)\\]\\.]?$",
|
|
685
|
+
"regex_roman_numbering": "^[\\[\\(]?M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})[\\)\\]\\.]?$",
|
|
686
|
+
"regex_letter_numbering": "^[\\[\\(]?[A-Za-z][\\)\\]\\.]$",
|
|
687
|
+
"regex_filling": "[._]{2,}",
|
|
688
|
+
"regex_filling_chars": "._",
|
|
689
|
+
"regex_page_number": "(^\\d+$)|(^M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})$)",
|
|
690
|
+
"regex_first_cap": "^[A-Z]",
|
|
691
|
+
"regex_terminal": "[\\.\\!\\?]$",
|
|
692
|
+
"regex_table_caption": "((^table)|(^tab\\.))",
|
|
693
|
+
"regex_image_caption": "((^image)|(^img\\.)|(^figure)|(^fig\\.))",
|
|
694
|
+
"regex_chart_caption": "((^chart)|(^map))",
|
|
695
|
+
"regex_note_caption": "((^source\\:)|(^note\\:))",
|
|
696
|
+
"regex_toc_caption": "((^content)|(^toc))",
|
|
697
|
+
"regex_colon": ":$",
|
|
698
|
+
"regex_comma": "[,;]$",
|
|
699
|
+
"regex_letter": "^[A-Za-z]$",
|
|
700
|
+
"number_chars": "-+.,%\\u20AC$\\u00A5\\u00A3",
|
|
701
|
+
"numbering_splitter_chars": ".()[]"
|
|
702
|
+
}
|
|
703
|
+
],
|
|
704
|
+
"@element_create<type>": "function",
|
|
705
|
+
"@element_create<desc>": "Define initial regions (anchors) such as pde_text, pde_list, or pde_table and optional template fields (flag, label, numbering, parent, and others) applied when those regions are recognized.",
|
|
706
|
+
"@element_create<use>": "Seed the layout with known bboxes or table shells; nest element_template for per-anchor overrides. Initial elements can pin list/table structure before auto-detection runs.",
|
|
707
|
+
"@element_create<insert>": [
|
|
708
|
+
{
|
|
709
|
+
"query": {
|
|
710
|
+
"$and": []
|
|
711
|
+
},
|
|
712
|
+
"elements": [
|
|
713
|
+
{
|
|
714
|
+
"type": "pde_text",
|
|
715
|
+
"bbox": {
|
|
716
|
+
"left": "0",
|
|
717
|
+
"bottom": "0",
|
|
718
|
+
"right": "0",
|
|
719
|
+
"top": "0"
|
|
720
|
+
}
|
|
721
|
+
}
|
|
722
|
+
]
|
|
723
|
+
}
|
|
724
|
+
],
|
|
725
|
+
"@element_create": [
|
|
726
|
+
{
|
|
727
|
+
"@disable": false,
|
|
728
|
+
"@disable<link>": "schema:keywords:disable",
|
|
729
|
+
"@statement": "$if",
|
|
730
|
+
"@statement<link>": "schema:keywords:statement",
|
|
731
|
+
"@query<insert>": {
|
|
732
|
+
"$and": []
|
|
733
|
+
},
|
|
734
|
+
"@query": {
|
|
735
|
+
"$and": []
|
|
736
|
+
},
|
|
737
|
+
"@elements<desc>": "Array of pre-defined elements.",
|
|
738
|
+
"@elements<type>": "array",
|
|
739
|
+
"@elements": [
|
|
740
|
+
{
|
|
741
|
+
"@alt<link>": "schema:values:alt",
|
|
742
|
+
"@alt": "",
|
|
743
|
+
"@contents<link>": "schema:values:contents",
|
|
744
|
+
"@contents": "",
|
|
745
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
746
|
+
"@actual_text": "",
|
|
747
|
+
"@expansion<link>": "schema:values:expansion",
|
|
748
|
+
"@expansion": "",
|
|
749
|
+
"@lang<link>": "schema:values:lang",
|
|
750
|
+
"@lang": "",
|
|
751
|
+
"@flag<link>": "schema:values:flag",
|
|
752
|
+
"@flag": "",
|
|
753
|
+
"@text_flag<link>": "schema:values:text_flag",
|
|
754
|
+
"@text_flag": "",
|
|
755
|
+
"@numbering<link>": "schema:values:numbering",
|
|
756
|
+
"@numbering": "",
|
|
757
|
+
"@label<link>": "schema:values:label",
|
|
758
|
+
"@label": "",
|
|
759
|
+
"@label_text<link>": "schema:values:label_text",
|
|
760
|
+
"@label_text": "",
|
|
761
|
+
"@type<desc>": "Set the element type; it drives which layout and tagging rules apply.",
|
|
762
|
+
"@type<type>": "string",
|
|
763
|
+
"@type<value>": [
|
|
764
|
+
"pde_text",
|
|
765
|
+
"pde_image",
|
|
766
|
+
"pde_container",
|
|
767
|
+
"pde_list",
|
|
768
|
+
"pde_line",
|
|
769
|
+
"pde_rect",
|
|
770
|
+
"pde_table",
|
|
771
|
+
"pde_cell",
|
|
772
|
+
"pde_toc",
|
|
773
|
+
"pde_header",
|
|
774
|
+
"pde_footer"
|
|
775
|
+
],
|
|
776
|
+
"type": "pde_element",
|
|
777
|
+
"@bbox<link>": "schema:values:bbox",
|
|
778
|
+
"bbox": {
|
|
779
|
+
"left": "0",
|
|
780
|
+
"bottom": "0",
|
|
781
|
+
"right": "0",
|
|
782
|
+
"top": "0"
|
|
783
|
+
},
|
|
784
|
+
"@start_bbox<link>": "schema:values:bbox",
|
|
785
|
+
"@start_bbox": {
|
|
786
|
+
"left": "0",
|
|
787
|
+
"bottom": "0",
|
|
788
|
+
"right": "0",
|
|
789
|
+
"top": "0"
|
|
790
|
+
},
|
|
791
|
+
"@end_bbox<link>": "schema:values:bbox",
|
|
792
|
+
"@end_bbox": {
|
|
793
|
+
"left": "0",
|
|
794
|
+
"bottom": "0",
|
|
795
|
+
"right": "0",
|
|
796
|
+
"top": "0"
|
|
797
|
+
},
|
|
798
|
+
"@tag<link>": "schema:values:tag_type",
|
|
799
|
+
"@tag": "",
|
|
800
|
+
"@heading<link>": "schema:values:heading",
|
|
801
|
+
"@heading": "normal",
|
|
802
|
+
"@name<link>": "schema:values:name",
|
|
803
|
+
"@name": "element1",
|
|
804
|
+
"@parent<link>": "schema:values:parent",
|
|
805
|
+
"@parent": "",
|
|
806
|
+
"@caption<link>": "schema:values:caption",
|
|
807
|
+
"@caption": "",
|
|
808
|
+
"@id<link>": "schema:values:id",
|
|
809
|
+
"@id": "",
|
|
810
|
+
"@sort_direction<desc>": "Specifies the primary direction for sort: 0 - inbuild, 1 - columns, 2 - rows.",
|
|
811
|
+
"@sort_direction<type>": "int",
|
|
812
|
+
"@sort_direction": 0,
|
|
813
|
+
"@rd_index<desc>": "Specifies the reading order index of the element in the parent element or the page. Starts at 0 for the first element. Only used when rd_sort is set to 3 (explicit rd_index).",
|
|
814
|
+
"@rd_index<type>": "int",
|
|
815
|
+
"@rd_index": -1,
|
|
816
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
817
|
+
"@single_instance": "font_size|font_name|left",
|
|
818
|
+
"@col_num": 1,
|
|
819
|
+
"@col_num<link>": "schema:values:col_num",
|
|
820
|
+
"@row_num": 1,
|
|
821
|
+
"@row_num<link>": "schema:values:row_num",
|
|
822
|
+
"@cell_column": "1",
|
|
823
|
+
"@cell_column<link>": "schema:values:cell_column",
|
|
824
|
+
"@cell_row": "1",
|
|
825
|
+
"@cell_row<link>": "schema:values:cell_row",
|
|
826
|
+
"@cell_row_span": "1",
|
|
827
|
+
"@cell_row_span<link>": "schema:values:cell_row_span",
|
|
828
|
+
"@cell_column_span": "1",
|
|
829
|
+
"@cell_column_span<link>": "schema:values:cell_column_span",
|
|
830
|
+
"@cell_scope": "column",
|
|
831
|
+
"@cell_scope<link>": "schema:values:cell_scope",
|
|
832
|
+
"@cell_header": false,
|
|
833
|
+
"@cell_header<link>": "schema:values:cell_header",
|
|
834
|
+
"@cell_associated_header<link>": "schema:values:associated_header",
|
|
835
|
+
"@cell_associated_header": "",
|
|
836
|
+
"@splitter": "pde_table",
|
|
837
|
+
"@splitter<link>": "schema:values:type",
|
|
838
|
+
"@splitter<desc>": "Used together with flag \"splitter\": element types (pde_table, pde_list, pde_text, …) whose active cross-page continuous chains are cleared at this element. Multiple types may be separated with |.",
|
|
839
|
+
"@toc<link>": "schema:values:label",
|
|
840
|
+
"@toc": "",
|
|
841
|
+
"@element_template<desc>": "Defines the template configuration that will be applied to this element.",
|
|
842
|
+
"@element_template<insert>": {
|
|
843
|
+
"template": {
|
|
844
|
+
"pagemap": [
|
|
845
|
+
{}
|
|
846
|
+
]
|
|
847
|
+
}
|
|
848
|
+
},
|
|
849
|
+
"@element_template": {
|
|
850
|
+
"template": {}
|
|
851
|
+
}
|
|
852
|
+
}
|
|
853
|
+
]
|
|
854
|
+
}
|
|
855
|
+
],
|
|
856
|
+
"@object_update<type>": "function",
|
|
857
|
+
"@object_update<desc>": "Runs while each low-level PDF content object (pds_text, pds_path, pds_image, pds_form, …) is classified before it becomes pde_* layout elements.",
|
|
858
|
+
"@object_update<use>": "Match pds_* types to set flags or names early, or override substructure_form_xobject per Form XObject before its content is expanded into the layout tree.",
|
|
859
|
+
"@object_update<insert>": [
|
|
860
|
+
{
|
|
861
|
+
"query": {
|
|
862
|
+
"param": [
|
|
863
|
+
"pds_object"
|
|
864
|
+
],
|
|
865
|
+
"$and": []
|
|
866
|
+
}
|
|
867
|
+
}
|
|
868
|
+
],
|
|
869
|
+
"@object_update": [
|
|
870
|
+
{
|
|
871
|
+
"@disable": false,
|
|
872
|
+
"@disable<link>": "schema:keywords:disable",
|
|
873
|
+
"@statement": "$if",
|
|
874
|
+
"@statement<link>": "schema:keywords:statement",
|
|
875
|
+
"@query": {
|
|
876
|
+
"@param<type>": "query_param",
|
|
877
|
+
"@param<insert>": [
|
|
878
|
+
"pds_object"
|
|
879
|
+
],
|
|
880
|
+
"param": [
|
|
881
|
+
[
|
|
882
|
+
"pds_object",
|
|
883
|
+
"pds_text",
|
|
884
|
+
"pds_path",
|
|
885
|
+
"pds_image",
|
|
886
|
+
"pds_shading",
|
|
887
|
+
"pds_form"
|
|
888
|
+
]
|
|
889
|
+
]
|
|
890
|
+
},
|
|
891
|
+
"@flag<link>": "schema:values:flag",
|
|
892
|
+
"@flag": "",
|
|
893
|
+
"@name<link>": "schema:values:name",
|
|
894
|
+
"@name": "object1",
|
|
895
|
+
"@label<link>": "schema:values:label",
|
|
896
|
+
"@label": "",
|
|
897
|
+
"substructure_form_xobject": "true",
|
|
898
|
+
"substructure_form_xobject<desc>": "If false for Form XObject, the Form XObject is treated as a single object; otherwise, its content is tagged individually.",
|
|
899
|
+
"substructure_form_xobject<type>": "bool",
|
|
900
|
+
"substructure_form_xobject<value>": [
|
|
901
|
+
"true",
|
|
902
|
+
"false"
|
|
903
|
+
]
|
|
904
|
+
}
|
|
905
|
+
],
|
|
906
|
+
"@text_run_update<type>": "function",
|
|
907
|
+
"@text_run_update<desc>": "Runs per pde_text_run after PDF text objects are parsed: adjust text_state_flag and other run-level keys before words are built.",
|
|
908
|
+
"@text_run_update<use>": "Rarely needed; use when a specific font or graphics state on a run should drive downstream splitting or suppression.",
|
|
909
|
+
"@text_run_update<insert>": [
|
|
910
|
+
{
|
|
911
|
+
"query": {
|
|
912
|
+
"param": [
|
|
913
|
+
"pde_text_run"
|
|
914
|
+
],
|
|
915
|
+
"$and": []
|
|
916
|
+
}
|
|
917
|
+
}
|
|
918
|
+
],
|
|
919
|
+
"@text_run_update": [
|
|
920
|
+
{
|
|
921
|
+
"@disable": false,
|
|
922
|
+
"@disable<link>": "schema:keywords:disable",
|
|
923
|
+
"@statement": "$if",
|
|
924
|
+
"@statement<link>": "schema:keywords:statement",
|
|
925
|
+
"@query": {
|
|
926
|
+
"@param<type>": "query_param",
|
|
927
|
+
"@param<insert>": [
|
|
928
|
+
"pde_text_run"
|
|
929
|
+
],
|
|
930
|
+
"param": [
|
|
931
|
+
[
|
|
932
|
+
"pde_text_run"
|
|
933
|
+
]
|
|
934
|
+
]
|
|
935
|
+
},
|
|
936
|
+
"@text_state_flag<link>": "schema:values:text_state_flag",
|
|
937
|
+
"@text_state_flag": ""
|
|
938
|
+
}
|
|
939
|
+
],
|
|
940
|
+
"@text_run_neighbours<type>": "function",
|
|
941
|
+
"@text_run_neighbours<desc>": "Binary rule on adjacent pde_text_run segments (left then right) while runs are merged into words or lines.",
|
|
942
|
+
"@text_run_neighbours<use>": "Same role as word_neighbours but at text-run granularity for awkward font-encoded PDFs.",
|
|
943
|
+
"@text_run_neighbours<rule>": "First parameter is the left text run, second is the right text run.",
|
|
944
|
+
"@text_run_neighbours<insert>": [
|
|
945
|
+
{
|
|
946
|
+
"query": {
|
|
947
|
+
"param": [
|
|
948
|
+
"pde_text_run",
|
|
949
|
+
"pde_text_run"
|
|
950
|
+
],
|
|
951
|
+
"$and": []
|
|
952
|
+
}
|
|
953
|
+
}
|
|
954
|
+
],
|
|
955
|
+
"@text_run_neighbours": [
|
|
956
|
+
{
|
|
957
|
+
"@disable": false,
|
|
958
|
+
"@disable<link>": "schema:keywords:disable",
|
|
959
|
+
"@statement": "$if",
|
|
960
|
+
"@statement<link>": "schema:keywords:statement",
|
|
961
|
+
"@query": {
|
|
962
|
+
"@param<type>": "query_param",
|
|
963
|
+
"@param<insert>": [
|
|
964
|
+
"pde_text_run",
|
|
965
|
+
"pde_text_run"
|
|
966
|
+
],
|
|
967
|
+
"param": [
|
|
968
|
+
[
|
|
969
|
+
"pde_text_run"
|
|
970
|
+
],
|
|
971
|
+
[
|
|
972
|
+
"pde_text_run"
|
|
973
|
+
]
|
|
974
|
+
]
|
|
975
|
+
},
|
|
976
|
+
"@join<desc>": "If true, the two matched text runs may merge into one line segment; if false, a hard boundary is kept between them.",
|
|
977
|
+
"@join<type>": "bool",
|
|
978
|
+
"@join<value>": [
|
|
979
|
+
"true",
|
|
980
|
+
"false"
|
|
981
|
+
],
|
|
982
|
+
"@join": true
|
|
983
|
+
}
|
|
984
|
+
],
|
|
985
|
+
"@line_update<type>": "function",
|
|
986
|
+
"@line_update<desc>": "Runs once per detected text line after path-based line detection. Matching rules here can set header, footer, artifact, parent, tag, alt, and related template keys on that line.",
|
|
987
|
+
"@line_update<use>": "Promote separators to artifacts, move running rules to header/footer bands, or rename lines for anchors before tables use them as guides.",
|
|
988
|
+
"@line_update<insert>": [
|
|
989
|
+
{
|
|
990
|
+
"query": {
|
|
991
|
+
"param": [
|
|
992
|
+
"pde_line"
|
|
993
|
+
],
|
|
994
|
+
"$and": []
|
|
995
|
+
}
|
|
996
|
+
}
|
|
997
|
+
],
|
|
998
|
+
"@line_update": [
|
|
999
|
+
{
|
|
1000
|
+
"@disable": false,
|
|
1001
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1002
|
+
"@statement": "$if",
|
|
1003
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1004
|
+
"@query": {
|
|
1005
|
+
"@param<type>": "query_param",
|
|
1006
|
+
"@param<insert>": [
|
|
1007
|
+
"pde_line"
|
|
1008
|
+
],
|
|
1009
|
+
"param": [
|
|
1010
|
+
[
|
|
1011
|
+
"pde_line"
|
|
1012
|
+
]
|
|
1013
|
+
]
|
|
1014
|
+
},
|
|
1015
|
+
"@flag<link>": "schema:values:flag",
|
|
1016
|
+
"@flag": "",
|
|
1017
|
+
"@alt<link>": "schema:values:alt",
|
|
1018
|
+
"@alt": "",
|
|
1019
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1020
|
+
"@actual_text": "",
|
|
1021
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1022
|
+
"@expansion": "",
|
|
1023
|
+
"@lang<link>": "schema:values:lang",
|
|
1024
|
+
"@lang": "",
|
|
1025
|
+
"@label<link>": "schema:values:label",
|
|
1026
|
+
"@label": "",
|
|
1027
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1028
|
+
"@tag": "",
|
|
1029
|
+
"@name<link>": "schema:values:name",
|
|
1030
|
+
"@name": "line1",
|
|
1031
|
+
"@parent<link>": "schema:values:parent",
|
|
1032
|
+
"@parent": ""
|
|
1033
|
+
}
|
|
1034
|
+
],
|
|
1035
|
+
"@rect_update<type>": "function",
|
|
1036
|
+
"@rect_update<desc>": "Runs per pde_rect still stored in the line-elements list after rect detection and extension: same promotion path as line_update (header/footer/artifact).",
|
|
1037
|
+
"@rect_update<use>": "Treat decorative boxes differently from structural lines, or strip rects that should not participate in graphic-table clustering.",
|
|
1038
|
+
"@rect_update<insert>": [
|
|
1039
|
+
{
|
|
1040
|
+
"query": {
|
|
1041
|
+
"param": [
|
|
1042
|
+
"pde_rect"
|
|
1043
|
+
],
|
|
1044
|
+
"$and": []
|
|
1045
|
+
}
|
|
1046
|
+
}
|
|
1047
|
+
],
|
|
1048
|
+
"@rect_update": [
|
|
1049
|
+
{
|
|
1050
|
+
"@disable": false,
|
|
1051
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1052
|
+
"@statement": "$if",
|
|
1053
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1054
|
+
"@query": {
|
|
1055
|
+
"@param<type>": "query_param",
|
|
1056
|
+
"@param<insert>": [
|
|
1057
|
+
"pde_rect"
|
|
1058
|
+
],
|
|
1059
|
+
"param": [
|
|
1060
|
+
[
|
|
1061
|
+
"pde_rect"
|
|
1062
|
+
]
|
|
1063
|
+
]
|
|
1064
|
+
},
|
|
1065
|
+
"@flag<link>": "schema:values:flag",
|
|
1066
|
+
"@flag": "",
|
|
1067
|
+
"@alt<link>": "schema:values:alt",
|
|
1068
|
+
"@alt": "",
|
|
1069
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1070
|
+
"@actual_text": "",
|
|
1071
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1072
|
+
"@expansion": "",
|
|
1073
|
+
"@lang<link>": "schema:values:lang",
|
|
1074
|
+
"@lang": "",
|
|
1075
|
+
"@label<link>": "schema:values:label",
|
|
1076
|
+
"@label": "",
|
|
1077
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1078
|
+
"@tag": "",
|
|
1079
|
+
"@name<link>": "schema:values:name",
|
|
1080
|
+
"@name": "rect1",
|
|
1081
|
+
"@parent<link>": "schema:values:parent",
|
|
1082
|
+
"@parent": ""
|
|
1083
|
+
}
|
|
1084
|
+
],
|
|
1085
|
+
"@element_graphic_neighbours<type>": "function",
|
|
1086
|
+
"@element_graphic_neighbours<desc>": "Binary rule on two neighbouring graphic candidates (line, rect, table fragment, or generic pde_element) during graphic-table merging.",
|
|
1087
|
+
"@element_graphic_neighbours<use>": "Set join true to allow the pair under one parent table hypothesis; false forces them to stay separate when the default heuristic over-merges.",
|
|
1088
|
+
"@element_graphic_neighbours<insert>": [
|
|
1089
|
+
{
|
|
1090
|
+
"query": {
|
|
1091
|
+
"param": [
|
|
1092
|
+
"pde_element",
|
|
1093
|
+
"pde_element"
|
|
1094
|
+
],
|
|
1095
|
+
"$and": []
|
|
1096
|
+
}
|
|
1097
|
+
}
|
|
1098
|
+
],
|
|
1099
|
+
"@element_graphic_neighbours": [
|
|
1100
|
+
{
|
|
1101
|
+
"@disable": false,
|
|
1102
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1103
|
+
"@statement": "$if",
|
|
1104
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1105
|
+
"@query": {
|
|
1106
|
+
"@param<type>": "query_param",
|
|
1107
|
+
"@param<insert>": [
|
|
1108
|
+
"pde_element",
|
|
1109
|
+
"pde_element"
|
|
1110
|
+
],
|
|
1111
|
+
"param": [
|
|
1112
|
+
[
|
|
1113
|
+
"pde_element",
|
|
1114
|
+
"pde_line",
|
|
1115
|
+
"pde_rect",
|
|
1116
|
+
"pde_table"
|
|
1117
|
+
],
|
|
1118
|
+
[
|
|
1119
|
+
"pde_element",
|
|
1120
|
+
"pde_line",
|
|
1121
|
+
"pde_rect",
|
|
1122
|
+
"pde_table"
|
|
1123
|
+
]
|
|
1124
|
+
]
|
|
1125
|
+
},
|
|
1126
|
+
"@join<desc>": "If true, both elements may share the same graphic-table parent; if false, the merge is rejected for this pair.",
|
|
1127
|
+
"@join<type>": "bool",
|
|
1128
|
+
"@join<value>": [
|
|
1129
|
+
"true",
|
|
1130
|
+
"false"
|
|
1131
|
+
],
|
|
1132
|
+
"@join": true
|
|
1133
|
+
}
|
|
1134
|
+
],
|
|
1135
|
+
"@element_graphic_update<type>": "function",
|
|
1136
|
+
"@element_graphic_update<desc>": "Runs on graphic layout nodes (pde_line, pde_rect, pde_table) after graphic detection: shared update keys (flag, parent, tag, single_instance, id, …).",
|
|
1137
|
+
"@element_graphic_update<use>": "Post-process vector tables and borders in one rule set instead of separate line_update and rect_update copies.",
|
|
1138
|
+
"@element_graphic_update<insert>": [
|
|
1139
|
+
{
|
|
1140
|
+
"query": {
|
|
1141
|
+
"param": [
|
|
1142
|
+
"pde_element"
|
|
1143
|
+
],
|
|
1144
|
+
"$and": []
|
|
1145
|
+
}
|
|
1146
|
+
}
|
|
1147
|
+
],
|
|
1148
|
+
"@element_graphic_update": [
|
|
1149
|
+
{
|
|
1150
|
+
"@disable": false,
|
|
1151
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1152
|
+
"@statement": "$if",
|
|
1153
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1154
|
+
"@query": {
|
|
1155
|
+
"@param<type>": "query_param",
|
|
1156
|
+
"@param<insert>": [
|
|
1157
|
+
"pde_element"
|
|
1158
|
+
],
|
|
1159
|
+
"param": [
|
|
1160
|
+
[
|
|
1161
|
+
"pde_element",
|
|
1162
|
+
"pde_line",
|
|
1163
|
+
"pde_rect",
|
|
1164
|
+
"pde_table"
|
|
1165
|
+
]
|
|
1166
|
+
]
|
|
1167
|
+
},
|
|
1168
|
+
"@flag<link>": "schema:values:flag",
|
|
1169
|
+
"@flag": "",
|
|
1170
|
+
"@alt<link>": "schema:values:alt",
|
|
1171
|
+
"@alt": "",
|
|
1172
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1173
|
+
"@actual_text": "",
|
|
1174
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1175
|
+
"@expansion": "",
|
|
1176
|
+
"@lang<link>": "schema:values:lang",
|
|
1177
|
+
"@lang": "",
|
|
1178
|
+
"@label<link>": "schema:values:label",
|
|
1179
|
+
"@label": "",
|
|
1180
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1181
|
+
"@tag": "",
|
|
1182
|
+
"@name<link>": "schema:values:name",
|
|
1183
|
+
"@name": "table1",
|
|
1184
|
+
"@parent<link>": "schema:values:parent",
|
|
1185
|
+
"@parent": "",
|
|
1186
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1187
|
+
"@single_instance": "font_size|font_name|left",
|
|
1188
|
+
"@id<link>": "schema:values:id",
|
|
1189
|
+
"@id": ""
|
|
1190
|
+
}
|
|
1191
|
+
],
|
|
1192
|
+
"@word_update<type>": "function",
|
|
1193
|
+
"@word_update<desc>": "Invoked for each extracted word during layout. (1) While words are collected, matching rules can set flags (header, footer, artifact, no_join, …), language, numbering, labels, and related keys; when header, footer, or artifact is set, the word may be moved into the corresponding band or region. (2) During font preparation the engine may evaluate this hook again so word_space can lock the space width for that font before lines are built. Global word_space_ratio in the main pagemap block still scales those widths afterward; this hook does not replace that multiplier.",
|
|
1194
|
+
"@word_update<use>": "Tag running heads/footers/artifacts, fix list markers, or set word_space on representative tokens so the font’s space width is locked before line grouping. For line-level spacing tweaks after lines exist, use text_line_update.",
|
|
1195
|
+
"@word_update<insert>": [
|
|
1196
|
+
{
|
|
1197
|
+
"query": {
|
|
1198
|
+
"param": [
|
|
1199
|
+
"pde_word"
|
|
1200
|
+
],
|
|
1201
|
+
"$and": []
|
|
1202
|
+
}
|
|
1203
|
+
}
|
|
1204
|
+
],
|
|
1205
|
+
"@word_update": [
|
|
1206
|
+
{
|
|
1207
|
+
"@disable": false,
|
|
1208
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1209
|
+
"@statement": "$if",
|
|
1210
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1211
|
+
"@query": {
|
|
1212
|
+
"@param<type>": "query_param",
|
|
1213
|
+
"@param<insert>": [
|
|
1214
|
+
"pde_word"
|
|
1215
|
+
],
|
|
1216
|
+
"param": [
|
|
1217
|
+
[
|
|
1218
|
+
"pde_word"
|
|
1219
|
+
]
|
|
1220
|
+
]
|
|
1221
|
+
},
|
|
1222
|
+
"@flag<link>": "schema:values:flag",
|
|
1223
|
+
"@flag": "",
|
|
1224
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1225
|
+
"@actual_text": "",
|
|
1226
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1227
|
+
"@expansion": "",
|
|
1228
|
+
"@lang<link>": "schema:values:lang",
|
|
1229
|
+
"@lang": "",
|
|
1230
|
+
"@numbering<link>": "schema:values:numbering",
|
|
1231
|
+
"@numbering": "",
|
|
1232
|
+
"@label<link>": "schema:values:label",
|
|
1233
|
+
"@label": "",
|
|
1234
|
+
"@label_text<link>": "schema:values:label_text",
|
|
1235
|
+
"@label_text": "",
|
|
1236
|
+
"@toc<link>": "schema:values:toc",
|
|
1237
|
+
"@toc": "",
|
|
1238
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1239
|
+
"@tag": "",
|
|
1240
|
+
"@heading<link>": "schema:values:heading",
|
|
1241
|
+
"@heading": "normal",
|
|
1242
|
+
"@word_flag<link>": "schema:values:word_flag",
|
|
1243
|
+
"@word_flag": "",
|
|
1244
|
+
"@name<link>": "schema:values:name",
|
|
1245
|
+
"@name": "word1",
|
|
1246
|
+
"@parent<link>": "schema:values:parent",
|
|
1247
|
+
"@parent": "",
|
|
1248
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1249
|
+
"@single_instance": "font_size|font_name|text",
|
|
1250
|
+
"@word_space<link>": "schema:values:word_space",
|
|
1251
|
+
"@word_space": 0
|
|
1252
|
+
}
|
|
1253
|
+
],
|
|
1254
|
+
"@word_neighbours<type>": "function",
|
|
1255
|
+
"@word_neighbours<desc>": "Binary rule on adjacent pde_word tokens (left then right) while the engine merges words into a text line.",
|
|
1256
|
+
"@word_neighbours<use>": "Force @join true to merge hyphenated or split tokens, or false to keep a hard boundary where heuristics would glue words.",
|
|
1257
|
+
"@word_neighbours<rule>": "The function takes two parameters: the first (pde_word) refers to the left-side word, and the second (pde_word) refers to the right-side word.",
|
|
1258
|
+
"@word_neighbours<insert>": [
|
|
1259
|
+
{
|
|
1260
|
+
"query": {
|
|
1261
|
+
"param": [
|
|
1262
|
+
"pde_word",
|
|
1263
|
+
"pde_word"
|
|
1264
|
+
],
|
|
1265
|
+
"$and": []
|
|
1266
|
+
}
|
|
1267
|
+
}
|
|
1268
|
+
],
|
|
1269
|
+
"@word_neighbours": [
|
|
1270
|
+
{
|
|
1271
|
+
"@disable": false,
|
|
1272
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1273
|
+
"@statement": "$if",
|
|
1274
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1275
|
+
"@query": {
|
|
1276
|
+
"@param<type>": "query_param",
|
|
1277
|
+
"@param<insert>": [
|
|
1278
|
+
"pde_word",
|
|
1279
|
+
"pde_word"
|
|
1280
|
+
],
|
|
1281
|
+
"param": [
|
|
1282
|
+
[
|
|
1283
|
+
"pde_word"
|
|
1284
|
+
],
|
|
1285
|
+
[
|
|
1286
|
+
"pde_word"
|
|
1287
|
+
]
|
|
1288
|
+
]
|
|
1289
|
+
},
|
|
1290
|
+
"@join<desc>": "If true, merge the left and right words into one line segment; if false, keep a word boundary between them.",
|
|
1291
|
+
"@join<type>": "bool",
|
|
1292
|
+
"@join<value>": [
|
|
1293
|
+
"true",
|
|
1294
|
+
"false"
|
|
1295
|
+
],
|
|
1296
|
+
"@join": true
|
|
1297
|
+
}
|
|
1298
|
+
],
|
|
1299
|
+
"@text_line_update<type>": "function",
|
|
1300
|
+
"@text_line_update<desc>": "Runs per pde_text_line after line assembly: same keys as other *_update steps (flag, heading, label, word_space, split, splitter, …).",
|
|
1301
|
+
"@text_line_update<use>": "Fix line-level tags, force split back to words, tune word_space for justified rows, or clear continuous chains via splitter before paragraph/text grouping.",
|
|
1302
|
+
"@text_line_update<insert>": [
|
|
1303
|
+
{
|
|
1304
|
+
"query": {
|
|
1305
|
+
"param": [
|
|
1306
|
+
"pde_text_line"
|
|
1307
|
+
],
|
|
1308
|
+
"$and": []
|
|
1309
|
+
}
|
|
1310
|
+
}
|
|
1311
|
+
],
|
|
1312
|
+
"@text_line_update": [
|
|
1313
|
+
{
|
|
1314
|
+
"@disable": false,
|
|
1315
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1316
|
+
"@statement": "$if",
|
|
1317
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1318
|
+
"@query": {
|
|
1319
|
+
"@param<type>": "query_param",
|
|
1320
|
+
"@param<insert>": [
|
|
1321
|
+
"pde_text_line"
|
|
1322
|
+
],
|
|
1323
|
+
"param": [
|
|
1324
|
+
[
|
|
1325
|
+
"pde_text_line"
|
|
1326
|
+
]
|
|
1327
|
+
]
|
|
1328
|
+
},
|
|
1329
|
+
"@flag<link>": "schema:values:flag",
|
|
1330
|
+
"@flag": "",
|
|
1331
|
+
"@numbering<link>": "schema:values:numbering",
|
|
1332
|
+
"@numbering": "",
|
|
1333
|
+
"@label<link>": "schema:values:label",
|
|
1334
|
+
"@label": "",
|
|
1335
|
+
"@label_text<link>": "schema:values:label_text",
|
|
1336
|
+
"@label_text": "",
|
|
1337
|
+
"@toc<link>": "schema:values:toc",
|
|
1338
|
+
"@toc": "",
|
|
1339
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1340
|
+
"@tag": "",
|
|
1341
|
+
"@heading<link>": "schema:values:heading",
|
|
1342
|
+
"@heading": "normal",
|
|
1343
|
+
"@text_line_flag<link>": "schema:values:text_line_flag",
|
|
1344
|
+
"@text_line_flag": "",
|
|
1345
|
+
"@name<link>": "schema:values:name",
|
|
1346
|
+
"@name": "textline1",
|
|
1347
|
+
"@parent<link>": "schema:values:parent",
|
|
1348
|
+
"@parent": "",
|
|
1349
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1350
|
+
"@single_instance": "font_size|font_name|text",
|
|
1351
|
+
"@word_space<link>": "schema:values:word_space",
|
|
1352
|
+
"@word_space": 0,
|
|
1353
|
+
"@split<desc>": "If true, split this line back to individual words.",
|
|
1354
|
+
"@split<type>": "bool",
|
|
1355
|
+
"@split<value>": [
|
|
1356
|
+
"true",
|
|
1357
|
+
"false"
|
|
1358
|
+
],
|
|
1359
|
+
"@split_filling<desc>": "If true, do not remove filling words from line and do not mark a line as filling.",
|
|
1360
|
+
"@split_filling<type>": "bool",
|
|
1361
|
+
"@split_filling<value>": [
|
|
1362
|
+
"true",
|
|
1363
|
+
"false"
|
|
1364
|
+
],
|
|
1365
|
+
"@split": true,
|
|
1366
|
+
"@splitter": "pde_table",
|
|
1367
|
+
"@splitter<link>": "schema:values:type",
|
|
1368
|
+
"@splitter<desc>": "Used together with flag \"splitter\": element types (pde_table, pde_list, pde_text, …) whose active cross-page continuous chains are cleared at this element. Multiple types may be separated with |."
|
|
1369
|
+
}
|
|
1370
|
+
],
|
|
1371
|
+
"@text_line_neighbours<type>": "function",
|
|
1372
|
+
"@text_line_neighbours<desc>": "Binary rule on an upper and lower pde_text_line: decide whether they may belong to one paragraph or must split.",
|
|
1373
|
+
"@text_line_neighbours<use>": "Override default paragraph glue for edge cases (e.g. headings glued to body, or two-column lines merged). First query param is the line above, second is below.",
|
|
1374
|
+
"@text_line_neighbours<rule>": "The first parameter, (pde_text_line), is above, and the second parameter, (pde_text_line), is below, with the direction moving from top to bottom.",
|
|
1375
|
+
"@text_line_neighbours<insert>": [
|
|
1376
|
+
{
|
|
1377
|
+
"query": {
|
|
1378
|
+
"param": [
|
|
1379
|
+
"pde_text_line",
|
|
1380
|
+
"pde_text_line"
|
|
1381
|
+
],
|
|
1382
|
+
"$and": []
|
|
1383
|
+
}
|
|
1384
|
+
}
|
|
1385
|
+
],
|
|
1386
|
+
"@text_line_neighbours": [
|
|
1387
|
+
{
|
|
1388
|
+
"@disable": false,
|
|
1389
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1390
|
+
"@statement": "$if",
|
|
1391
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1392
|
+
"@query": {
|
|
1393
|
+
"@param<type>": "query_param",
|
|
1394
|
+
"@param<insert>": [
|
|
1395
|
+
"pde_text_line",
|
|
1396
|
+
"pde_text_line"
|
|
1397
|
+
],
|
|
1398
|
+
"param": [
|
|
1399
|
+
[
|
|
1400
|
+
"pde_text_line"
|
|
1401
|
+
],
|
|
1402
|
+
[
|
|
1403
|
+
"pde_text_line"
|
|
1404
|
+
]
|
|
1405
|
+
]
|
|
1406
|
+
},
|
|
1407
|
+
"@join<desc>": "If true, the upper and lower lines may belong to one paragraph; if false, force a paragraph break between them. If omitted, default paragraph heuristics apply.",
|
|
1408
|
+
"@join<type>": "bool",
|
|
1409
|
+
"@join<value>": [
|
|
1410
|
+
"true",
|
|
1411
|
+
"false"
|
|
1412
|
+
],
|
|
1413
|
+
"@join": true,
|
|
1414
|
+
"@split<desc>": "If split is set to true, the text is divided between these lines.",
|
|
1415
|
+
"@split<type>": "bool",
|
|
1416
|
+
"@split<value>": [
|
|
1417
|
+
"true",
|
|
1418
|
+
"false"
|
|
1419
|
+
],
|
|
1420
|
+
"@split": true
|
|
1421
|
+
}
|
|
1422
|
+
],
|
|
1423
|
+
"@text_update<type>": "function",
|
|
1424
|
+
"@text_update<desc>": "Runs per pde_text after paragraph/chunk grouping: set tag, heading, lang, flags, numbering, label fields, id, and other text-level template keys.",
|
|
1425
|
+
"@text_update<use>": "Refine accessibility and structure on whole text blocks (not single words or lines); you may set the same template keys documented for element_update (tag, parent, flag, dimensions, contents, and related fields).",
|
|
1426
|
+
"@text_update<insert>": [
|
|
1427
|
+
{
|
|
1428
|
+
"query": {
|
|
1429
|
+
"param": [
|
|
1430
|
+
"pde_text"
|
|
1431
|
+
],
|
|
1432
|
+
"$and": []
|
|
1433
|
+
}
|
|
1434
|
+
}
|
|
1435
|
+
],
|
|
1436
|
+
"@text_update": [
|
|
1437
|
+
{
|
|
1438
|
+
"@disable": false,
|
|
1439
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1440
|
+
"@statement": "$if",
|
|
1441
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1442
|
+
"@query": {
|
|
1443
|
+
"@param<type>": "query_param",
|
|
1444
|
+
"@param<insert>": [
|
|
1445
|
+
"pde_text"
|
|
1446
|
+
],
|
|
1447
|
+
"param": [
|
|
1448
|
+
[
|
|
1449
|
+
"pde_text"
|
|
1450
|
+
]
|
|
1451
|
+
]
|
|
1452
|
+
},
|
|
1453
|
+
"@flag<link>": "schema:values:flag",
|
|
1454
|
+
"@flag": "",
|
|
1455
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1456
|
+
"@actual_text": "",
|
|
1457
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1458
|
+
"@expansion": "",
|
|
1459
|
+
"@lang<link>": "schema:values:lang",
|
|
1460
|
+
"@lang": "",
|
|
1461
|
+
"@numbering<link>": "schema:values:numbering",
|
|
1462
|
+
"@numbering": "",
|
|
1463
|
+
"@label<link>": "schema:values:label",
|
|
1464
|
+
"@label": "",
|
|
1465
|
+
"@label_text<link>": "schema:values:label_text",
|
|
1466
|
+
"@label_text": "",
|
|
1467
|
+
"@toc<link>": "schema:values:toc",
|
|
1468
|
+
"@toc": "",
|
|
1469
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1470
|
+
"@tag": "",
|
|
1471
|
+
"@heading<link>": "schema:values:heading",
|
|
1472
|
+
"@heading": "normal",
|
|
1473
|
+
"@text_flag<link>": "schema:values:text_flag",
|
|
1474
|
+
"@text_flag": "",
|
|
1475
|
+
"@name<link>": "schema:values:name",
|
|
1476
|
+
"@name": "text1",
|
|
1477
|
+
"@parent<link>": "schema:values:parent",
|
|
1478
|
+
"@parent": "",
|
|
1479
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1480
|
+
"@single_instance": "font_size|font_name|text",
|
|
1481
|
+
"@id<link>": "schema:values:id",
|
|
1482
|
+
"@id": ""
|
|
1483
|
+
}
|
|
1484
|
+
],
|
|
1485
|
+
"@image_update<type>": "function",
|
|
1486
|
+
"@image_update<desc>": "Runs per pde_image after raster/vector image objects are recognized: alt, actual_text, expansion, lang, tag, flag, single_instance, id, etc.",
|
|
1487
|
+
"@image_update<use>": "Assign figure semantics, deduplicate decorative icons via single_instance, or mark images artifact without re-running detection.",
|
|
1488
|
+
"@image_update<insert>": [
|
|
1489
|
+
{
|
|
1490
|
+
"query": {
|
|
1491
|
+
"param": [
|
|
1492
|
+
"pde_image"
|
|
1493
|
+
],
|
|
1494
|
+
"$and": []
|
|
1495
|
+
}
|
|
1496
|
+
}
|
|
1497
|
+
],
|
|
1498
|
+
"@image_update": [
|
|
1499
|
+
{
|
|
1500
|
+
"@disable": false,
|
|
1501
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1502
|
+
"@statement": "$if",
|
|
1503
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1504
|
+
"@query": {
|
|
1505
|
+
"@param<type>": "query_param",
|
|
1506
|
+
"@param<insert>": [
|
|
1507
|
+
"pde_image"
|
|
1508
|
+
],
|
|
1509
|
+
"param": [
|
|
1510
|
+
[
|
|
1511
|
+
"pde_image"
|
|
1512
|
+
]
|
|
1513
|
+
]
|
|
1514
|
+
},
|
|
1515
|
+
"@alt<link>": "schema:values:alt",
|
|
1516
|
+
"@alt": "",
|
|
1517
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1518
|
+
"@actual_text": "",
|
|
1519
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1520
|
+
"@expansion": "",
|
|
1521
|
+
"@lang<link>": "schema:values:lang",
|
|
1522
|
+
"@lang": "",
|
|
1523
|
+
"@numbering<link>": "schema:values:numbering",
|
|
1524
|
+
"@numbering": "",
|
|
1525
|
+
"@label<link>": "schema:values:label",
|
|
1526
|
+
"@label": "",
|
|
1527
|
+
"@label_text<link>": "schema:values:label_text",
|
|
1528
|
+
"@label_text": "",
|
|
1529
|
+
"@toc<link>": "schema:values:toc",
|
|
1530
|
+
"@toc": "",
|
|
1531
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1532
|
+
"@tag": "",
|
|
1533
|
+
"@flag<link>": "schema:values:flag",
|
|
1534
|
+
"@flag": "",
|
|
1535
|
+
"@name<link>": "schema:values:name",
|
|
1536
|
+
"@name": "element1",
|
|
1537
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1538
|
+
"@single_instance": "type|width|height",
|
|
1539
|
+
"@id<link>": "schema:values:id",
|
|
1540
|
+
"@id": ""
|
|
1541
|
+
}
|
|
1542
|
+
],
|
|
1543
|
+
"@element_update<type>": "function",
|
|
1544
|
+
"@element_update<desc>": "Generic template hook for any element type listed in the query (text, image, line, rect, table, and other layout types). Use it to set tag, parent, flag, dimensions, contents, and related keys on the matched element.",
|
|
1545
|
+
"@element_update<use>": "One place to post-process mixed elements; only keys you set are applied—other values stay as detected by the layout engine. See the knowledge-base topics for element_update and applying rules.",
|
|
1546
|
+
"@element_update<insert>": [
|
|
1547
|
+
{
|
|
1548
|
+
"query": {
|
|
1549
|
+
"param": [
|
|
1550
|
+
"pde_element"
|
|
1551
|
+
],
|
|
1552
|
+
"$and": []
|
|
1553
|
+
}
|
|
1554
|
+
}
|
|
1555
|
+
],
|
|
1556
|
+
"@element_update": [
|
|
1557
|
+
{
|
|
1558
|
+
"@disable": false,
|
|
1559
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1560
|
+
"@statement": "$if",
|
|
1561
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1562
|
+
"@query": {
|
|
1563
|
+
"@param<type>": "query_param",
|
|
1564
|
+
"@param<insert>": [
|
|
1565
|
+
"pde_element"
|
|
1566
|
+
],
|
|
1567
|
+
"param": [
|
|
1568
|
+
[
|
|
1569
|
+
"pde_element",
|
|
1570
|
+
"pde_text",
|
|
1571
|
+
"pde_image",
|
|
1572
|
+
"pde_line",
|
|
1573
|
+
"pde_rect",
|
|
1574
|
+
"pde_table"
|
|
1575
|
+
]
|
|
1576
|
+
]
|
|
1577
|
+
},
|
|
1578
|
+
"@alt<link>": "schema:values:alt",
|
|
1579
|
+
"@alt": "",
|
|
1580
|
+
"@contents<link>": "schema:values:contents",
|
|
1581
|
+
"@contents": "",
|
|
1582
|
+
"@actual_text<link>": "schema:values:actual_text",
|
|
1583
|
+
"@actual_text": "",
|
|
1584
|
+
"@expansion<link>": "schema:values:expansion",
|
|
1585
|
+
"@expansion": "",
|
|
1586
|
+
"@lang<link>": "schema:values:lang",
|
|
1587
|
+
"@lang": "",
|
|
1588
|
+
"@flag<link>": "schema:values:flag",
|
|
1589
|
+
"@flag": "",
|
|
1590
|
+
"@tag<link>": "schema:values:tag_type",
|
|
1591
|
+
"@tag": "",
|
|
1592
|
+
"@name<link>": "schema:values:name",
|
|
1593
|
+
"@name": "element1",
|
|
1594
|
+
"@parent<link>": "schema:values:parent",
|
|
1595
|
+
"@parent": "",
|
|
1596
|
+
"@caption<link>": "schema:values:caption",
|
|
1597
|
+
"@caption": "",
|
|
1598
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1599
|
+
"@single_instance": "font_size|font_name|left",
|
|
1600
|
+
"@id<link>": "schema:values:id",
|
|
1601
|
+
"@id": ""
|
|
1602
|
+
}
|
|
1603
|
+
],
|
|
1604
|
+
"@element_nest<type>": "function",
|
|
1605
|
+
"@element_nest<desc>": "Binary template: the query matches two elements by structural kind. The first `param` operand is the candidate parent; the second is the candidate child. When a row matches, the optional `join` value can force nesting (first becomes parent of the second) or forbid it; if `join` is omitted after a match, default nesting logic still runs.",
|
|
1606
|
+
"@element_nest<use>": "Tighten each `param` slot to the `pde_*` kinds that should form a pair (see allowed lists under `query.param`). Set `join` true when that pair should nest even if default rules hesitate; false when they must stay separate despite overlap. Add separate rows for different parent/child combinations.",
|
|
1607
|
+
"@element_nest<insert>": [
|
|
1608
|
+
{
|
|
1609
|
+
"query": {
|
|
1610
|
+
"param": [
|
|
1611
|
+
"pde_element",
|
|
1612
|
+
"pde_element"
|
|
1613
|
+
],
|
|
1614
|
+
"$and": []
|
|
1615
|
+
}
|
|
1616
|
+
}
|
|
1617
|
+
],
|
|
1618
|
+
"@element_nest": [
|
|
1619
|
+
{
|
|
1620
|
+
"@disable": false,
|
|
1621
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1622
|
+
"@statement": "$if",
|
|
1623
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1624
|
+
"@query": {
|
|
1625
|
+
"@param<type>": "query_param",
|
|
1626
|
+
"@param<insert>": [
|
|
1627
|
+
"pde_element",
|
|
1628
|
+
"pde_element"
|
|
1629
|
+
],
|
|
1630
|
+
"param": [
|
|
1631
|
+
[
|
|
1632
|
+
"pde_element",
|
|
1633
|
+
"pde_image",
|
|
1634
|
+
"pde_table",
|
|
1635
|
+
"pde_container",
|
|
1636
|
+
"pde_list",
|
|
1637
|
+
"pde_rect"
|
|
1638
|
+
],
|
|
1639
|
+
[
|
|
1640
|
+
"pde_element",
|
|
1641
|
+
"pde_text",
|
|
1642
|
+
"pde_image",
|
|
1643
|
+
"pde_table",
|
|
1644
|
+
"pde_container",
|
|
1645
|
+
"pde_list",
|
|
1646
|
+
"pde_line",
|
|
1647
|
+
"pde_rect",
|
|
1648
|
+
"pde_form_field",
|
|
1649
|
+
"pde_annot"
|
|
1650
|
+
]
|
|
1651
|
+
]
|
|
1652
|
+
},
|
|
1653
|
+
"@join<desc>": "When true, make first element parent of another.",
|
|
1654
|
+
"@join<type>": "bool",
|
|
1655
|
+
"@join<value>": [
|
|
1656
|
+
"true",
|
|
1657
|
+
"false"
|
|
1658
|
+
],
|
|
1659
|
+
"@join": false
|
|
1660
|
+
}
|
|
1661
|
+
],
|
|
1662
|
+
"@table_update<type>": "function",
|
|
1663
|
+
"@table_update<desc>": "Runs while the engine is building each detected table, after the grid is recognized. Matching rules here load values for the table_update step and apply them to the current table. The boolean fields below, when set in this block, override the document’s default pagemap table thresholds for that table only: delete_empty, split_by_row, split_label, split_by_cell, span_empty, span_row_first, span_row, join, cell_header, and the optional flag. Note: split_by_cell usually behaves as intended only when text-based row detection (text_table_detect_row) is also enabled in pagemap, unless your template explicitly changes that relationship.",
|
|
1664
|
+
"@table_update<use>": "Per-table post-process: e.g. split_by_row + split_label to rebuild cells, turn off delete_empty to retain blank grid lines for tagging, or set flag for tag_table. For individual cells after structure exists, use cell_update.",
|
|
1665
|
+
"@table_update<insert>": [
|
|
1666
|
+
{
|
|
1667
|
+
"query": {
|
|
1668
|
+
"param": [
|
|
1669
|
+
"pde_table"
|
|
1670
|
+
],
|
|
1671
|
+
"$and": []
|
|
1672
|
+
}
|
|
1673
|
+
}
|
|
1674
|
+
],
|
|
1675
|
+
"@table_update": [
|
|
1676
|
+
{
|
|
1677
|
+
"@disable": false,
|
|
1678
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1679
|
+
"@statement": "$if",
|
|
1680
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1681
|
+
"@query": {
|
|
1682
|
+
"@param<type>": "query_param",
|
|
1683
|
+
"@param<insert>": [
|
|
1684
|
+
"pde_table"
|
|
1685
|
+
],
|
|
1686
|
+
"param": [
|
|
1687
|
+
[
|
|
1688
|
+
"pde_table"
|
|
1689
|
+
]
|
|
1690
|
+
]
|
|
1691
|
+
},
|
|
1692
|
+
"@delete_empty<desc>": "When true, drops empty rows/columns after recognition for this table when the rule matches (overrides pagemap table_update_delete_empty).",
|
|
1693
|
+
"@delete_empty<type>": "bool",
|
|
1694
|
+
"@delete_empty<value>": [
|
|
1695
|
+
"true",
|
|
1696
|
+
"false"
|
|
1697
|
+
],
|
|
1698
|
+
"@delete_empty": true,
|
|
1699
|
+
"@split_by_cell<desc>": "When the rule matches, replaces the engine defaults for split-by-cell and for text-based row detection (the table_update_split_by_cell and text_table_detect_row pagemap fields). If true, the engine splits merged cells that were wrongly treated as one.",
|
|
1700
|
+
"@split_by_cell<type>": "bool",
|
|
1701
|
+
"@split_by_cell<value>": [
|
|
1702
|
+
"true",
|
|
1703
|
+
"false"
|
|
1704
|
+
],
|
|
1705
|
+
"@split_by_cell": false,
|
|
1706
|
+
"@split_by_row<desc>": "When true, breaks multi-line cell text to one line per row for this table (overrides pagemap default).",
|
|
1707
|
+
"@split_by_row<type>": "bool",
|
|
1708
|
+
"@split_by_row<value>": [
|
|
1709
|
+
"true",
|
|
1710
|
+
"false"
|
|
1711
|
+
],
|
|
1712
|
+
"@split_by_row": false,
|
|
1713
|
+
"@split_label<desc>": "When true, separates list/label markers from body text inside cells where they were fused (overrides pagemap default).",
|
|
1714
|
+
"@split_label<type>": "bool",
|
|
1715
|
+
"@split_label<value>": [
|
|
1716
|
+
"true",
|
|
1717
|
+
"false"
|
|
1718
|
+
],
|
|
1719
|
+
"@split_label": false,
|
|
1720
|
+
"@span_empty<desc>": "When true, merges visually empty cells using colspan/rowspan metadata (overrides pagemap default).",
|
|
1721
|
+
"@span_empty<type>": "bool",
|
|
1722
|
+
"@span_empty<value>": [
|
|
1723
|
+
"true",
|
|
1724
|
+
"false"
|
|
1725
|
+
],
|
|
1726
|
+
"@span_empty": true,
|
|
1727
|
+
"@span_row_first<desc>": "When true, row-span repair runs before column-span repair; when false, column-first (overrides pagemap default).",
|
|
1728
|
+
"@span_row_first<type>": "bool",
|
|
1729
|
+
"@span_row_first<value>": [
|
|
1730
|
+
"true",
|
|
1731
|
+
"false"
|
|
1732
|
+
],
|
|
1733
|
+
"@span_row_first": false,
|
|
1734
|
+
"@span_row<desc>": "When the rule matches, overrides pagemap table_update_span_row. If true, the engine merges table rows using row-span metadata.",
|
|
1735
|
+
"@span_row<type>": "bool",
|
|
1736
|
+
"@span_row<value>": [
|
|
1737
|
+
"true",
|
|
1738
|
+
"false"
|
|
1739
|
+
],
|
|
1740
|
+
"@span_row": false,
|
|
1741
|
+
"@join<desc>": "When true, concatenates all text fragments in each cell into one paragraph element (overrides pagemap default).",
|
|
1742
|
+
"@join<type>": "bool",
|
|
1743
|
+
"@join<value>": [
|
|
1744
|
+
"true",
|
|
1745
|
+
"false"
|
|
1746
|
+
],
|
|
1747
|
+
"@join": false,
|
|
1748
|
+
"@cell_header<link>": "schema:values:cell_header",
|
|
1749
|
+
"@cell_header": true,
|
|
1750
|
+
"@flag<link>": "schema:values:flag",
|
|
1751
|
+
"@flag": ""
|
|
1752
|
+
}
|
|
1753
|
+
],
|
|
1754
|
+
"@cell_update<type>": "function",
|
|
1755
|
+
"@cell_update<desc>": "Runs for each table cell after the table’s global passes, when the cell is a span and not only an initial placeholder. Matching rules can set name (anchor), language, flag, column/row span, scope, cell_header, id, heading, single_instance, associated_header, join (merge text in that cell), and join_row (steer row merges with borders in later passes).",
|
|
1756
|
+
"@cell_update<use>": "Fine-tune one cell’s metadata or force joining text in that cell; use join_row -1, 0, or 1 to steer row-merging when borders are present.",
|
|
1757
|
+
"@cell_update<insert>": [
|
|
1758
|
+
{
|
|
1759
|
+
"query": {
|
|
1760
|
+
"param": [
|
|
1761
|
+
"pde_table",
|
|
1762
|
+
"pde_cell"
|
|
1763
|
+
],
|
|
1764
|
+
"$and": []
|
|
1765
|
+
}
|
|
1766
|
+
}
|
|
1767
|
+
],
|
|
1768
|
+
"@cell_update": [
|
|
1769
|
+
{
|
|
1770
|
+
"@disable": false,
|
|
1771
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1772
|
+
"@statement": "$if",
|
|
1773
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1774
|
+
"@query": {
|
|
1775
|
+
"@param<type>": "query_param",
|
|
1776
|
+
"@param<insert>": [
|
|
1777
|
+
"pde_table",
|
|
1778
|
+
"pde_cell"
|
|
1779
|
+
],
|
|
1780
|
+
"param": [
|
|
1781
|
+
[
|
|
1782
|
+
"pde_table"
|
|
1783
|
+
],
|
|
1784
|
+
[
|
|
1785
|
+
"pde_cell"
|
|
1786
|
+
]
|
|
1787
|
+
]
|
|
1788
|
+
},
|
|
1789
|
+
"@cell_row_span": "",
|
|
1790
|
+
"@cell_row_span<link>": "schema:values:cell_row_span",
|
|
1791
|
+
"@cell_column_span": "",
|
|
1792
|
+
"@cell_column_span<link>": "schema:values:cell_column_span",
|
|
1793
|
+
"@cell_scope": "column",
|
|
1794
|
+
"@cell_scope<link>": "schema:values:cell_scope",
|
|
1795
|
+
"@join<desc>": "When true for this cell, runs join texts on that cell so all child texts become one paragraph.",
|
|
1796
|
+
"@join<type>": "bool",
|
|
1797
|
+
"@join<value>": [
|
|
1798
|
+
"true",
|
|
1799
|
+
"false"
|
|
1800
|
+
],
|
|
1801
|
+
"@join": false,
|
|
1802
|
+
"@cell_header<link>": "schema:values:cell_header",
|
|
1803
|
+
"@cell_header": true,
|
|
1804
|
+
"@heading<link>": "schema:values:heading",
|
|
1805
|
+
"@heading": "normal",
|
|
1806
|
+
"@id<link>": "schema:values:id",
|
|
1807
|
+
"@id": "",
|
|
1808
|
+
"@name<link>": "schema:values:name",
|
|
1809
|
+
"@name": "cell1",
|
|
1810
|
+
"@cell_associated_header<link>": "schema:values:associated_header",
|
|
1811
|
+
"@cell_associated_header": "",
|
|
1812
|
+
"@single_instance<link>": "schema:values:single_instance",
|
|
1813
|
+
"@single_instance": "font_size|font_name|text|cell_column",
|
|
1814
|
+
"@join_row<desc>": "-1 favors merging this row with the previous, +1 with behavior tied to the following row, 0 uses border-aware heuristics during the cell pass.",
|
|
1815
|
+
"@join_row<type>": "int",
|
|
1816
|
+
"@join_row": 0
|
|
1817
|
+
}
|
|
1818
|
+
],
|
|
1819
|
+
"@table_split<type>": "function",
|
|
1820
|
+
"@table_split<desc>": "Binary template on pde_table plus one child element: the engine scans rows from the bottom upward; if the rule matches and row is true, that row index becomes a table split boundary. Children with a table splitter flag can also force a split row.",
|
|
1821
|
+
"@table_split<use>": "Declare which content row should break a continuous or oversized table; also aligns with the table-splitter flag on cells when pde_table appears in the splitter type list.",
|
|
1822
|
+
"@table_split<insert>": [
|
|
1823
|
+
{
|
|
1824
|
+
"query": {
|
|
1825
|
+
"param": [
|
|
1826
|
+
"pde_table",
|
|
1827
|
+
"pde_element"
|
|
1828
|
+
],
|
|
1829
|
+
"$and": []
|
|
1830
|
+
}
|
|
1831
|
+
}
|
|
1832
|
+
],
|
|
1833
|
+
"@table_split": [
|
|
1834
|
+
{
|
|
1835
|
+
"@disable": false,
|
|
1836
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1837
|
+
"@statement": "$if",
|
|
1838
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1839
|
+
"@query": {
|
|
1840
|
+
"@param<type>": "query_param",
|
|
1841
|
+
"@param<insert>": [
|
|
1842
|
+
"pde_table",
|
|
1843
|
+
"pde_element"
|
|
1844
|
+
],
|
|
1845
|
+
"param": [
|
|
1846
|
+
[
|
|
1847
|
+
"pde_table"
|
|
1848
|
+
],
|
|
1849
|
+
[
|
|
1850
|
+
"pde_element",
|
|
1851
|
+
"pde_text",
|
|
1852
|
+
"pde_image",
|
|
1853
|
+
"pde_container",
|
|
1854
|
+
"pde_list",
|
|
1855
|
+
"pde_line",
|
|
1856
|
+
"pde_rect",
|
|
1857
|
+
"pde_form_field",
|
|
1858
|
+
"pde_annot"
|
|
1859
|
+
]
|
|
1860
|
+
]
|
|
1861
|
+
},
|
|
1862
|
+
"@row<desc>": "Must be true for a matching rule to return the current row index as the split point during the bottom-up scan.",
|
|
1863
|
+
"@row<type>": "bool",
|
|
1864
|
+
"@row<default>": false,
|
|
1865
|
+
"@row<value>": [
|
|
1866
|
+
"true",
|
|
1867
|
+
"false"
|
|
1868
|
+
],
|
|
1869
|
+
"@row": false
|
|
1870
|
+
}
|
|
1871
|
+
],
|
|
1872
|
+
"@element_prop_update<type>": "function",
|
|
1873
|
+
"@element_prop_update<desc>": "Updates a specific property (Alt, ActualText, Expansion, or Contents) of an element. The value is determined by a step-by-step fallback process. The engine checks each step in order (1 to 7); if a step is enabled (true) and yields a value, that value is used, and subsequent steps are skipped.",
|
|
1874
|
+
"@element_prop_update<use>": "Use this function to ensure PDF/UA accessibility compliance by automatically detecting or assigning missing descriptions. For example, you can automatically pull Alt text for an image from a nearby caption, or set the ActualText for a link based on its URL.",
|
|
1875
|
+
"@element_prop_update<rule>": "If the property was already set explicitly in the template (e.g. during element creation), this update function is ignored to prevent overwriting custom values.",
|
|
1876
|
+
"@element_prop_update<insert>": [
|
|
1877
|
+
{
|
|
1878
|
+
"query": {
|
|
1879
|
+
"param": [
|
|
1880
|
+
"pde_element"
|
|
1881
|
+
],
|
|
1882
|
+
"$and": []
|
|
1883
|
+
},
|
|
1884
|
+
"property": "alt",
|
|
1885
|
+
"mc": "true",
|
|
1886
|
+
"children": "true",
|
|
1887
|
+
"prev": "false",
|
|
1888
|
+
"next": "false",
|
|
1889
|
+
"bbox_padding": "true",
|
|
1890
|
+
"destination": "true"
|
|
1891
|
+
}
|
|
1892
|
+
],
|
|
1893
|
+
"@element_prop_update": [
|
|
1894
|
+
{
|
|
1895
|
+
"@disable": false,
|
|
1896
|
+
"@disable<link>": "schema:keywords:disable",
|
|
1897
|
+
"@statement": "$if",
|
|
1898
|
+
"@statement<link>": "schema:keywords:statement",
|
|
1899
|
+
"@query": {
|
|
1900
|
+
"@param<type>": "query_param",
|
|
1901
|
+
"@param<insert>": [
|
|
1902
|
+
"pde_element"
|
|
1903
|
+
],
|
|
1904
|
+
"param": [
|
|
1905
|
+
[
|
|
1906
|
+
"pde_element",
|
|
1907
|
+
"pde_text",
|
|
1908
|
+
"pde_image",
|
|
1909
|
+
"pde_container",
|
|
1910
|
+
"pde_list",
|
|
1911
|
+
"pde_line",
|
|
1912
|
+
"pde_rect",
|
|
1913
|
+
"pde_table",
|
|
1914
|
+
"pde_toc",
|
|
1915
|
+
"pde_form_field",
|
|
1916
|
+
"pde_annot"
|
|
1917
|
+
]
|
|
1918
|
+
]
|
|
1919
|
+
},
|
|
1920
|
+
"@property<desc>": "The target property to update. 'contents' applies to Annotations/Form Fields; 'alt', 'actual_text', and 'expansion' apply to structural elements.",
|
|
1921
|
+
"@property<type>": "string",
|
|
1922
|
+
"@property<value>": [
|
|
1923
|
+
"alt",
|
|
1924
|
+
"actual_text",
|
|
1925
|
+
"expansion",
|
|
1926
|
+
"contents"
|
|
1927
|
+
],
|
|
1928
|
+
"@property": "alt",
|
|
1929
|
+
"@mc<desc>": "Step 1: Check Marked Content (MC). If true, tries to retrieve the property value directly from the PDF's logical structure tags (MCID).",
|
|
1930
|
+
"@mc<type>": "bool",
|
|
1931
|
+
"@mc<value>": [
|
|
1932
|
+
"true",
|
|
1933
|
+
"false"
|
|
1934
|
+
],
|
|
1935
|
+
"@mc": true,
|
|
1936
|
+
"@children<desc>": "Step 2: Check Children. If true, combines the text content of all child elements (e.g., text runs inside a Figure) to form the property value.",
|
|
1937
|
+
"@children<type>": "bool",
|
|
1938
|
+
"@children<value>": [
|
|
1939
|
+
"true",
|
|
1940
|
+
"false"
|
|
1941
|
+
],
|
|
1942
|
+
"@children": true,
|
|
1943
|
+
"@prev<desc>": "Step 3: Check Previous Sibling. If true, takes the text from the element immediately preceding this one in the structure tree (e.g., a caption above an image).",
|
|
1944
|
+
"@prev<type>": "bool",
|
|
1945
|
+
"@prev<value>": [
|
|
1946
|
+
"true",
|
|
1947
|
+
"false"
|
|
1948
|
+
],
|
|
1949
|
+
"@prev": false,
|
|
1950
|
+
"@next<desc>": "Step 4: Check Next Sibling. If true, takes the text from the element immediately following this one in the structure tree (e.g., a caption below an image).",
|
|
1951
|
+
"@next<type>": "bool",
|
|
1952
|
+
"@next<value>": [
|
|
1953
|
+
"true",
|
|
1954
|
+
"false"
|
|
1955
|
+
],
|
|
1956
|
+
"@next": false,
|
|
1957
|
+
"@bbox_padding<desc>": "Step 5: Check Nearby Text (Spatial). If true, searches for text on the page within a specific distance (padding) from this element's bounding box.",
|
|
1958
|
+
"@bbox_padding<type>": "bool",
|
|
1959
|
+
"@bbox_padding<value>": [
|
|
1960
|
+
"true",
|
|
1961
|
+
"false"
|
|
1962
|
+
],
|
|
1963
|
+
"@bbox_padding": true,
|
|
1964
|
+
"@bbox_padding_left<desc>": "Defines the search range (in points) to the LEFT of the element when Step 5 is active.",
|
|
1965
|
+
"@bbox_padding_left<type>": "float",
|
|
1966
|
+
"@bbox_padding_left": 0,
|
|
1967
|
+
"@bbox_padding_right<desc>": "Defines the search range (in points) to the RIGHT of the element when Step 5 is active.",
|
|
1968
|
+
"@bbox_padding_right<type>": "float",
|
|
1969
|
+
"@bbox_padding_right": 0,
|
|
1970
|
+
"@bbox_padding_bottom<desc>": "Defines the search range (in points) BELOW the element when Step 5 is active. Useful for finding captions.",
|
|
1971
|
+
"@bbox_padding_bottom<type>": "float",
|
|
1972
|
+
"@bbox_padding_bottom": 80,
|
|
1973
|
+
"@bbox_padding_top<desc>": "Defines the search range (in points) ABOVE the element when Step 5 is active. Useful for finding headers or titles.",
|
|
1974
|
+
"@bbox_padding_top<type>": "float",
|
|
1975
|
+
"@bbox_padding_top": 0,
|
|
1976
|
+
"@destination<desc>": "Step 6: Check Link Destination. If true and the element is a Link Annotation, the property value is derived from the link's target URL or destination.",
|
|
1977
|
+
"@destination<type>": "bool",
|
|
1978
|
+
"@destination<value>": [
|
|
1979
|
+
"true",
|
|
1980
|
+
"false"
|
|
1981
|
+
],
|
|
1982
|
+
"@destination": true,
|
|
1983
|
+
"@text<desc>": "Step 7: Hardcoded Text. If no value was found in Steps 1-6, this specific text string is applied as the final fallback.",
|
|
1984
|
+
"@text<use>": "To enforce a specific static text (e.g. 'Company Logo') regardless of other content, you must set this value AND set all previous steps (mc, children, prev, next, bbox_padding, destination) to false.",
|
|
1985
|
+
"@text<type>": "string",
|
|
1986
|
+
"@text": ""
|
|
1987
|
+
}
|
|
1988
|
+
],
|
|
1989
|
+
"@artifact_update<type>": "function",
|
|
1990
|
+
"@artifact_update<desc>": "Runs while post-processing the artifact candidate list: the template can keep an item as artifact (artifact true) or clear the artifact flag so the element moves back to header, footer, page body, or its initial-element parent.",
|
|
1991
|
+
"@artifact_update<use>": "Recover real content misclassified as decorative, or force keep-as-artifact when heuristics would promote an image back into the layout.",
|
|
1992
|
+
"@artifact_update<insert>": [
|
|
1993
|
+
{
|
|
1994
|
+
"query": {
|
|
1995
|
+
"param": [
|
|
1996
|
+
"pde_element"
|
|
1997
|
+
],
|
|
1998
|
+
"$and": []
|
|
1999
|
+
}
|
|
2000
|
+
}
|
|
2001
|
+
],
|
|
2002
|
+
"@artifact_update": [
|
|
2003
|
+
{
|
|
2004
|
+
"@disable": false,
|
|
2005
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2006
|
+
"@statement": "$if",
|
|
2007
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2008
|
+
"@query": {
|
|
2009
|
+
"@param<type>": "query_param",
|
|
2010
|
+
"@param<insert>": [
|
|
2011
|
+
"pde_element"
|
|
2012
|
+
],
|
|
2013
|
+
"param": [
|
|
2014
|
+
[
|
|
2015
|
+
"pde_element",
|
|
2016
|
+
"pde_text",
|
|
2017
|
+
"pde_text_line",
|
|
2018
|
+
"pde_word",
|
|
2019
|
+
"pde_text_run",
|
|
2020
|
+
"pde_image",
|
|
2021
|
+
"pde_container",
|
|
2022
|
+
"pde_list",
|
|
2023
|
+
"pde_line",
|
|
2024
|
+
"pde_rect",
|
|
2025
|
+
"pde_table",
|
|
2026
|
+
"pde_cell",
|
|
2027
|
+
"pde_toc",
|
|
2028
|
+
"pde_header",
|
|
2029
|
+
"pde_footer",
|
|
2030
|
+
"pde_form_field",
|
|
2031
|
+
"pde_annot"
|
|
2032
|
+
]
|
|
2033
|
+
]
|
|
2034
|
+
},
|
|
2035
|
+
"@artifact": true,
|
|
2036
|
+
"@artifact<desc>": "If false on a matching rule, the artifact classification is cleared and the element is removed from the artifact list and reinserted into layout containers. If true, the item stays an artifact.",
|
|
2037
|
+
"@artifact<type>": "bool",
|
|
2038
|
+
"@artifact<value>": [
|
|
2039
|
+
"true",
|
|
2040
|
+
"false"
|
|
2041
|
+
]
|
|
2042
|
+
}
|
|
2043
|
+
],
|
|
2044
|
+
"@label_update<type>": "function",
|
|
2045
|
+
"@label_update<desc>": "Refine words or elements already marked as list markers (label levels, patterns) before lists are assembled.",
|
|
2046
|
+
"@label_update<use>": "Runs late in the labeling pass, immediately before list_detect builds nested list structure; use it to correct marker levels or template-driven label assignments.",
|
|
2047
|
+
"@label_update<insert>": [
|
|
2048
|
+
{
|
|
2049
|
+
"query": {
|
|
2050
|
+
"param": [
|
|
2051
|
+
"pde_element",
|
|
2052
|
+
"pde_word"
|
|
2053
|
+
],
|
|
2054
|
+
"$and": []
|
|
2055
|
+
}
|
|
2056
|
+
}
|
|
2057
|
+
],
|
|
2058
|
+
"@label_update": [
|
|
2059
|
+
{
|
|
2060
|
+
"@disable": false,
|
|
2061
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2062
|
+
"@statement": "$if",
|
|
2063
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2064
|
+
"@query": {
|
|
2065
|
+
"@param<type>": "query_param",
|
|
2066
|
+
"@param<insert>": [
|
|
2067
|
+
"pde_element",
|
|
2068
|
+
"pde_word"
|
|
2069
|
+
],
|
|
2070
|
+
"param": [
|
|
2071
|
+
[
|
|
2072
|
+
"pde_element",
|
|
2073
|
+
"pde_word",
|
|
2074
|
+
"pde_image",
|
|
2075
|
+
"pde_line",
|
|
2076
|
+
"pde_rect"
|
|
2077
|
+
],
|
|
2078
|
+
[
|
|
2079
|
+
"pde_word"
|
|
2080
|
+
]
|
|
2081
|
+
]
|
|
2082
|
+
},
|
|
2083
|
+
"@label<link>": "schema:values:label",
|
|
2084
|
+
"@label": "",
|
|
2085
|
+
"@name<link>": "schema:values:name",
|
|
2086
|
+
"@name": "label1"
|
|
2087
|
+
}
|
|
2088
|
+
],
|
|
2089
|
+
"@list_update<type>": "function",
|
|
2090
|
+
"@list_update<desc>": "After list_detect builds pde_list containers, refine each list: set flag (e.g. continuous), numbering style, or split back to plain texts.",
|
|
2091
|
+
"@list_update<use>": "Runs on nested list branches; use split on the main branch to dismantle a bad list while preserving item labels, or set numbering for consistent output style.",
|
|
2092
|
+
"@list_update<insert>": [
|
|
2093
|
+
{
|
|
2094
|
+
"query": {
|
|
2095
|
+
"param": [
|
|
2096
|
+
"pde_list"
|
|
2097
|
+
],
|
|
2098
|
+
"$and": []
|
|
2099
|
+
}
|
|
2100
|
+
}
|
|
2101
|
+
],
|
|
2102
|
+
"@list_update": [
|
|
2103
|
+
{
|
|
2104
|
+
"@disable": false,
|
|
2105
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2106
|
+
"@statement": "$if",
|
|
2107
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2108
|
+
"@query": {
|
|
2109
|
+
"@param<type>": "query_param",
|
|
2110
|
+
"@param<insert>": [
|
|
2111
|
+
"pde_list"
|
|
2112
|
+
],
|
|
2113
|
+
"param": [
|
|
2114
|
+
[
|
|
2115
|
+
"pde_list"
|
|
2116
|
+
]
|
|
2117
|
+
]
|
|
2118
|
+
},
|
|
2119
|
+
"@flag<link>": "schema:values:flag",
|
|
2120
|
+
"@flag": "",
|
|
2121
|
+
"@numbering<link>": "schema:values:numbering",
|
|
2122
|
+
"@numbering": "None",
|
|
2123
|
+
"@split<desc>": "If set to true, splits the list back into individual elements. The label flag remains on each list item.",
|
|
2124
|
+
"@split<type>": "bool",
|
|
2125
|
+
"@split<value>": [
|
|
2126
|
+
"true",
|
|
2127
|
+
"false"
|
|
2128
|
+
],
|
|
2129
|
+
"@split": false
|
|
2130
|
+
}
|
|
2131
|
+
],
|
|
2132
|
+
"@tag_image<type>": "function",
|
|
2133
|
+
"@tag_image<desc>": "Runs while emitting structure for a pde_image: supplies Alt / ActualText, optional artifact escape hatch, and (PDF 2.0+) whether child widgets or annotations live under the Figure.",
|
|
2134
|
+
"@tag_image<use>": "Match images to set alt/actual_text from KB; set artifact true only when both are empty and the image is decorative; enable children only on PDF 2.0 when inline controls must stay inside the figure subtree.",
|
|
2135
|
+
"@tag_image<insert>": [
|
|
2136
|
+
{
|
|
2137
|
+
"query": {
|
|
2138
|
+
"param": [
|
|
2139
|
+
"pde_image"
|
|
2140
|
+
],
|
|
2141
|
+
"$and": []
|
|
2142
|
+
}
|
|
2143
|
+
}
|
|
2144
|
+
],
|
|
2145
|
+
"@tag_image": [
|
|
2146
|
+
{
|
|
2147
|
+
"@disable": false,
|
|
2148
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2149
|
+
"@statement": "$if",
|
|
2150
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2151
|
+
"@query": {
|
|
2152
|
+
"@param<type>": "query_param",
|
|
2153
|
+
"@param<insert>": [
|
|
2154
|
+
"pde_image"
|
|
2155
|
+
],
|
|
2156
|
+
"param": [
|
|
2157
|
+
[
|
|
2158
|
+
"pde_image"
|
|
2159
|
+
]
|
|
2160
|
+
]
|
|
2161
|
+
},
|
|
2162
|
+
"@artifact<desc>": "If true and both Alt and ActualText are empty after KB, the image is tagged as an artifact instead of a Figure. If false, a Figure with empty alt/actual may still be emitted (not PDF/UA compliant).",
|
|
2163
|
+
"@artifact<type>": "bool",
|
|
2164
|
+
"@artifact<value>": [
|
|
2165
|
+
"true",
|
|
2166
|
+
"false"
|
|
2167
|
+
],
|
|
2168
|
+
"@artifact": false,
|
|
2169
|
+
"@children<desc>": "PDF 2.0+: if true, nested children (text, form fields, annotations) are tagged under the Figure; if false, they are emitted beside the figure or as artifacts per engine defaults.",
|
|
2170
|
+
"@children<type>": "bool",
|
|
2171
|
+
"@children<value>": [
|
|
2172
|
+
"true",
|
|
2173
|
+
"false"
|
|
2174
|
+
],
|
|
2175
|
+
"@children": false
|
|
2176
|
+
}
|
|
2177
|
+
],
|
|
2178
|
+
"@tag_table<type>": "function",
|
|
2179
|
+
"@tag_table<desc>": "Emits Table structure for a pde_table: borders/isolated chrome as artifacts, optional split into per-cell traversal when headers are missing, and continuous tables that reuse one Table node across pages when the element continuous flag is set.",
|
|
2180
|
+
"@tag_table<use>": "Use split+row/col only for headerless tables that must still be read cell-by-cell; set summary for AT; rely on cell_update for TH scope before tagging when possible.",
|
|
2181
|
+
"@tag_table<insert>": [
|
|
2182
|
+
{
|
|
2183
|
+
"query": {
|
|
2184
|
+
"param": [
|
|
2185
|
+
"pde_table"
|
|
2186
|
+
],
|
|
2187
|
+
"$and": []
|
|
2188
|
+
}
|
|
2189
|
+
}
|
|
2190
|
+
],
|
|
2191
|
+
"@tag_table": [
|
|
2192
|
+
{
|
|
2193
|
+
"@disable": false,
|
|
2194
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2195
|
+
"@statement": "$if",
|
|
2196
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2197
|
+
"@query": {
|
|
2198
|
+
"@param<type>": "query_param",
|
|
2199
|
+
"@param<insert>": [
|
|
2200
|
+
"pde_table"
|
|
2201
|
+
],
|
|
2202
|
+
"param": [
|
|
2203
|
+
[
|
|
2204
|
+
"pde_table"
|
|
2205
|
+
]
|
|
2206
|
+
]
|
|
2207
|
+
},
|
|
2208
|
+
"@split<desc>": "When true flatten table to individual cells in row or column order.",
|
|
2209
|
+
"@split<type>": "bool",
|
|
2210
|
+
"@split<value>": [
|
|
2211
|
+
"true",
|
|
2212
|
+
"false"
|
|
2213
|
+
],
|
|
2214
|
+
"@split": false,
|
|
2215
|
+
"@row<desc>": "Used when split is true. When true, splits by table rows (default) when flattening. When false, splits by columns.",
|
|
2216
|
+
"@row<type>": "bool",
|
|
2217
|
+
"@row<value>": [
|
|
2218
|
+
"true",
|
|
2219
|
+
"false"
|
|
2220
|
+
],
|
|
2221
|
+
"@row": true,
|
|
2222
|
+
"@summary<desc>": "Set Table Summary attribute (owner Table) for assistive technologies.",
|
|
2223
|
+
"@summary<type>": "string",
|
|
2224
|
+
"@summary": ""
|
|
2225
|
+
}
|
|
2226
|
+
],
|
|
2227
|
+
"@tag_update<type>": "function",
|
|
2228
|
+
"@tag_update<desc>": "Template hook on an existing pds_struct_elem during tagging commands: lets regex or programmatic filters accept or reject structure nodes by running KB rules against the live struct element.",
|
|
2229
|
+
"@tag_update<use>": "Used when tagging or content actions run template-based filters instead of a plain type regex; return a match to include the node in the processed set.",
|
|
2230
|
+
"@tag_update<insert>": [
|
|
2231
|
+
{
|
|
2232
|
+
"query": {
|
|
2233
|
+
"param": [
|
|
2234
|
+
"pds_struct_elem"
|
|
2235
|
+
],
|
|
2236
|
+
"$and": []
|
|
2237
|
+
}
|
|
2238
|
+
}
|
|
2239
|
+
],
|
|
2240
|
+
"@tag_update": [
|
|
2241
|
+
{
|
|
2242
|
+
"@disable": false,
|
|
2243
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2244
|
+
"@statement": "$if",
|
|
2245
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2246
|
+
"@query": {
|
|
2247
|
+
"@param<type>": "query_param",
|
|
2248
|
+
"@param<insert>": [
|
|
2249
|
+
"pds_struct_elem"
|
|
2250
|
+
],
|
|
2251
|
+
"param": [
|
|
2252
|
+
[
|
|
2253
|
+
"pds_struct_elem"
|
|
2254
|
+
]
|
|
2255
|
+
]
|
|
2256
|
+
}
|
|
2257
|
+
}
|
|
2258
|
+
],
|
|
2259
|
+
"@annot_update<type>": "function",
|
|
2260
|
+
"@annot_update<desc>": "Runs when associating structure with a pdf_annot: adjust alt and Contents (and linked schema values) after the annotation is tied to content.",
|
|
2261
|
+
"@annot_update<use>": "Normalize link or markup annotations for accessibility text; pair with annot_char_overlap in pagemap for geometry-based character association.",
|
|
2262
|
+
"@annot_update<insert>": [
|
|
2263
|
+
{
|
|
2264
|
+
"query": {
|
|
2265
|
+
"param": [
|
|
2266
|
+
"pdf_annot"
|
|
2267
|
+
],
|
|
2268
|
+
"$and": []
|
|
2269
|
+
}
|
|
2270
|
+
}
|
|
2271
|
+
],
|
|
2272
|
+
"@annot_update": [
|
|
2273
|
+
{
|
|
2274
|
+
"@disable": false,
|
|
2275
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2276
|
+
"@statement": "$if",
|
|
2277
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2278
|
+
"@query": {
|
|
2279
|
+
"@param<type>": "query_param",
|
|
2280
|
+
"@param<insert>": [
|
|
2281
|
+
"pdf_annot"
|
|
2282
|
+
],
|
|
2283
|
+
"param": [
|
|
2284
|
+
[
|
|
2285
|
+
"pdf_annot"
|
|
2286
|
+
]
|
|
2287
|
+
]
|
|
2288
|
+
},
|
|
2289
|
+
"@alt<link>": "schema:values:alt",
|
|
2290
|
+
"@alt": "",
|
|
2291
|
+
"@contents<link>": "schema:values:contents",
|
|
2292
|
+
"@contents": ""
|
|
2293
|
+
}
|
|
2294
|
+
],
|
|
2295
|
+
"@font_update<type>": "function",
|
|
2296
|
+
"@font_update<desc>": "Template hook on a pdf_font during font commands: lets regex or programmatic filters accept or reject fonts by running KB rules against the live font dictionary.",
|
|
2297
|
+
"@font_update<use>": "Used when font actions run template-based filters instead of a plain name regex; return a match to include the font in the processed set.",
|
|
2298
|
+
"@font_update<insert>": [
|
|
2299
|
+
{
|
|
2300
|
+
"query": {
|
|
2301
|
+
"param": [
|
|
2302
|
+
"pdf_font"
|
|
2303
|
+
],
|
|
2304
|
+
"$and": []
|
|
2305
|
+
}
|
|
2306
|
+
}
|
|
2307
|
+
],
|
|
2308
|
+
"@font_update": [
|
|
2309
|
+
{
|
|
2310
|
+
"@disable": false,
|
|
2311
|
+
"@disable<link>": "schema:keywords:disable",
|
|
2312
|
+
"@statement": "$if",
|
|
2313
|
+
"@statement<link>": "schema:keywords:statement",
|
|
2314
|
+
"@query": {
|
|
2315
|
+
"@param<type>": "query_param",
|
|
2316
|
+
"@param<insert>": [
|
|
2317
|
+
"pdf_font"
|
|
2318
|
+
],
|
|
2319
|
+
"param": [
|
|
2320
|
+
[
|
|
2321
|
+
"pdf_font"
|
|
2322
|
+
]
|
|
2323
|
+
]
|
|
2324
|
+
}
|
|
2325
|
+
}
|
|
2326
|
+
]
|
|
2327
|
+
},
|
|
2328
|
+
"@schema": {
|
|
2329
|
+
"statement<desc>": "Controls conditional evaluation: $if / $elif / $else branches decide which template nodes run against the current query.",
|
|
2330
|
+
"statement<use>": "Can be used in all template functions. Evaluation stops when a branch matches ($if) or falls through ($elif/$else). If omitted, the default is $if.",
|
|
2331
|
+
"statement<value>": [
|
|
2332
|
+
"$if",
|
|
2333
|
+
"$elif",
|
|
2334
|
+
"$else"
|
|
2335
|
+
],
|
|
2336
|
+
"statement<default>": "$if",
|
|
2337
|
+
"statement": {
|
|
2338
|
+
"$if": "",
|
|
2339
|
+
"$if<type>": "statement",
|
|
2340
|
+
"$if<desc>": "Primary branch: the following query is evaluated when no earlier branch matched.",
|
|
2341
|
+
"$elif": "",
|
|
2342
|
+
"$elif<type>": "statement",
|
|
2343
|
+
"$elif<desc>": "Secondary branch evaluated only if previous $if/$elif queries did not match.",
|
|
2344
|
+
"$else": "",
|
|
2345
|
+
"$else<type>": "statement",
|
|
2346
|
+
"$else<desc>": "Fallback branch when no $if or $elif matched."
|
|
2347
|
+
},
|
|
2348
|
+
"query<type>": "query",
|
|
2349
|
+
"query<desc>": "Predicate and context for a template node: parameters, logical operators, and optional variables that must pass before values are applied.",
|
|
2350
|
+
"query": {
|
|
2351
|
+
"param<desc>": "Define the number and type of input parameters.",
|
|
2352
|
+
"param<type>": "query_param",
|
|
2353
|
+
"param": {
|
|
2354
|
+
"pds_object_params": "",
|
|
2355
|
+
"pds_object_params<desc>": "A parameter that represents PdsObject. The value starts with the character $, followed by a number (e.g., $0_width). The number represents the index of the parameter in the param array.",
|
|
2356
|
+
"pds_object_params<link>": "schema:pds_object_params",
|
|
2357
|
+
"pde_element_params": "",
|
|
2358
|
+
"pde_element_params<desc>": "A parameter that represents PdeElement. The value starts with the character $, followed by a number (e.g., $0_width). The number represents the index of the parameter in the param array.",
|
|
2359
|
+
"pde_element_params<link>": "schema:pde_element_params",
|
|
2360
|
+
"pds_struct_elem_params": "",
|
|
2361
|
+
"pds_struct_elem_params<desc>": "A parameter that represents a structure-tree element. The value starts with the character $, followed by a number (e.g., $0_width). The number represents the index of the parameter in the param array.",
|
|
2362
|
+
"pds_struct_elem_params<link>": "schema:pds_struct_elem_params",
|
|
2363
|
+
"pdf_annot_params": "",
|
|
2364
|
+
"pdf_annot_params<desc>": "A parameter that represents a PDF annotation. The value starts with the character $, followed by a number (e.g., $0_width). The number represents the index of the parameter in the param array.",
|
|
2365
|
+
"pdf_annot_params<link>": "schema:pdf_annot_params",
|
|
2366
|
+
"pdf_font_params": "",
|
|
2367
|
+
"pdf_font_params<desc>": "A parameter that represents a PDF font. The value starts with the character $, followed by a number (e.g., $0_font_name). The number represents the index of the parameter in the param array.",
|
|
2368
|
+
"pdf_font_params<link>": "schema:pdf_font_params",
|
|
2369
|
+
"pdf_rect": "",
|
|
2370
|
+
"pdf_rect<link>": "schema:values:pdf_rect",
|
|
2371
|
+
"pdf_rgb": "",
|
|
2372
|
+
"pdf_rgb<link>": "schema:values:pdf_rgb",
|
|
2373
|
+
"int": "",
|
|
2374
|
+
"int<desc>": "Integer operand in a condition (no fractional part).",
|
|
2375
|
+
"int<type>": "int",
|
|
2376
|
+
"bool": "",
|
|
2377
|
+
"bool<desc>": "Boolean true/false operand in a condition.",
|
|
2378
|
+
"bool<type>": "bool",
|
|
2379
|
+
"float": "",
|
|
2380
|
+
"float<desc>": "Floating-point operand (geometry, font size, ratios).",
|
|
2381
|
+
"float<type>": "float",
|
|
2382
|
+
"string": "",
|
|
2383
|
+
"string<desc>": "String operand (text, names, regex targets).",
|
|
2384
|
+
"string<type>": "string"
|
|
2385
|
+
},
|
|
2386
|
+
"var<desc>": "User-defined variables; reference bbox macros such as left, width, or top in conditions.",
|
|
2387
|
+
"var<type>": "var",
|
|
2388
|
+
"var": {
|
|
2389
|
+
"0_value": "",
|
|
2390
|
+
"0_value<type>": "var_value",
|
|
2391
|
+
"0_value<value>": [
|
|
2392
|
+
"left",
|
|
2393
|
+
"bottom",
|
|
2394
|
+
"right",
|
|
2395
|
+
"top",
|
|
2396
|
+
"width",
|
|
2397
|
+
"height"
|
|
2398
|
+
]
|
|
2399
|
+
},
|
|
2400
|
+
"logical_operators<link>": "schema:logical_operators",
|
|
2401
|
+
"logical_operators": ""
|
|
2402
|
+
},
|
|
2403
|
+
"logical_operators<desc>": "Combine multiple query conditions: $and requires all branches, $or any branch, $not inverts one branch.",
|
|
2404
|
+
"logical_operators<use>": "Nest under query to build compound predicates on bbox, text, or parameter fields.",
|
|
2405
|
+
"logical_operators<type>": "string",
|
|
2406
|
+
"logical_operators<value>": [
|
|
2407
|
+
"$and",
|
|
2408
|
+
"$or",
|
|
2409
|
+
"$not"
|
|
2410
|
+
],
|
|
2411
|
+
"logical_operators": {
|
|
2412
|
+
"$and<type>": "logical_operator",
|
|
2413
|
+
"$and<desc>": "Logical AND. All sub-conditions must be true.",
|
|
2414
|
+
"$and<num_values>": 2,
|
|
2415
|
+
"$and<insert>": [],
|
|
2416
|
+
"$and": {
|
|
2417
|
+
"$and": "",
|
|
2418
|
+
"$or": "",
|
|
2419
|
+
"$not": "",
|
|
2420
|
+
"condition": "",
|
|
2421
|
+
"condition<link>": "schema:condition"
|
|
2422
|
+
},
|
|
2423
|
+
"$or<type>": "logical_operator",
|
|
2424
|
+
"$or<desc>": "Logical OR. At least one sub-condition must be true.",
|
|
2425
|
+
"$or<num_values>": 2,
|
|
2426
|
+
"$or<insert>": [],
|
|
2427
|
+
"$or": {
|
|
2428
|
+
"$and": "",
|
|
2429
|
+
"$or": "",
|
|
2430
|
+
"$not": "",
|
|
2431
|
+
"condition": "",
|
|
2432
|
+
"condition<link>": "schema:condition"
|
|
2433
|
+
},
|
|
2434
|
+
"$not<type>": "logical_operator",
|
|
2435
|
+
"$not<desc>": "Logical NOT.",
|
|
2436
|
+
"$not<num_values>": 1,
|
|
2437
|
+
"$not<insert>": [],
|
|
2438
|
+
"$not": {
|
|
2439
|
+
"$and": "",
|
|
2440
|
+
"$or": "",
|
|
2441
|
+
"$not": "",
|
|
2442
|
+
"condition": "",
|
|
2443
|
+
"condition<link>": "schema:condition"
|
|
2444
|
+
}
|
|
2445
|
+
},
|
|
2446
|
+
"comparison_operators<desc>": "Scalar comparisons ($eq, $ne, $lt, …), regex match ($regex), and bbox containment ($in / $nin).",
|
|
2447
|
+
"comparison_operators<use>": "Combine under logical_operators inside query.$and / $or to filter template matches.",
|
|
2448
|
+
"comparison_operators<type>": "string",
|
|
2449
|
+
"comparison_operators<value>": [
|
|
2450
|
+
"$eq",
|
|
2451
|
+
"$ne",
|
|
2452
|
+
"$lt",
|
|
2453
|
+
"$lte",
|
|
2454
|
+
"$gt",
|
|
2455
|
+
"$gte",
|
|
2456
|
+
"$regex",
|
|
2457
|
+
"$in",
|
|
2458
|
+
"$nin"
|
|
2459
|
+
],
|
|
2460
|
+
"comparison_operators": {
|
|
2461
|
+
"$eq": "",
|
|
2462
|
+
"$eq<type>": "comparison_operator",
|
|
2463
|
+
"$eq<desc>": "Field equals the given literal (string, number, or bool).",
|
|
2464
|
+
"$eq<use>": "Exact match on name, tag_type, flag string, or numeric geometry.",
|
|
2465
|
+
"$eq<num_values>": 1,
|
|
2466
|
+
"$eq<types>": [
|
|
2467
|
+
"int",
|
|
2468
|
+
"float",
|
|
2469
|
+
"string",
|
|
2470
|
+
"bool"
|
|
2471
|
+
],
|
|
2472
|
+
"$ne": "",
|
|
2473
|
+
"$ne<type>": "comparison_operator",
|
|
2474
|
+
"$ne<desc>": "Field differs from the given literal.",
|
|
2475
|
+
"$ne<use>": "Exclude a specific tag type or marker level from a broad rule.",
|
|
2476
|
+
"$ne<num_values>": 1,
|
|
2477
|
+
"$ne<types>": [
|
|
2478
|
+
"int",
|
|
2479
|
+
"float",
|
|
2480
|
+
"string",
|
|
2481
|
+
"bool"
|
|
2482
|
+
],
|
|
2483
|
+
"$lt": "",
|
|
2484
|
+
"$lt<type>": "comparison_operator",
|
|
2485
|
+
"$lt<desc>": "Numeric field strictly less than the operand.",
|
|
2486
|
+
"$lt<use>": "Upper-bound font size, width, or column index comparisons.",
|
|
2487
|
+
"$lt<num_values>": 1,
|
|
2488
|
+
"$lt<types>": [
|
|
2489
|
+
"int",
|
|
2490
|
+
"float"
|
|
2491
|
+
],
|
|
2492
|
+
"$lte": "",
|
|
2493
|
+
"$lte<type>": "comparison_operator",
|
|
2494
|
+
"$lte<desc>": "Numeric field less than or equal to the operand.",
|
|
2495
|
+
"$lte<use>": "Inclusive limits for spacing or table geometry.",
|
|
2496
|
+
"$lte<num_values>": 1,
|
|
2497
|
+
"$lte<types>": [
|
|
2498
|
+
"int",
|
|
2499
|
+
"float"
|
|
2500
|
+
],
|
|
2501
|
+
"$gt": "",
|
|
2502
|
+
"$gt<type>": "comparison_operator",
|
|
2503
|
+
"$gt<desc>": "Numeric field strictly greater than the operand.",
|
|
2504
|
+
"$gt<use>": "Minimum-size filters (e.g. drop tiny icons).",
|
|
2505
|
+
"$gt<num_values>": 1,
|
|
2506
|
+
"$gt<types>": [
|
|
2507
|
+
"int",
|
|
2508
|
+
"float"
|
|
2509
|
+
],
|
|
2510
|
+
"$gte": "",
|
|
2511
|
+
"$gte<type>": "comparison_operator",
|
|
2512
|
+
"$gte<desc>": "Numeric field greater than or equal to the operand.",
|
|
2513
|
+
"$gte<use>": "Minimum inclusive thresholds for area or font metrics.",
|
|
2514
|
+
"$gte<num_values>": 1,
|
|
2515
|
+
"$gte<types>": [
|
|
2516
|
+
"int",
|
|
2517
|
+
"float"
|
|
2518
|
+
],
|
|
2519
|
+
"$regex": "",
|
|
2520
|
+
"$regex<type>": "comparison_operator",
|
|
2521
|
+
"$regex<desc>": "String field must match the given regular expression pattern.",
|
|
2522
|
+
"$regex<use>": "Use on text, name, or tag fields; escape special characters per engine rules.",
|
|
2523
|
+
"$regex<num_values>": 1,
|
|
2524
|
+
"$regex<types>": [
|
|
2525
|
+
"string"
|
|
2526
|
+
],
|
|
2527
|
+
"$in": "",
|
|
2528
|
+
"$in<type>": "comparison_operator",
|
|
2529
|
+
"$in<desc>": "True when the parameter’s bbox lies inside the comparison bbox (spatial containment).",
|
|
2530
|
+
"$in<use>": "Typical for hit-testing a region against an anchor or column band.",
|
|
2531
|
+
"$in<num_values>": 1,
|
|
2532
|
+
"$in<types>": [
|
|
2533
|
+
"bbox"
|
|
2534
|
+
],
|
|
2535
|
+
"$nin": "",
|
|
2536
|
+
"$nin<type>": "comparison_operator",
|
|
2537
|
+
"$nin<desc>": "True when the parameter’s bbox does not lie inside the comparison bbox.",
|
|
2538
|
+
"$nin<use>": "Exclude elements whose geometry falls inside a masked region.",
|
|
2539
|
+
"$nin<num_values>": 1,
|
|
2540
|
+
"$nin<types>": [
|
|
2541
|
+
"bbox"
|
|
2542
|
+
]
|
|
2543
|
+
},
|
|
2544
|
+
"pds_object_params<desc>": "Catalog of low-level PDF content object shapes (pds_text, pds_path, …) usable as QUERY param types for object_update.",
|
|
2545
|
+
"pds_object_params": {
|
|
2546
|
+
"pds_text<desc>": "Text page object",
|
|
2547
|
+
"pds_text": {
|
|
2548
|
+
"width": "",
|
|
2549
|
+
"width<link>": "schema:values:width",
|
|
2550
|
+
"height": "",
|
|
2551
|
+
"height<link>": "schema:values:height",
|
|
2552
|
+
"left": "",
|
|
2553
|
+
"left<link>": "schema:values:left",
|
|
2554
|
+
"right": "",
|
|
2555
|
+
"right<link>": "schema:values:right",
|
|
2556
|
+
"top": "",
|
|
2557
|
+
"top<link>": "schema:values:top",
|
|
2558
|
+
"bottom": "",
|
|
2559
|
+
"bottom<link>": "schema:values:bottom",
|
|
2560
|
+
"bbox<link>": "schema:values:bbox",
|
|
2561
|
+
"bbox": {
|
|
2562
|
+
"left": 0,
|
|
2563
|
+
"bottom": 0,
|
|
2564
|
+
"right": 0,
|
|
2565
|
+
"top": 0
|
|
2566
|
+
},
|
|
2567
|
+
"has_fill": "",
|
|
2568
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2569
|
+
"has_stroke": "",
|
|
2570
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2571
|
+
"fill_color": "",
|
|
2572
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2573
|
+
"stroke_color": "",
|
|
2574
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2575
|
+
"fill_alpha": "",
|
|
2576
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
2577
|
+
"stroke_alpha": "",
|
|
2578
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
2579
|
+
"has_transparency": "",
|
|
2580
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
2581
|
+
"children_num": "",
|
|
2582
|
+
"children_num<link>": "schema:values:children_num",
|
|
2583
|
+
"artifact": "",
|
|
2584
|
+
"artifact<link>": "schema:values:artifact",
|
|
2585
|
+
"mcid": "",
|
|
2586
|
+
"mcid<link>": "schema:values:mcid",
|
|
2587
|
+
"font_size": "",
|
|
2588
|
+
"font_size<link>": "schema:values:font_size",
|
|
2589
|
+
"font_name": "",
|
|
2590
|
+
"font_name<link>": "schema:values:font_name",
|
|
2591
|
+
"text": "",
|
|
2592
|
+
"text<link>": "schema:values:text",
|
|
2593
|
+
"lang": "",
|
|
2594
|
+
"lang<link>": "schema:values:lang",
|
|
2595
|
+
"alt": "",
|
|
2596
|
+
"alt<link>": "schema:values:alt",
|
|
2597
|
+
"actual_text": "",
|
|
2598
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
2599
|
+
"expansion": "",
|
|
2600
|
+
"expansion<link>": "schema:values:expansion"
|
|
2601
|
+
},
|
|
2602
|
+
"pds_path": "",
|
|
2603
|
+
"pds_path<link>": "schema:pds_object_params:pds_text",
|
|
2604
|
+
"pds_image": "",
|
|
2605
|
+
"pds_image<link>": "schema:pds_object_params:pds_text",
|
|
2606
|
+
"pds_shading": "",
|
|
2607
|
+
"pds_shading<link>": "schema:pds_object_params:pds_text",
|
|
2608
|
+
"pds_form": "",
|
|
2609
|
+
"pds_form<link>": "schema:pds_object_params:pds_text",
|
|
2610
|
+
"pds_object": {
|
|
2611
|
+
"type": "",
|
|
2612
|
+
"type<desc>": "Concrete pds_* subtype for this object instance.",
|
|
2613
|
+
"type<type>": "string",
|
|
2614
|
+
"type<value>": [
|
|
2615
|
+
"pds_object",
|
|
2616
|
+
"pds_text",
|
|
2617
|
+
"pds_path",
|
|
2618
|
+
"pds_image",
|
|
2619
|
+
"pds_shading",
|
|
2620
|
+
"pds_form"
|
|
2621
|
+
],
|
|
2622
|
+
"width": "",
|
|
2623
|
+
"width<link>": "schema:values:width",
|
|
2624
|
+
"height": "",
|
|
2625
|
+
"height<link>": "schema:values:height",
|
|
2626
|
+
"left": "",
|
|
2627
|
+
"left<link>": "schema:values:left",
|
|
2628
|
+
"right": "",
|
|
2629
|
+
"right<link>": "schema:values:right",
|
|
2630
|
+
"top": "",
|
|
2631
|
+
"top<link>": "schema:values:top",
|
|
2632
|
+
"bottom": "",
|
|
2633
|
+
"bottom<link>": "schema:values:bottom",
|
|
2634
|
+
"bbox<link>": "schema:values:bbox",
|
|
2635
|
+
"bbox": {
|
|
2636
|
+
"left": 0,
|
|
2637
|
+
"bottom": 0,
|
|
2638
|
+
"right": 0,
|
|
2639
|
+
"top": 0
|
|
2640
|
+
},
|
|
2641
|
+
"has_fill": "",
|
|
2642
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2643
|
+
"has_stroke": "",
|
|
2644
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2645
|
+
"fill_color": "",
|
|
2646
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2647
|
+
"stroke_color": "",
|
|
2648
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2649
|
+
"fill_alpha": "",
|
|
2650
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
2651
|
+
"stroke_alpha": "",
|
|
2652
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
2653
|
+
"has_transparency": "",
|
|
2654
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
2655
|
+
"children_num": "",
|
|
2656
|
+
"children_num<link>": "schema:values:children_num",
|
|
2657
|
+
"artifact": "",
|
|
2658
|
+
"artifact<link>": "schema:values:artifact",
|
|
2659
|
+
"mcid": "",
|
|
2660
|
+
"mcid<link>": "schema:values:mcid",
|
|
2661
|
+
"font_size": "",
|
|
2662
|
+
"font_size<link>": "schema:values:font_size",
|
|
2663
|
+
"font_name": "",
|
|
2664
|
+
"font_name<link>": "schema:values:font_name",
|
|
2665
|
+
"text": "",
|
|
2666
|
+
"text<link>": "schema:values:text",
|
|
2667
|
+
"lang": "",
|
|
2668
|
+
"lang<link>": "schema:values:lang",
|
|
2669
|
+
"alt": "",
|
|
2670
|
+
"alt<link>": "schema:values:alt",
|
|
2671
|
+
"actual_text": "",
|
|
2672
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
2673
|
+
"expansion": "",
|
|
2674
|
+
"expansion<link>": "schema:values:expansion"
|
|
2675
|
+
}
|
|
2676
|
+
},
|
|
2677
|
+
"pds_struct_elem_params<desc>": "Structure-tree element parameters for tag_update and similar hooks (pds_struct_elem).",
|
|
2678
|
+
"pdf_font_params<desc>": "Font dictionary fields available when pdf_font is a query parameter (font_update, font filters).",
|
|
2679
|
+
"pdf_font_params": {
|
|
2680
|
+
"pdf_font": {
|
|
2681
|
+
"font_name": "",
|
|
2682
|
+
"font_name<link>": "schema:values:font_name",
|
|
2683
|
+
"cos_obj": "",
|
|
2684
|
+
"cos_obj<link>": "schema:values:cos_obj",
|
|
2685
|
+
"font_type": "",
|
|
2686
|
+
"font_type<link>": "schema:values:font_type",
|
|
2687
|
+
"embedded": "",
|
|
2688
|
+
"embedded<link>": "schema:values:embedded",
|
|
2689
|
+
"bold": "",
|
|
2690
|
+
"bold<link>": "schema:values:bold",
|
|
2691
|
+
"italic": "",
|
|
2692
|
+
"italic<link>": "schema:values:italic",
|
|
2693
|
+
"fixed_width": "",
|
|
2694
|
+
"fixed_width<link>": "schema:values:fixed_width",
|
|
2695
|
+
"vertical": "",
|
|
2696
|
+
"vertical<link>": "schema:values:vertical",
|
|
2697
|
+
"missing_unicode": "",
|
|
2698
|
+
"missing_unicode<link>": "schema:values:missing_unicode"
|
|
2699
|
+
}
|
|
2700
|
+
},
|
|
2701
|
+
"pds_struct_elem_params": {
|
|
2702
|
+
"pds_struct_elem": {
|
|
2703
|
+
"type": "",
|
|
2704
|
+
"type<desc>": "Structure element role/type name as in the PDF structure tree.",
|
|
2705
|
+
"type<type>": "string",
|
|
2706
|
+
"type<value>": [
|
|
2707
|
+
"pds_struct_elem"
|
|
2708
|
+
],
|
|
2709
|
+
"width": "",
|
|
2710
|
+
"width<link>": "schema:values:width",
|
|
2711
|
+
"height": "",
|
|
2712
|
+
"height<link>": "schema:values:height",
|
|
2713
|
+
"left": "",
|
|
2714
|
+
"left<link>": "schema:values:left",
|
|
2715
|
+
"right": "",
|
|
2716
|
+
"right<link>": "schema:values:right",
|
|
2717
|
+
"top": "",
|
|
2718
|
+
"top<link>": "schema:values:top",
|
|
2719
|
+
"bottom": "",
|
|
2720
|
+
"bottom<link>": "schema:values:bottom",
|
|
2721
|
+
"bbox<link>": "schema:values:bbox",
|
|
2722
|
+
"bbox": {
|
|
2723
|
+
"left": 0,
|
|
2724
|
+
"bottom": 0,
|
|
2725
|
+
"right": 0,
|
|
2726
|
+
"top": 0
|
|
2727
|
+
},
|
|
2728
|
+
"tag_type": "",
|
|
2729
|
+
"tag_type<link>": "schema:values:tag_type",
|
|
2730
|
+
"role_mapped_tag_type": "",
|
|
2731
|
+
"role_mapped_tag_type<link>": "schema:values:tag_type",
|
|
2732
|
+
"parent_tag_type": "",
|
|
2733
|
+
"parent_tag_type<link>": "schema:values:parent_tag_type",
|
|
2734
|
+
"children_tag_type": "",
|
|
2735
|
+
"children_tag_type<link>": "schema:values:children_tag_type",
|
|
2736
|
+
"is_standard_tag": "",
|
|
2737
|
+
"is_standard_tag<desc>": "A value identifying whether the tag is a standard tag for detected PDF.",
|
|
2738
|
+
"is_standard_tag<type>": "bool",
|
|
2739
|
+
"is_standard_tag<value>": [
|
|
2740
|
+
"true",
|
|
2741
|
+
"false"
|
|
2742
|
+
],
|
|
2743
|
+
"id": "",
|
|
2744
|
+
"id<link>": "schema:values:id",
|
|
2745
|
+
"title": "",
|
|
2746
|
+
"title<link>": "schema:values:title",
|
|
2747
|
+
"lang": "",
|
|
2748
|
+
"lang<link>": "schema:values:lang",
|
|
2749
|
+
"alt": "",
|
|
2750
|
+
"alt<link>": "schema:values:alt",
|
|
2751
|
+
"actual_text": "",
|
|
2752
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
2753
|
+
"expansion": "",
|
|
2754
|
+
"expansion<link>": "schema:values:expansion",
|
|
2755
|
+
"text": "",
|
|
2756
|
+
"text<link>": "schema:values:text",
|
|
2757
|
+
"children_num": "",
|
|
2758
|
+
"children_num<link>": "schema:values:children_num",
|
|
2759
|
+
"has_fill": "",
|
|
2760
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2761
|
+
"has_stroke": "",
|
|
2762
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2763
|
+
"fill_color": "",
|
|
2764
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2765
|
+
"stroke_color": "",
|
|
2766
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2767
|
+
"fill_alpha": "",
|
|
2768
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
2769
|
+
"stroke_alpha": "",
|
|
2770
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
2771
|
+
"has_transparency": "",
|
|
2772
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
2773
|
+
"font_size": "",
|
|
2774
|
+
"font_size<link>": "schema:values:font_size",
|
|
2775
|
+
"font_name": "",
|
|
2776
|
+
"font_name<link>": "schema:values:font_name",
|
|
2777
|
+
"has_content": "",
|
|
2778
|
+
"has_content<link>": "schema:values:has_content",
|
|
2779
|
+
"has_annot": "",
|
|
2780
|
+
"has_annot<link>": "schema:values:has_annot",
|
|
2781
|
+
"row_num": "",
|
|
2782
|
+
"row_num<link>": "schema:values:row_num",
|
|
2783
|
+
"col_num": "",
|
|
2784
|
+
"col_num<link>": "schema:values:col_num",
|
|
2785
|
+
"cell_column": "",
|
|
2786
|
+
"cell_column<link>": "schema:values:cell_column",
|
|
2787
|
+
"cell_row": "",
|
|
2788
|
+
"cell_row<link>": "schema:values:cell_row",
|
|
2789
|
+
"cell_row_span": "",
|
|
2790
|
+
"cell_row_span<link>": "schema:values:cell_row_span",
|
|
2791
|
+
"cell_column_span": "",
|
|
2792
|
+
"cell_column_span<link>": "schema:values:cell_column_span",
|
|
2793
|
+
"cell_scope": "column",
|
|
2794
|
+
"cell_scope<link>": "schema:values:cell_scope"
|
|
2795
|
+
}
|
|
2796
|
+
},
|
|
2797
|
+
"pdf_annot_params<desc>": "Annotation fields available when pdf_annot is a query parameter (annot_update, linking).",
|
|
2798
|
+
"pdf_annot_params": {
|
|
2799
|
+
"pdf_annot": {
|
|
2800
|
+
"type": "",
|
|
2801
|
+
"type<desc>": "Annotation subtype (link, text, highlight, …) exposed to the template.",
|
|
2802
|
+
"type<type>": "string",
|
|
2803
|
+
"type<value>": [
|
|
2804
|
+
"pdf_annot"
|
|
2805
|
+
],
|
|
2806
|
+
"width": "",
|
|
2807
|
+
"width<link>": "schema:values:width",
|
|
2808
|
+
"height": "",
|
|
2809
|
+
"height<link>": "schema:values:height",
|
|
2810
|
+
"left": "",
|
|
2811
|
+
"left<link>": "schema:values:left",
|
|
2812
|
+
"right": "",
|
|
2813
|
+
"right<link>": "schema:values:right",
|
|
2814
|
+
"top": "",
|
|
2815
|
+
"top<link>": "schema:values:top",
|
|
2816
|
+
"bottom": "",
|
|
2817
|
+
"bottom<link>": "schema:values:bottom",
|
|
2818
|
+
"bbox<link>": "schema:values:bbox",
|
|
2819
|
+
"bbox": {
|
|
2820
|
+
"left": 0,
|
|
2821
|
+
"bottom": 0,
|
|
2822
|
+
"right": 0,
|
|
2823
|
+
"top": 0
|
|
2824
|
+
},
|
|
2825
|
+
"annot_type": "",
|
|
2826
|
+
"annot_type<link>": "schema:values:annot_type",
|
|
2827
|
+
"font_size": "",
|
|
2828
|
+
"font_size<link>": "schema:values:font_size",
|
|
2829
|
+
"font_name": "",
|
|
2830
|
+
"font_name<link>": "schema:values:font_name",
|
|
2831
|
+
"contents": "",
|
|
2832
|
+
"contents<link>": "schema:values:contents",
|
|
2833
|
+
"annot_flag": "",
|
|
2834
|
+
"annot_flag<link>": "schema:values:annot_flag",
|
|
2835
|
+
"tag_type": "",
|
|
2836
|
+
"tag_type<link>": "schema:values:tag_type"
|
|
2837
|
+
}
|
|
2838
|
+
},
|
|
2839
|
+
"pde_element_params<desc>": "Layout element field catalog (pde_text, pde_image, …) referenced from QUERY param arrays.",
|
|
2840
|
+
"pde_element_params": {
|
|
2841
|
+
"pde_text": {
|
|
2842
|
+
"width": "",
|
|
2843
|
+
"width<link>": "schema:values:width",
|
|
2844
|
+
"height": "",
|
|
2845
|
+
"height<link>": "schema:values:height",
|
|
2846
|
+
"left": "",
|
|
2847
|
+
"left<link>": "schema:values:left",
|
|
2848
|
+
"right": "",
|
|
2849
|
+
"right<link>": "schema:values:right",
|
|
2850
|
+
"top": "",
|
|
2851
|
+
"top<link>": "schema:values:top",
|
|
2852
|
+
"bottom": "",
|
|
2853
|
+
"bottom<link>": "schema:values:bottom",
|
|
2854
|
+
"baseline_x": "",
|
|
2855
|
+
"baseline_x<link>": "schema:values:baseline_x",
|
|
2856
|
+
"baseline_y": "",
|
|
2857
|
+
"baseline_y<link>": "schema:values:baseline_y",
|
|
2858
|
+
"bbox": {
|
|
2859
|
+
"left": 0,
|
|
2860
|
+
"bottom": 0,
|
|
2861
|
+
"right": 0,
|
|
2862
|
+
"top": 0
|
|
2863
|
+
},
|
|
2864
|
+
"bbox<link>": "schema:values:bbox",
|
|
2865
|
+
"label": "",
|
|
2866
|
+
"label<link>": "schema:values:label",
|
|
2867
|
+
"toc": "",
|
|
2868
|
+
"toc<link>": "schema:values:toc",
|
|
2869
|
+
"has_fill": "",
|
|
2870
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2871
|
+
"has_stroke": "",
|
|
2872
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2873
|
+
"fill_color": "",
|
|
2874
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2875
|
+
"stroke_color": "",
|
|
2876
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2877
|
+
"fill_alpha": "",
|
|
2878
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
2879
|
+
"stroke_alpha": "",
|
|
2880
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
2881
|
+
"has_transparency": "",
|
|
2882
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
2883
|
+
"lang": "",
|
|
2884
|
+
"lang<link>": "schema:values:lang",
|
|
2885
|
+
"alt": "",
|
|
2886
|
+
"alt<link>": "schema:values:alt",
|
|
2887
|
+
"actual_text": "",
|
|
2888
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
2889
|
+
"expansion": "",
|
|
2890
|
+
"expansion<link>": "schema:values:expansion",
|
|
2891
|
+
"font_size": "",
|
|
2892
|
+
"font_size<link>": "schema:values:font_size",
|
|
2893
|
+
"font_name": "",
|
|
2894
|
+
"font_name<link>": "schema:values:font_name",
|
|
2895
|
+
"text": "",
|
|
2896
|
+
"text<link>": "schema:values:text",
|
|
2897
|
+
"angle": "",
|
|
2898
|
+
"angle<link>": "schema:values:angle",
|
|
2899
|
+
"heading": "",
|
|
2900
|
+
"heading<link>": "schema:values:heading",
|
|
2901
|
+
"name": "",
|
|
2902
|
+
"name<link>": "schema:values:name"
|
|
2903
|
+
},
|
|
2904
|
+
"pde_text_line": "",
|
|
2905
|
+
"pde_text_line<link>": "schema:pde_element_params:pde_text",
|
|
2906
|
+
"pde_word": "",
|
|
2907
|
+
"pde_word<link>": "schema:pde_element_params:pde_text",
|
|
2908
|
+
"pde_text_run": "",
|
|
2909
|
+
"pde_text_run<link>": "schema:pde_element_params:pde_text",
|
|
2910
|
+
"pde_image": {
|
|
2911
|
+
"width": "",
|
|
2912
|
+
"width<link>": "schema:values:width",
|
|
2913
|
+
"height": "",
|
|
2914
|
+
"height<link>": "schema:values:height",
|
|
2915
|
+
"left": "",
|
|
2916
|
+
"left<link>": "schema:values:left",
|
|
2917
|
+
"right": "",
|
|
2918
|
+
"right<link>": "schema:values:right",
|
|
2919
|
+
"top": "",
|
|
2920
|
+
"top<link>": "schema:values:top",
|
|
2921
|
+
"bottom": "",
|
|
2922
|
+
"bottom<link>": "schema:values:bottom",
|
|
2923
|
+
"bbox": {
|
|
2924
|
+
"left": 0,
|
|
2925
|
+
"bottom": 0,
|
|
2926
|
+
"right": 0,
|
|
2927
|
+
"top": 0
|
|
2928
|
+
},
|
|
2929
|
+
"bbox<link>": "schema:values:bbox",
|
|
2930
|
+
"font_size": "",
|
|
2931
|
+
"font_size<link>": "schema:values:font_size",
|
|
2932
|
+
"font_name": "",
|
|
2933
|
+
"font_name<link>": "schema:values:font_name",
|
|
2934
|
+
"label": "",
|
|
2935
|
+
"label<link>": "schema:values:label",
|
|
2936
|
+
"has_fill": "",
|
|
2937
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2938
|
+
"has_stroke": "",
|
|
2939
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2940
|
+
"fill_color": "",
|
|
2941
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2942
|
+
"stroke_color": "",
|
|
2943
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2944
|
+
"fill_alpha": "",
|
|
2945
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
2946
|
+
"stroke_alpha": "",
|
|
2947
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
2948
|
+
"has_transparency": "",
|
|
2949
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
2950
|
+
"lang": "",
|
|
2951
|
+
"lang<link>": "schema:values:lang",
|
|
2952
|
+
"alt": "",
|
|
2953
|
+
"alt<link>": "schema:values:alt",
|
|
2954
|
+
"actual_text": "",
|
|
2955
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
2956
|
+
"expansion": "",
|
|
2957
|
+
"expansion<link>": "schema:values:expansion",
|
|
2958
|
+
"children_num": 0,
|
|
2959
|
+
"children_num<link>": "schema:values:children_num",
|
|
2960
|
+
"object_num": 0,
|
|
2961
|
+
"object_num<link>": "schema:values:object_num",
|
|
2962
|
+
"name": "",
|
|
2963
|
+
"name<link>": "schema:values:name"
|
|
2964
|
+
},
|
|
2965
|
+
"pde_container": "",
|
|
2966
|
+
"pde_container<link>": "schema:pde_element_params:pde_image",
|
|
2967
|
+
"pde_list": "",
|
|
2968
|
+
"pde_list<link>": "schema:pde_element_params:pde_image",
|
|
2969
|
+
"pde_rect": "",
|
|
2970
|
+
"pde_rect<link>": "schema:pde_element_params:pde_image",
|
|
2971
|
+
"pde_cell": {
|
|
2972
|
+
"width": "",
|
|
2973
|
+
"width<link>": "schema:values:width",
|
|
2974
|
+
"height": "",
|
|
2975
|
+
"height<link>": "schema:values:height",
|
|
2976
|
+
"left": "",
|
|
2977
|
+
"left<link>": "schema:values:left",
|
|
2978
|
+
"right": "",
|
|
2979
|
+
"right<link>": "schema:values:right",
|
|
2980
|
+
"top": "",
|
|
2981
|
+
"top<link>": "schema:values:top",
|
|
2982
|
+
"bottom": "",
|
|
2983
|
+
"bottom<link>": "schema:values:bottom",
|
|
2984
|
+
"bbox": {
|
|
2985
|
+
"left": 0,
|
|
2986
|
+
"bottom": 0,
|
|
2987
|
+
"right": 0,
|
|
2988
|
+
"top": 0
|
|
2989
|
+
},
|
|
2990
|
+
"bbox<link>": "schema:values:bbox",
|
|
2991
|
+
"has_fill": "",
|
|
2992
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
2993
|
+
"has_stroke": "",
|
|
2994
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
2995
|
+
"fill_color": "",
|
|
2996
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
2997
|
+
"stroke_color": "",
|
|
2998
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
2999
|
+
"fill_alpha": "",
|
|
3000
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
3001
|
+
"stroke_alpha": "",
|
|
3002
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
3003
|
+
"has_transparency": "",
|
|
3004
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
3005
|
+
"font_size": "",
|
|
3006
|
+
"font_size<link>": "schema:values:font_size",
|
|
3007
|
+
"font_name": "",
|
|
3008
|
+
"font_name<link>": "schema:values:font_name",
|
|
3009
|
+
"text": "",
|
|
3010
|
+
"text<link>": "schema:values:text",
|
|
3011
|
+
"cell_column": "",
|
|
3012
|
+
"cell_column<link>": "schema:values:cell_column",
|
|
3013
|
+
"cell_row": "",
|
|
3014
|
+
"cell_row<link>": "schema:values:cell_row",
|
|
3015
|
+
"cell_row_span": "",
|
|
3016
|
+
"cell_row_span<link>": "schema:values:cell_row_span",
|
|
3017
|
+
"cell_column_span": "",
|
|
3018
|
+
"cell_column_span<link>": "schema:values:cell_column_span",
|
|
3019
|
+
"cell_scope": "",
|
|
3020
|
+
"cell_scope<link>": "schema:values:cell_scope",
|
|
3021
|
+
"children_num": 0,
|
|
3022
|
+
"children_num<link>": "schema:values:children_num",
|
|
3023
|
+
"name": "",
|
|
3024
|
+
"name<link>": "schema:values:name"
|
|
3025
|
+
},
|
|
3026
|
+
"pde_toc": "",
|
|
3027
|
+
"pde_toc<link>": "schema:pde_element_params:pde_image",
|
|
3028
|
+
"pde_header": "",
|
|
3029
|
+
"pde_header<link>": "schema:pde_element_params:pde_image",
|
|
3030
|
+
"pde_footer": "",
|
|
3031
|
+
"pde_footer<link>": "schema:pde_element_params:pde_image",
|
|
3032
|
+
"pde_line": {
|
|
3033
|
+
"width": "",
|
|
3034
|
+
"width<link>": "schema:values:width",
|
|
3035
|
+
"height": "",
|
|
3036
|
+
"height<link>": "schema:values:height",
|
|
3037
|
+
"left": "",
|
|
3038
|
+
"left<link>": "schema:values:left",
|
|
3039
|
+
"right": "",
|
|
3040
|
+
"right<link>": "schema:values:right",
|
|
3041
|
+
"top": "",
|
|
3042
|
+
"top<link>": "schema:values:top",
|
|
3043
|
+
"bottom": "",
|
|
3044
|
+
"bottom<link>": "schema:values:bottom",
|
|
3045
|
+
"bbox": {
|
|
3046
|
+
"left": 0,
|
|
3047
|
+
"bottom": 0,
|
|
3048
|
+
"right": 0,
|
|
3049
|
+
"top": 0
|
|
3050
|
+
},
|
|
3051
|
+
"bbox<link>": "schema:values:bbox",
|
|
3052
|
+
"label": "",
|
|
3053
|
+
"label<link>": "schema:values:label",
|
|
3054
|
+
"has_fill": "",
|
|
3055
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
3056
|
+
"has_stroke": "",
|
|
3057
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
3058
|
+
"fill_color": "",
|
|
3059
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
3060
|
+
"stroke_color": "",
|
|
3061
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
3062
|
+
"fill_alpha": "",
|
|
3063
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
3064
|
+
"stroke_alpha": "",
|
|
3065
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
3066
|
+
"has_transparency": "",
|
|
3067
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
3068
|
+
"lang": "",
|
|
3069
|
+
"lang<link>": "schema:values:lang",
|
|
3070
|
+
"alt": "",
|
|
3071
|
+
"alt<link>": "schema:values:alt",
|
|
3072
|
+
"actual_text": "",
|
|
3073
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
3074
|
+
"expansion": "",
|
|
3075
|
+
"expansion<link>": "schema:values:expansion",
|
|
3076
|
+
"name": "",
|
|
3077
|
+
"name<link>": "schema:values:name"
|
|
3078
|
+
},
|
|
3079
|
+
"pde_form_field": "",
|
|
3080
|
+
"pde_form_field<link>": "schema:pde_element_params:pde_line",
|
|
3081
|
+
"pde_annot": "",
|
|
3082
|
+
"pde_annot<link>": "schema:pde_element_params:pde_line",
|
|
3083
|
+
"pde_table": {
|
|
3084
|
+
"width": "",
|
|
3085
|
+
"width<link>": "schema:values:width",
|
|
3086
|
+
"height": "",
|
|
3087
|
+
"height<link>": "schema:values:height",
|
|
3088
|
+
"left": "",
|
|
3089
|
+
"left<link>": "schema:values:left",
|
|
3090
|
+
"right": "",
|
|
3091
|
+
"right<link>": "schema:values:right",
|
|
3092
|
+
"top": "",
|
|
3093
|
+
"top<link>": "schema:values:top",
|
|
3094
|
+
"bottom": "",
|
|
3095
|
+
"bottom<link>": "schema:values:bottom",
|
|
3096
|
+
"bbox": {
|
|
3097
|
+
"left": 0,
|
|
3098
|
+
"bottom": 0,
|
|
3099
|
+
"right": 0,
|
|
3100
|
+
"top": 0
|
|
3101
|
+
},
|
|
3102
|
+
"bbox<link>": "schema:values:bbox",
|
|
3103
|
+
"label": "",
|
|
3104
|
+
"label<link>": "schema:values:label",
|
|
3105
|
+
"has_fill": "",
|
|
3106
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
3107
|
+
"has_stroke": "",
|
|
3108
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
3109
|
+
"fill_color": "",
|
|
3110
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
3111
|
+
"stroke_color": "",
|
|
3112
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
3113
|
+
"fill_alpha": "",
|
|
3114
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
3115
|
+
"stroke_alpha": "",
|
|
3116
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
3117
|
+
"has_transparency": "",
|
|
3118
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
3119
|
+
"lang": "",
|
|
3120
|
+
"lang<link>": "schema:values:lang",
|
|
3121
|
+
"alt": "",
|
|
3122
|
+
"alt<link>": "schema:values:alt",
|
|
3123
|
+
"actual_text": "",
|
|
3124
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
3125
|
+
"expansion": "",
|
|
3126
|
+
"expansion<link>": "schema:values:expansion",
|
|
3127
|
+
"row_num": "",
|
|
3128
|
+
"row_num<link>": "schema:values:row_num",
|
|
3129
|
+
"col_num": "",
|
|
3130
|
+
"col_num<link>": "schema:values:col_num",
|
|
3131
|
+
"cell_header": false,
|
|
3132
|
+
"cell_header<link>": "schema:values:cell_header",
|
|
3133
|
+
"table_type": "",
|
|
3134
|
+
"table_type<link>": "schema:values:table_type",
|
|
3135
|
+
"isolated_num": "",
|
|
3136
|
+
"isolated_num<desc>": "Number of isolated elements in the table.",
|
|
3137
|
+
"isolated_num<type>": "int",
|
|
3138
|
+
"name": "",
|
|
3139
|
+
"name<link>": "schema:values:name"
|
|
3140
|
+
},
|
|
3141
|
+
"pde_element": {
|
|
3142
|
+
"type": "pde_element",
|
|
3143
|
+
"type<link>": "schema:values:type",
|
|
3144
|
+
"width": "",
|
|
3145
|
+
"width<link>": "schema:values:width",
|
|
3146
|
+
"height": "",
|
|
3147
|
+
"height<link>": "schema:values:height",
|
|
3148
|
+
"left": "",
|
|
3149
|
+
"left<link>": "schema:values:left",
|
|
3150
|
+
"right": "",
|
|
3151
|
+
"right<link>": "schema:values:right",
|
|
3152
|
+
"top": "",
|
|
3153
|
+
"top<link>": "schema:values:top",
|
|
3154
|
+
"bottom": "",
|
|
3155
|
+
"bottom<link>": "schema:values:bottom",
|
|
3156
|
+
"baseline_x": "",
|
|
3157
|
+
"baseline_x<link>": "schema:values:baseline_x",
|
|
3158
|
+
"baseline_y": "",
|
|
3159
|
+
"baseline_y<link>": "schema:values:baseline_y",
|
|
3160
|
+
"bbox": {
|
|
3161
|
+
"left": 0,
|
|
3162
|
+
"bottom": 0,
|
|
3163
|
+
"right": 0,
|
|
3164
|
+
"top": 0
|
|
3165
|
+
},
|
|
3166
|
+
"bbox<link>": "schema:values:bbox",
|
|
3167
|
+
"label": "",
|
|
3168
|
+
"label<link>": "schema:values:label",
|
|
3169
|
+
"has_fill": "",
|
|
3170
|
+
"has_fill<link>": "schema:values:has_fill",
|
|
3171
|
+
"has_stroke": "",
|
|
3172
|
+
"has_stroke<link>": "schema:values:has_stroke",
|
|
3173
|
+
"fill_color": "",
|
|
3174
|
+
"fill_color<link>": "schema:values:fill_color",
|
|
3175
|
+
"stroke_color": "",
|
|
3176
|
+
"stroke_color<link>": "schema:values:stroke_color",
|
|
3177
|
+
"fill_alpha": "",
|
|
3178
|
+
"fill_alpha<link>": "schema:values:fill_alpha",
|
|
3179
|
+
"stroke_alpha": "",
|
|
3180
|
+
"stroke_alpha<link>": "schema:values:stroke_alpha",
|
|
3181
|
+
"has_transparency": "",
|
|
3182
|
+
"has_transparency<link>": "schema:values:has_transparency",
|
|
3183
|
+
"angle": "",
|
|
3184
|
+
"angle<link>": "schema:values:angle",
|
|
3185
|
+
"lang": "",
|
|
3186
|
+
"lang<link>": "schema:values:lang",
|
|
3187
|
+
"alt": "",
|
|
3188
|
+
"alt<link>": "schema:values:alt",
|
|
3189
|
+
"actual_text": "",
|
|
3190
|
+
"actual_text<link>": "schema:values:actual_text",
|
|
3191
|
+
"expansion": "",
|
|
3192
|
+
"expansion<link>": "schema:values:expansion",
|
|
3193
|
+
"flag": "",
|
|
3194
|
+
"flag<link>": "schema:values:flag",
|
|
3195
|
+
"name": "",
|
|
3196
|
+
"name<link>": "schema:values:name"
|
|
3197
|
+
}
|
|
3198
|
+
},
|
|
3199
|
+
"general_vars<desc>": "Built-in read-only variables ($page_*, $doc_*) usable in query conditions without declaring parameters.",
|
|
3200
|
+
"general_vars<use>": "Compare geometry or typography against page averages, or gate rules on anchor names already resolved for the page or document.",
|
|
3201
|
+
"general_vars<type>": "string",
|
|
3202
|
+
"general_vars": {
|
|
3203
|
+
"$page_num": "1",
|
|
3204
|
+
"$page_num<desc>": "1-based index of the page being processed.",
|
|
3205
|
+
"$page_num<use>": "First-page-only or last-page rules when combined with $doc_num_pages.",
|
|
3206
|
+
"$page_num<type>": "int",
|
|
3207
|
+
"$page_width": "",
|
|
3208
|
+
"$page_width<desc>": "Crop box width in default user space units (same basis as layout bboxes).",
|
|
3209
|
+
"$page_width<use>": "Normalize element widths (e.g. $0_width / $page_width) or margin rules.",
|
|
3210
|
+
"$page_width<type>": "float",
|
|
3211
|
+
"$page_height": "",
|
|
3212
|
+
"$page_height<desc>": "Crop box height in default user space units (same basis as layout bboxes).",
|
|
3213
|
+
"$page_height<use>": "Header/footer band tests as a fraction of $page_height.",
|
|
3214
|
+
"$page_height<type>": "float",
|
|
3215
|
+
"$page_font_size": "",
|
|
3216
|
+
"$page_font_size<desc>": "Dominant body font size estimated for the page.",
|
|
3217
|
+
"$page_font_size<use>": "Scale distance thresholds that are stored as font multiples in pagemap.",
|
|
3218
|
+
"$page_font_size<type>": "float",
|
|
3219
|
+
"$page_min_font_size": "",
|
|
3220
|
+
"$page_min_font_size<desc>": "Smallest font size observed on the page.",
|
|
3221
|
+
"$page_min_font_size<use>": "Catch footnotes or legal lines that must be treated differently from body text.",
|
|
3222
|
+
"$page_min_font_size<type>": "float",
|
|
3223
|
+
"$page_max_font_size": "",
|
|
3224
|
+
"$page_max_font_size<desc>": "Largest font size observed on the page.",
|
|
3225
|
+
"$page_max_font_size<use>": "Identify display headings or cover titles by size alone.",
|
|
3226
|
+
"$page_max_font_size<type>": "float",
|
|
3227
|
+
"$page_rotation": "",
|
|
3228
|
+
"$page_rotation<desc>": "PDF /Rotate value in degrees for the page (typically 0, 90, 180, or 270).",
|
|
3229
|
+
"$page_rotation<use>": "Branch rules for landscape vs portrait or skip geometry heuristics that assume rotation 0.",
|
|
3230
|
+
"$page_rotation<type>": "int",
|
|
3231
|
+
"$page_rotation<value>": [
|
|
3232
|
+
0,
|
|
3233
|
+
90,
|
|
3234
|
+
180,
|
|
3235
|
+
270
|
|
3236
|
+
],
|
|
3237
|
+
"$page_rtl": false,
|
|
3238
|
+
"$page_rtl<desc>": "True when the page is treated as right-to-left for layout. Set during page-content parsing when the knowledge base enables RTL handling and a text object contains a character classified as right-to-left; if RTL is disabled in the knowledge base, this flag is usually not set that way and stays false.",
|
|
3239
|
+
"$page_rtl<use>": "Branch template rules on page RTL state; pair with KB/settings RTL if you rely on parse-time detection rather than document-wide defaults alone.",
|
|
3240
|
+
"$page_rtl<type>": "bool",
|
|
3241
|
+
"$page_rtl<value>": [
|
|
3242
|
+
"true",
|
|
3243
|
+
"false"
|
|
3244
|
+
],
|
|
3245
|
+
"$page_anchor": "",
|
|
3246
|
+
"$page_anchor<desc>": "Comma-joined names of anchors already registered on this page during template application.",
|
|
3247
|
+
"$page_anchor<use>": "Test substring or regex membership to skip duplicate rules or target follow-up logic after an anchor fired.",
|
|
3248
|
+
"$page_anchor<type>": "string",
|
|
3249
|
+
"$doc_num_pages": "",
|
|
3250
|
+
"$doc_num_pages<desc>": "Total page count in the PDF document.",
|
|
3251
|
+
"$doc_num_pages<use>": "Enable rules only on single-page vs multi-page jobs.",
|
|
3252
|
+
"$doc_num_pages<type>": "int",
|
|
3253
|
+
"$doc_lang": "",
|
|
3254
|
+
"$doc_lang<desc>": "Document-level language hint from metadata or detection.",
|
|
3255
|
+
"$doc_lang<use>": "Pick regex tables or tag defaults when the pack is bilingual.",
|
|
3256
|
+
"$doc_lang<type>": "string",
|
|
3257
|
+
"$doc_title": "",
|
|
3258
|
+
"$doc_title<desc>": "Title string from document metadata when available.",
|
|
3259
|
+
"$doc_title<use>": "Match cover or running metadata in rare conditional templates.",
|
|
3260
|
+
"$doc_title<type>": "string",
|
|
3261
|
+
"$doc_anchor": "",
|
|
3262
|
+
"$doc_anchor<desc>": "Comma-joined anchor names resolved anywhere in the document so far.",
|
|
3263
|
+
"$doc_anchor<use>": "Apply doc-wide follow-up rules only after a named anchor (e.g. title block) was seen.",
|
|
3264
|
+
"$doc_anchor<type>": "string"
|
|
3265
|
+
},
|
|
3266
|
+
"values<desc>": "Reusable field definitions (alt, flag, bbox, …) linked from template nodes via schema:values:*.",
|
|
3267
|
+
"values<use>": "Each entry documents type, meaning, and allowed flag or enum strings for queries and *_update payloads.",
|
|
3268
|
+
"values": {
|
|
3269
|
+
"type": "",
|
|
3270
|
+
"type<desc>": "Discriminator string for the object or element in a query param slot.",
|
|
3271
|
+
"type<type>": "string",
|
|
3272
|
+
"type<value>": [
|
|
3273
|
+
"pds_object",
|
|
3274
|
+
"pds_text",
|
|
3275
|
+
"pds_path",
|
|
3276
|
+
"pds_image",
|
|
3277
|
+
"pds_shading",
|
|
3278
|
+
"pds_form",
|
|
3279
|
+
"pde_element",
|
|
3280
|
+
"pde_text",
|
|
3281
|
+
"pde_text_line",
|
|
3282
|
+
"pde_word",
|
|
3283
|
+
"pde_text_run",
|
|
3284
|
+
"pde_image",
|
|
3285
|
+
"pde_container",
|
|
3286
|
+
"pde_list",
|
|
3287
|
+
"pde_line",
|
|
3288
|
+
"pde_rect",
|
|
3289
|
+
"pde_table",
|
|
3290
|
+
"pde_cell",
|
|
3291
|
+
"pde_toc",
|
|
3292
|
+
"pde_header",
|
|
3293
|
+
"pde_footer",
|
|
3294
|
+
"pde_form_field",
|
|
3295
|
+
"pde_annot",
|
|
3296
|
+
"pds_struct_elem",
|
|
3297
|
+
"pdf_annot"
|
|
3298
|
+
],
|
|
3299
|
+
"alt": "",
|
|
3300
|
+
"alt<desc>": "Alternate description for figures, images, and other structure elements (PDF/UA).",
|
|
3301
|
+
"alt<use>": "Required for meaningful Figure tags; can be filled by element_prop_update if omitted in the PDF.",
|
|
3302
|
+
"alt<type>": "string",
|
|
3303
|
+
"actual_text": "",
|
|
3304
|
+
"actual_text<desc>": "Replacement Unicode string for structure or content when true text is not encoded as glyphs (or for accessibility overrides).",
|
|
3305
|
+
"actual_text<use>": "Set on images, artifacts, or ligature-heavy lines where Alt alone is insufficient.",
|
|
3306
|
+
"actual_text<type>": "string",
|
|
3307
|
+
"expansion": "",
|
|
3308
|
+
"expansion<desc>": "Spoken or spelled-out form of an abbreviation or acronym (E for expanded form in tagged PDF).",
|
|
3309
|
+
"expansion<use>": "Pair with short headings or metrics in body text so assistive tech can read the long form once.",
|
|
3310
|
+
"expansion<type>": "string",
|
|
3311
|
+
"lang": "",
|
|
3312
|
+
"lang<desc>": "BCP 47 / ISO language tag applied to the element (e.g. en-US).",
|
|
3313
|
+
"lang<use>": "Set on text containers for screen-reader language switching.",
|
|
3314
|
+
"lang<type>": "string",
|
|
3315
|
+
"id": "",
|
|
3316
|
+
"id<desc>": "Stable structure-element identifier string carried into the tag tree when set from the template.",
|
|
3317
|
+
"id<use>": "Correlate output with external CMS rows or cross-reference captions and figures.",
|
|
3318
|
+
"id<type>": "string",
|
|
3319
|
+
"associated_header": "",
|
|
3320
|
+
"associated_header<desc>": "Header cell id(s) this data cell relates to; use | to list multiple associations.",
|
|
3321
|
+
"associated_header<use>": "PDF/UA: link TD cells to TH ids when automatic scope detection is ambiguous.",
|
|
3322
|
+
"associated_header<type>": "string",
|
|
3323
|
+
"has_content": "",
|
|
3324
|
+
"has_content<desc>": "True when the structure node has real page-marked content (text, paths, images) under it in the tag tree—not counting OBJR-only annotation references.",
|
|
3325
|
+
"has_content<use>": "Distinguish empty wrapper tags from leaf nodes with MCID-backed content.",
|
|
3326
|
+
"has_content<type>": "bool",
|
|
3327
|
+
"has_content<value>": [
|
|
3328
|
+
"true",
|
|
3329
|
+
"false"
|
|
3330
|
+
],
|
|
3331
|
+
"has_annot": "",
|
|
3332
|
+
"has_annot<desc>": "True when the structure element references annotations via OBJR entries (links, widgets, markup) even if no direct text path is attached.",
|
|
3333
|
+
"has_annot<use>": "Target Link or Form structure without relying on has_content alone.",
|
|
3334
|
+
"has_annot<type>": "bool",
|
|
3335
|
+
"has_annot<value>": [
|
|
3336
|
+
"true",
|
|
3337
|
+
"false"
|
|
3338
|
+
],
|
|
3339
|
+
"tag_type": "",
|
|
3340
|
+
"tag_type<desc>": "PDF structure type name or regex (e.g. Figure, P, .*).",
|
|
3341
|
+
"tag_type<use>": "Filter tag_update or tagging pipelines to specific roles.",
|
|
3342
|
+
"tag_type<type>": "string",
|
|
3343
|
+
"tag_type<value>": [
|
|
3344
|
+
"Annot",
|
|
3345
|
+
"Art",
|
|
3346
|
+
"Artifact",
|
|
3347
|
+
"Aside",
|
|
3348
|
+
"BibEntry",
|
|
3349
|
+
"BlockQuote",
|
|
3350
|
+
"Caption",
|
|
3351
|
+
"Code",
|
|
3352
|
+
"Div",
|
|
3353
|
+
"Document",
|
|
3354
|
+
"DocumentFragment",
|
|
3355
|
+
"Em",
|
|
3356
|
+
"FENote",
|
|
3357
|
+
"Figure",
|
|
3358
|
+
"Form",
|
|
3359
|
+
"Formula",
|
|
3360
|
+
"H",
|
|
3361
|
+
"H1",
|
|
3362
|
+
"H2",
|
|
3363
|
+
"H3",
|
|
3364
|
+
"H4",
|
|
3365
|
+
"H5",
|
|
3366
|
+
"H6",
|
|
3367
|
+
"Index",
|
|
3368
|
+
"L",
|
|
3369
|
+
"Lbl",
|
|
3370
|
+
"LBody",
|
|
3371
|
+
"LI",
|
|
3372
|
+
"Link",
|
|
3373
|
+
"NonStruct",
|
|
3374
|
+
"Note",
|
|
3375
|
+
"P",
|
|
3376
|
+
"Part",
|
|
3377
|
+
"Private",
|
|
3378
|
+
"Quote",
|
|
3379
|
+
"RB",
|
|
3380
|
+
"Reference",
|
|
3381
|
+
"RP",
|
|
3382
|
+
"RT",
|
|
3383
|
+
"Ruby",
|
|
3384
|
+
"Sect",
|
|
3385
|
+
"Span",
|
|
3386
|
+
"Strong",
|
|
3387
|
+
"Sub",
|
|
3388
|
+
"Table",
|
|
3389
|
+
"TBody",
|
|
3390
|
+
"TD",
|
|
3391
|
+
"TFoot",
|
|
3392
|
+
"TH",
|
|
3393
|
+
"THead",
|
|
3394
|
+
"Title",
|
|
3395
|
+
"TOC",
|
|
3396
|
+
"TOCI",
|
|
3397
|
+
"TR",
|
|
3398
|
+
"Warichu",
|
|
3399
|
+
"WP",
|
|
3400
|
+
"WT"
|
|
3401
|
+
],
|
|
3402
|
+
"parent_tag_type": "",
|
|
3403
|
+
"parent_tag_type<desc>": "Same vocabulary as tag_type: role name or regex for the parent structure node in tag_update queries.",
|
|
3404
|
+
"parent_tag_type<use>": "Match only when the immediate ancestor is e.g. Link, Table, or Sect.",
|
|
3405
|
+
"parent_tag_type<link>": "schema:values:tag_type",
|
|
3406
|
+
"children_tag_type": "",
|
|
3407
|
+
"children_tag_type<desc>": "Pipe-separated direct child types. Struct element children use tag_type role names (P, Span, Link, …). OBJR annotation children use the annotation subtype with -OBJR suffix (Link-OBJR, Widget-OBJR, Highlight-OBJR, …). Use $eq on the full string or $regex to match one segment.",
|
|
3408
|
+
"children_tag_type<use>": "Require that a container wraps specific child roles or annotation OBJR entries, e.g. Span|Link-OBJR on a Link tag.",
|
|
3409
|
+
"children_tag_type<link>": "schema:values:tag_type",
|
|
3410
|
+
"annot_type": "",
|
|
3411
|
+
"annot_type<desc>": "PDF annotation subtype name (Link, Widget, Highlight, …) or regex; use .* to match all.",
|
|
3412
|
+
"annot_type<use>": "Narrow annot_update or tagging rules to links vs form fields vs markup.",
|
|
3413
|
+
"annot_type<type>": "string",
|
|
3414
|
+
"annot_type<value>": [
|
|
3415
|
+
"Text",
|
|
3416
|
+
"Link",
|
|
3417
|
+
"FreeText",
|
|
3418
|
+
"Line",
|
|
3419
|
+
"Square",
|
|
3420
|
+
"Circle",
|
|
3421
|
+
"Polygon",
|
|
3422
|
+
"PolyLine",
|
|
3423
|
+
"Highlight",
|
|
3424
|
+
"Underline",
|
|
3425
|
+
"Squiggly",
|
|
3426
|
+
"StrikeOut",
|
|
3427
|
+
"Stamp",
|
|
3428
|
+
"Caret",
|
|
3429
|
+
"Ink",
|
|
3430
|
+
"Popup",
|
|
3431
|
+
"FileAttachment",
|
|
3432
|
+
"Sound",
|
|
3433
|
+
"Movie",
|
|
3434
|
+
"Widget",
|
|
3435
|
+
"Screen",
|
|
3436
|
+
"PrinterMark",
|
|
3437
|
+
"TrapNet",
|
|
3438
|
+
"Watermark",
|
|
3439
|
+
"3D",
|
|
3440
|
+
"Redact",
|
|
3441
|
+
"Projection",
|
|
3442
|
+
"RichMedia"
|
|
3443
|
+
],
|
|
3444
|
+
"contents": "",
|
|
3445
|
+
"contents<desc>": "Human-readable string written to the annotation Contents entry, or TU for widget tooltips when mapping to accessibility.",
|
|
3446
|
+
"contents<use>": "Set via annot_update when the PDF lacks a proper tooltip or link description.",
|
|
3447
|
+
"contents<type>": "string",
|
|
3448
|
+
"annot_flag": 0,
|
|
3449
|
+
"annot_flag<desc>": "Comma-separated list of PDF annotation flag keywords (invisible, print, read_only, …). The token contents refers to the flag bit, not this string property.",
|
|
3450
|
+
"annot_flag<use>": "Filter out hidden or no-view annotations, or select print-only stamps.",
|
|
3451
|
+
"annot_flag<type>": "string",
|
|
3452
|
+
"annot_flag<value>": [
|
|
3453
|
+
"invisible",
|
|
3454
|
+
"hidden",
|
|
3455
|
+
"print",
|
|
3456
|
+
"no_zoom",
|
|
3457
|
+
"no_rotate",
|
|
3458
|
+
"no_view",
|
|
3459
|
+
"read_only",
|
|
3460
|
+
"locked",
|
|
3461
|
+
"toggle",
|
|
3462
|
+
"contents"
|
|
3463
|
+
],
|
|
3464
|
+
"title": "",
|
|
3465
|
+
"title<desc>": "Human-readable title string on annotations or widgets when exposed by the PDF.",
|
|
3466
|
+
"title<type>": "string",
|
|
3467
|
+
"name": "",
|
|
3468
|
+
"name<desc>": "Template element `name` for this element (letters, digits, hyphen; no $ or _).",
|
|
3469
|
+
"name<use>": "Referenced by parent, variables like $NAME_left, and single_instance deduplication keys.",
|
|
3470
|
+
"name<type>": "string",
|
|
3471
|
+
"parent": "",
|
|
3472
|
+
"parent<desc>": "Optional parent override by element `name`; changes default element nesting.",
|
|
3473
|
+
"parent<use>": "Use when you need to re-parent an element and that parenting was not already defined by the initial-element tree.",
|
|
3474
|
+
"parent<type>": "string",
|
|
3475
|
+
"caption": "",
|
|
3476
|
+
"caption<desc>": "Element `name` of an element that should be attached as this element's caption/title child.",
|
|
3477
|
+
"caption<use>": "Set when a specific element must be inserted during tagging as the first child (caption-first/title behavior).",
|
|
3478
|
+
"caption<type>": "string",
|
|
3479
|
+
"angle": "",
|
|
3480
|
+
"angle<desc>": "Rotation of the text or element in radians (positive = counter-clockwise in user space).",
|
|
3481
|
+
"angle<type>": "float",
|
|
3482
|
+
"bbox<desc>": "Axis-aligned bounds in default user space: left, bottom, right, top. Coordinates may be literals, $page_* macros, anchor offsets, or simple math. One bbox ties to one anchor for variable expansion.",
|
|
3483
|
+
"bbox<type>": "bbox",
|
|
3484
|
+
"bbox": {
|
|
3485
|
+
"left": 0,
|
|
3486
|
+
"bottom": 0,
|
|
3487
|
+
"right": 0,
|
|
3488
|
+
"top": 0
|
|
3489
|
+
},
|
|
3490
|
+
"cell_column": "",
|
|
3491
|
+
"cell_column<desc>": "The specific column index (1-based) of the cell within the table.",
|
|
3492
|
+
"cell_column<use>": "Use this in queries to target specific columns. For example, to mark all cells in column 1 as row headers.",
|
|
3493
|
+
"cell_column<type>": "int",
|
|
3494
|
+
"cell_row": "",
|
|
3495
|
+
"cell_row<desc>": "The specific row index (1-based) of the cell within the table.",
|
|
3496
|
+
"cell_row<use>": "Use this in queries to target specific rows. For example, 'If cell_row is 1, set as Header'.",
|
|
3497
|
+
"cell_row<type>": "int",
|
|
3498
|
+
"cell_row_span": "",
|
|
3499
|
+
"cell_row_span<desc>": "The number of vertical rows this cell merges or covers.",
|
|
3500
|
+
"cell_row_span<use>": "Important for complex tables. Use to identify or set cells that cover multiple data points vertically.",
|
|
3501
|
+
"cell_row_span<type>": "int",
|
|
3502
|
+
"cell_column_span": "",
|
|
3503
|
+
"cell_column_span<desc>": "The number of horizontal columns this cell merges or covers.",
|
|
3504
|
+
"cell_column_span<use>": "Commonly used for top-level headers that group multiple sub-columns (e.g., 'Q1 Results' spanning 'Jan', 'Feb', 'Mar').",
|
|
3505
|
+
"cell_column_span<type>": "int",
|
|
3506
|
+
"cell_scope": "column",
|
|
3507
|
+
"cell_scope<desc>": "Defines the accessibility scope of a header cell (Row, Column, or Both).",
|
|
3508
|
+
"cell_scope<use>": "Crucial for PDF/UA compliance. Set to 'Column' for top headers and 'Row' for side headers so screen readers associate data correctly.",
|
|
3509
|
+
"cell_scope<type>": "string",
|
|
3510
|
+
"cell_scope<value>": [
|
|
3511
|
+
"row",
|
|
3512
|
+
"column",
|
|
3513
|
+
"both"
|
|
3514
|
+
],
|
|
3515
|
+
"col_num": "",
|
|
3516
|
+
"col_num<desc>": "Logical column count of the parent pde_table after recognition (grid width, not visual guides).",
|
|
3517
|
+
"col_num<use>": "Target narrow vs wide tables (e.g. only style two-column grids) or skip rules on single-column layouts.",
|
|
3518
|
+
"col_num<type>": "int",
|
|
3519
|
+
"children_num": "",
|
|
3520
|
+
"children_num<desc>": "Direct child count: layout children for pde_*, or tag-tree children including OBJR for structure nodes.",
|
|
3521
|
+
"children_num<use>": "Detect empty wrappers vs populated containers; trim leaf tags with zero children in tag_update.",
|
|
3522
|
+
"children_num<type>": "int",
|
|
3523
|
+
"object_num": "",
|
|
3524
|
+
"object_num<desc>": "Number of low-level PDF objects (paths, text runs, images) merged into this layout element.",
|
|
3525
|
+
"object_num<use>": "Heuristics for widgets (e.g. checkbox: one path empty, multiple paths when checked) or icon complexity.",
|
|
3526
|
+
"object_num<type>": "int",
|
|
3527
|
+
"artifact": "",
|
|
3528
|
+
"artifact<desc>": "Indicates if the page content object is explicitly marked as an Artifact via a Marked Content Sequence (BMC/BDC tag with /Artifact property).",
|
|
3529
|
+
"artifact<use>": "Use to filter out content that is already defined as background/decorative at the PDF syntax level. If true, the object should typically be ignored by layout analysis.",
|
|
3530
|
+
"artifact<type>": "bool",
|
|
3531
|
+
"artifact<value>": [
|
|
3532
|
+
"true",
|
|
3533
|
+
"false"
|
|
3534
|
+
],
|
|
3535
|
+
"mcid": "",
|
|
3536
|
+
"mcid<desc>": "The Marked Content Identifier (MCID) assigned to this content stream object.",
|
|
3537
|
+
"mcid<use>": "Crucial for Tagged PDF (PDF/UA). A value >= 0 links the content to the tag tree. A value of -1 indicates the content is untagged or not properly associated.",
|
|
3538
|
+
"mcid<type>": "int",
|
|
3539
|
+
"has_fill": false,
|
|
3540
|
+
"has_fill<desc>": "Indicates whether the object has a fill color applied.",
|
|
3541
|
+
"has_fill<use>": "Use as a safeguard before checking 'fill_color'. Ensures you are evaluating color only on objects that actually have a fill, preventing logical errors on transparent or unfilled shapes.",
|
|
3542
|
+
"has_fill<type>": "bool",
|
|
3543
|
+
"has_fill<value>": [
|
|
3544
|
+
"true",
|
|
3545
|
+
"false"
|
|
3546
|
+
],
|
|
3547
|
+
"fill_color<desc>": "The RGB components of the object's fill (interior) color.",
|
|
3548
|
+
"fill_color<use>": "Color-based detection (e.g. shaded table headers). Prefer checking has_fill first unless the content model guarantees a fill.",
|
|
3549
|
+
"fill_color<type>": "rgb",
|
|
3550
|
+
"fill_color": {
|
|
3551
|
+
"red": 0,
|
|
3552
|
+
"green": 0,
|
|
3553
|
+
"blue": 0
|
|
3554
|
+
},
|
|
3555
|
+
"has_stroke": false,
|
|
3556
|
+
"has_stroke<desc>": "Indicates whether the object has a stroke (border/outline) applied.",
|
|
3557
|
+
"has_stroke<use>": "Use as a safeguard before checking 'stroke_color'. Ensures you are evaluating color only on objects that actually have a stroke, avoiding errors on borderless shapes.",
|
|
3558
|
+
"has_stroke<type>": "bool",
|
|
3559
|
+
"has_stroke<value>": [
|
|
3560
|
+
"true",
|
|
3561
|
+
"false"
|
|
3562
|
+
],
|
|
3563
|
+
"stroke_color<desc>": "The RGB components of the object's stroke (border) color.",
|
|
3564
|
+
"stroke_color<use>": "Defines the target border color values. Best practice is to evaluate this only when 'has_stroke' is true, or when you are certain the object possesses a stroke.",
|
|
3565
|
+
"stroke_color<type>": "rgb",
|
|
3566
|
+
"stroke_color": {
|
|
3567
|
+
"red": 0,
|
|
3568
|
+
"green": 0,
|
|
3569
|
+
"blue": 0
|
|
3570
|
+
},
|
|
3571
|
+
"fill_alpha": "",
|
|
3572
|
+
"fill_alpha<desc>": "The opacity level of the fill color, ranging from 0 (invisible) to 255 (fully opaque).",
|
|
3573
|
+
"fill_alpha<use>": "Use to detect semi-transparent elements like watermarks or highlights (e.g., if fill_alpha < 255, mark as Artifact).",
|
|
3574
|
+
"fill_alpha<type>": "int",
|
|
3575
|
+
"stroke_alpha": "",
|
|
3576
|
+
"stroke_alpha<desc>": "The opacity level of the stroke color, ranging from 0 (invisible) to 255 (fully opaque).",
|
|
3577
|
+
"stroke_alpha<use>": "Use to detect faint or transparent lines that should be ignored or treated as background decoration.",
|
|
3578
|
+
"stroke_alpha<type>": "int",
|
|
3579
|
+
"has_transparency": false,
|
|
3580
|
+
"has_transparency<desc>": "Indicates if the element or object contains any transparency features.",
|
|
3581
|
+
"has_transparency<use>": "Primarily used to detect Form XObjects that contain transparency. This helps determine if an XObject can be safely flattened or must be preserved to maintain correct visual rendering.",
|
|
3582
|
+
"has_transparency<type>": "bool",
|
|
3583
|
+
"has_transparency<value>": [
|
|
3584
|
+
"true",
|
|
3585
|
+
"false"
|
|
3586
|
+
],
|
|
3587
|
+
"flag": "",
|
|
3588
|
+
"flag<desc>": "Template-driven behavior on an element: for example no_join and no_split control merging; continuous keeps one structure tag across pages for tables, lists, or text when tagging; splitter ends continuous chains for the types listed in splitter; artifact, header, and footer move content out of the main reading flow.",
|
|
3589
|
+
"flag<use>": "Combine with splitter type list to end multi-page tables; use continuous only when tag output must span pages without a new Table/L root.",
|
|
3590
|
+
"flag<type>": "string",
|
|
3591
|
+
"flag<value>": [
|
|
3592
|
+
"none",
|
|
3593
|
+
"no_join",
|
|
3594
|
+
"no_split",
|
|
3595
|
+
"artifact",
|
|
3596
|
+
"header",
|
|
3597
|
+
"footer",
|
|
3598
|
+
"splitter",
|
|
3599
|
+
"no_table",
|
|
3600
|
+
"no_image",
|
|
3601
|
+
"no_expand",
|
|
3602
|
+
"continuous",
|
|
3603
|
+
"anchor"
|
|
3604
|
+
],
|
|
3605
|
+
"numbering": "None",
|
|
3606
|
+
"numbering<desc>": "List numbering style for the element (None, Unordered, Decimal, UpperRoman, LowerRoman, and other schema values). Drives list attributes and ordered-marker continuation when ordinals are detected.",
|
|
3607
|
+
"numbering<use>": "Force Decimal vs Unordered on list_update, or normalize markers that were misclassified.",
|
|
3608
|
+
"numbering<type>": "string",
|
|
3609
|
+
"numbering<value>": [
|
|
3610
|
+
"None",
|
|
3611
|
+
"Unordered",
|
|
3612
|
+
"Disc",
|
|
3613
|
+
"Circle",
|
|
3614
|
+
"Square",
|
|
3615
|
+
"Ordered",
|
|
3616
|
+
"Decimal",
|
|
3617
|
+
"UpperRoman",
|
|
3618
|
+
"LowerRoman",
|
|
3619
|
+
"UpperAlpha",
|
|
3620
|
+
"LowerAlpha",
|
|
3621
|
+
"Description"
|
|
3622
|
+
],
|
|
3623
|
+
"single_instance": "font_size|font_name|text|left",
|
|
3624
|
+
"single_instance<desc>": "Pipe-separated property keys; duplicates sharing those keys collapse so only the first instance is kept for tagging or layout.",
|
|
3625
|
+
"single_instance<use>": "Dedupe repeated logos, headers, or boilerplate that share font, bbox, or text.",
|
|
3626
|
+
"single_instance<type>": "string",
|
|
3627
|
+
"single_instance<value>": [
|
|
3628
|
+
"type",
|
|
3629
|
+
"width",
|
|
3630
|
+
"height",
|
|
3631
|
+
"left",
|
|
3632
|
+
"right",
|
|
3633
|
+
"top",
|
|
3634
|
+
"bottom",
|
|
3635
|
+
"bbox",
|
|
3636
|
+
"font_size",
|
|
3637
|
+
"font_name",
|
|
3638
|
+
"text",
|
|
3639
|
+
"fill_color",
|
|
3640
|
+
"stroke_color",
|
|
3641
|
+
"angle",
|
|
3642
|
+
"alt",
|
|
3643
|
+
"actual_text",
|
|
3644
|
+
"flag",
|
|
3645
|
+
"word_flag",
|
|
3646
|
+
"text_line_flag",
|
|
3647
|
+
"text_flag",
|
|
3648
|
+
"lang",
|
|
3649
|
+
"cell_column",
|
|
3650
|
+
"cell_row",
|
|
3651
|
+
"cell_column_span",
|
|
3652
|
+
"cell_row_span",
|
|
3653
|
+
"cell_scope",
|
|
3654
|
+
"row_num",
|
|
3655
|
+
"col_num"
|
|
3656
|
+
],
|
|
3657
|
+
"word_space": 0,
|
|
3658
|
+
"word_space<desc>": "When word_update matches and sets this float, the engine stores it as that font’s fixed space width before lines are built, so later layout uses that spacing for the font.",
|
|
3659
|
+
"word_space<use>": "Set on a representative word so the whole font’s spacing shifts; pagemap word_space_ratio still multiplies those KB-locked widths afterward.",
|
|
3660
|
+
"word_space<type>": "float",
|
|
3661
|
+
"font_name": "",
|
|
3662
|
+
"font_name<desc>": "PDF font base name for the text object or run.",
|
|
3663
|
+
"font_name<type>": "string",
|
|
3664
|
+
"cos_obj": "",
|
|
3665
|
+
"cos_obj<desc>": "Indirect object number of the object dictionary in the PDF file.",
|
|
3666
|
+
"cos_obj<use>": "Match or filter objects by COS object.",
|
|
3667
|
+
"cos_obj<type>": "int",
|
|
3668
|
+
"font_type": "",
|
|
3669
|
+
"font_type<desc>": "PDF font type from GetFontState (Type1, TrueType, Type3, CIDFont, Unknown).",
|
|
3670
|
+
"font_type<use>": "Filter font_update rules by font program type without parsing BaseFont names.",
|
|
3671
|
+
"font_type<type>": "string",
|
|
3672
|
+
"font_type<value>": [
|
|
3673
|
+
"Unknown",
|
|
3674
|
+
"Type1",
|
|
3675
|
+
"TrueType",
|
|
3676
|
+
"Type3",
|
|
3677
|
+
"CIDFont"
|
|
3678
|
+
],
|
|
3679
|
+
"embedded": "",
|
|
3680
|
+
"embedded<desc>": "True when the font program is embedded in the PDF.",
|
|
3681
|
+
"embedded<use>": "Target non-embedded fonts for subsetting or replacement in font_update.",
|
|
3682
|
+
"embedded<type>": "bool",
|
|
3683
|
+
"embedded<value>": [
|
|
3684
|
+
"true",
|
|
3685
|
+
"false"
|
|
3686
|
+
],
|
|
3687
|
+
"bold": "",
|
|
3688
|
+
"bold<desc>": "True when the font is treated as bold (font metrics or BaseFont name heuristic).",
|
|
3689
|
+
"bold<use>": "Classify font families in font_update; not the same as system substitution bold.",
|
|
3690
|
+
"bold<type>": "bool",
|
|
3691
|
+
"bold<value>": [
|
|
3692
|
+
"true",
|
|
3693
|
+
"false"
|
|
3694
|
+
],
|
|
3695
|
+
"italic": "",
|
|
3696
|
+
"italic<desc>": "True when the font has a non-zero italic angle.",
|
|
3697
|
+
"italic<use>": "Filter oblique or italic faces in font_update.",
|
|
3698
|
+
"italic<type>": "bool",
|
|
3699
|
+
"italic<value>": [
|
|
3700
|
+
"true",
|
|
3701
|
+
"false"
|
|
3702
|
+
],
|
|
3703
|
+
"fixed_width": "",
|
|
3704
|
+
"fixed_width<desc>": "True for fixed-pitch (monospace) fonts.",
|
|
3705
|
+
"fixed_width<use>": "Match code or table fonts in font_update.",
|
|
3706
|
+
"fixed_width<type>": "bool",
|
|
3707
|
+
"fixed_width<value>": [
|
|
3708
|
+
"true",
|
|
3709
|
+
"false"
|
|
3710
|
+
],
|
|
3711
|
+
"vertical": "",
|
|
3712
|
+
"vertical<desc>": "True when the font uses vertical writing mode.",
|
|
3713
|
+
"vertical<use>": "Separate CJK vertical fonts in font_update.",
|
|
3714
|
+
"vertical<type>": "bool",
|
|
3715
|
+
"vertical<value>": [
|
|
3716
|
+
"true",
|
|
3717
|
+
"false"
|
|
3718
|
+
],
|
|
3719
|
+
"missing_unicode": "",
|
|
3720
|
+
"missing_unicode<desc>": "True when the font has charcodes without a Unicode mapping.",
|
|
3721
|
+
"missing_unicode<use>": "Select fonts that need Unicode repair in font commands.",
|
|
3722
|
+
"missing_unicode<type>": "bool",
|
|
3723
|
+
"missing_unicode<value>": [
|
|
3724
|
+
"true",
|
|
3725
|
+
"false"
|
|
3726
|
+
],
|
|
3727
|
+
"font_size": "",
|
|
3728
|
+
"font_size<desc>": "Font size in points (text height scale) for the object or run.",
|
|
3729
|
+
"font_size<type>": "float",
|
|
3730
|
+
"red": 0,
|
|
3731
|
+
"red<desc>": "The red component of an RGB color.",
|
|
3732
|
+
"red<type>": "int",
|
|
3733
|
+
"green": 0,
|
|
3734
|
+
"green<desc>": "The green component of an RGB color.",
|
|
3735
|
+
"green<type>": "int",
|
|
3736
|
+
"blue": 0,
|
|
3737
|
+
"blue<desc>": "The blue component of an RGB color.",
|
|
3738
|
+
"blue<type>": "int",
|
|
3739
|
+
"cell_header": false,
|
|
3740
|
+
"cell_header<desc>": "True when this table cell should be emitted or treated as a header (TH) for accessibility.",
|
|
3741
|
+
"cell_header<use>": "Set from cell_update when detection missed a shaded header row, or clear false for false positives.",
|
|
3742
|
+
"cell_header<type>": "bool",
|
|
3743
|
+
"cell_header<value>": [
|
|
3744
|
+
"true",
|
|
3745
|
+
"false"
|
|
3746
|
+
],
|
|
3747
|
+
"cell_associated_header": "",
|
|
3748
|
+
"cell_associated_header<desc>": "Pipe-separated list of header cell ids associated with this cell (see associated_header).",
|
|
3749
|
+
"cell_associated_header<use>": "Mirrors associated_header at cell scope for complex TH/TD relationships in cell_update.",
|
|
3750
|
+
"cell_associated_header<type>": "string",
|
|
3751
|
+
"heading": "",
|
|
3752
|
+
"heading<desc>": "Logical heading level for structure output (normal, h1–h6, h, note, title).",
|
|
3753
|
+
"heading<use>": "Map visual styles to PDF/UA heading roles without relying on font size alone.",
|
|
3754
|
+
"heading<type>": "string",
|
|
3755
|
+
"heading<value>": [
|
|
3756
|
+
"normal",
|
|
3757
|
+
"h1",
|
|
3758
|
+
"h2",
|
|
3759
|
+
"h3",
|
|
3760
|
+
"h4",
|
|
3761
|
+
"h5",
|
|
3762
|
+
"h6",
|
|
3763
|
+
"h",
|
|
3764
|
+
"note",
|
|
3765
|
+
"title"
|
|
3766
|
+
],
|
|
3767
|
+
"width": "",
|
|
3768
|
+
"width<desc>": "Width of the bbox in default user space units (points).",
|
|
3769
|
+
"width<use>": "Compare to page width for full-bleed blocks, or scale thresholds with $page_width.",
|
|
3770
|
+
"width<type>": "float",
|
|
3771
|
+
"height": "",
|
|
3772
|
+
"height<desc>": "Height of the bbox in default user space units (points).",
|
|
3773
|
+
"height<use>": "Detect thin separator lines vs tall text blocks using height ratios.",
|
|
3774
|
+
"height<type>": "float",
|
|
3775
|
+
"label": "",
|
|
3776
|
+
"label<desc>": "List-marker classification: label_no (not a marker), label (generic), label_list (list body), or li_1–li_6 for outline depth used by list_detect.",
|
|
3777
|
+
"label<use>": "On an initial pde_list, pinning fixes the level for every item; elsewhere use label_update to correct mis-leveled markers before list_detect.",
|
|
3778
|
+
"label<type>": "string",
|
|
3779
|
+
"label<value>": [
|
|
3780
|
+
"label_no",
|
|
3781
|
+
"label",
|
|
3782
|
+
"label_list",
|
|
3783
|
+
"li_1",
|
|
3784
|
+
"li_2",
|
|
3785
|
+
"li_3",
|
|
3786
|
+
"li_4",
|
|
3787
|
+
"li_5",
|
|
3788
|
+
"li_6"
|
|
3789
|
+
],
|
|
3790
|
+
"label_text": "",
|
|
3791
|
+
"label_text<desc>": "String or regex for the visible marker token (digits, bullets, roman numerals) consistent with pagemap_regex numbering and bullet patterns.",
|
|
3792
|
+
"label_text<use>": "Pin on initial pde_list so every item shares one marker pattern; or narrow list_update to custom bullet glyphs not in default regex tables.",
|
|
3793
|
+
"label_text<type>": "string",
|
|
3794
|
+
"left": "",
|
|
3795
|
+
"left<desc>": "Left edge x of the bbox in default user space.",
|
|
3796
|
+
"left<use>": "Compare to $page_width or anchor left for column and margin rules.",
|
|
3797
|
+
"left<type>": "float",
|
|
3798
|
+
"right": "",
|
|
3799
|
+
"right<desc>": "Right edge x of the bbox in default user space.",
|
|
3800
|
+
"right<use>": "Test alignment to page margin or table width.",
|
|
3801
|
+
"right<type>": "float",
|
|
3802
|
+
"top": "",
|
|
3803
|
+
"top<desc>": "Top edge y of the bbox in default user space.",
|
|
3804
|
+
"top<use>": "Vertical bands (headers) relative to $page_height.",
|
|
3805
|
+
"top<type>": "float",
|
|
3806
|
+
"bottom": "",
|
|
3807
|
+
"bottom<desc>": "Bottom edge y of the bbox in default user space.",
|
|
3808
|
+
"bottom<use>": "Footer detection or minimum distance to page bottom.",
|
|
3809
|
+
"bottom<type>": "float",
|
|
3810
|
+
"baseline_x": "",
|
|
3811
|
+
"baseline_x<desc>": "Horizontal position of the text baseline origin for this text object.",
|
|
3812
|
+
"baseline_x<use>": "Align captions, rules, or anchors to typographic baseline instead of bbox edges.",
|
|
3813
|
+
"baseline_x<type>": "float",
|
|
3814
|
+
"baseline_y": "",
|
|
3815
|
+
"baseline_y<desc>": "Vertical position of the text baseline for this text object.",
|
|
3816
|
+
"baseline_y<use>": "Compare line positions or detect superscript offset relative to body lines.",
|
|
3817
|
+
"baseline_y<type>": "float",
|
|
3818
|
+
"pdf_rect<desc>": "Rectangle in default user space as left, bottom, right, top (same meaning as bbox for query macros).",
|
|
3819
|
+
"pdf_rect<type>": "rec",
|
|
3820
|
+
"pdf_rect": {
|
|
3821
|
+
"left": 0,
|
|
3822
|
+
"bottom": 0,
|
|
3823
|
+
"right": 0,
|
|
3824
|
+
"top": 0
|
|
3825
|
+
},
|
|
3826
|
+
"pdf_rgb<desc>": "Device RGB triplet 0–255 per channel for comparisons against fill_color or stroke_color.",
|
|
3827
|
+
"pdf_rgb<type>": "rgb",
|
|
3828
|
+
"pdf_rgb": {
|
|
3829
|
+
"red": 0,
|
|
3830
|
+
"green": 0,
|
|
3831
|
+
"blue": 0
|
|
3832
|
+
},
|
|
3833
|
+
"reflow": "",
|
|
3834
|
+
"reflow<desc>": "When false, hard line breaks in the PDF are preserved in logical text; when true, softer wrapping may be applied depending on export path.",
|
|
3835
|
+
"reflow<use>": "Disable reflow for poetry or code blocks where each line must stay separate.",
|
|
3836
|
+
"reflow<type>": "bool",
|
|
3837
|
+
"reflow<value>": [
|
|
3838
|
+
"true",
|
|
3839
|
+
"false"
|
|
3840
|
+
],
|
|
3841
|
+
"row_num": "",
|
|
3842
|
+
"row_num<desc>": "Detected row count for the parent table element.",
|
|
3843
|
+
"row_num<use>": "Target rules at tables with a specific vertical extent (e.g. header-only grids).",
|
|
3844
|
+
"row_num<type>": "int",
|
|
3845
|
+
"table_type": "",
|
|
3846
|
+
"table_type<desc>": "Classifier label for how the table was formed: graphic (paths), isolated, row/column text layout, or form-derived.",
|
|
3847
|
+
"table_type<use>": "Branch template logic: e.g. stricter tag_table rules for graphic tables only.",
|
|
3848
|
+
"table_type<type>": "string",
|
|
3849
|
+
"table_type<value>": [
|
|
3850
|
+
"graphic",
|
|
3851
|
+
"isolated",
|
|
3852
|
+
"row",
|
|
3853
|
+
"col",
|
|
3854
|
+
"form"
|
|
3855
|
+
],
|
|
3856
|
+
"text": "",
|
|
3857
|
+
"text<desc>": "Unicode string extracted for the object or tag: prefers marked-content text (MCID), then related annotation strings; length may be capped by max_char in tooling.",
|
|
3858
|
+
"text<use>": "Regex or equality tests on body copy, headers, or link visible text.",
|
|
3859
|
+
"text<type>": "string",
|
|
3860
|
+
"text_flag": "",
|
|
3861
|
+
"text_flag<desc>": "Paragraph-level caption and line-break hints set from detection or template (table_caption, filling, …).",
|
|
3862
|
+
"text_flag<use>": "Route captions into tag roles or suppress paragraph merge across caption boundaries.",
|
|
3863
|
+
"text_flag<type>": "string",
|
|
3864
|
+
"text_flag<value>": [
|
|
3865
|
+
"none",
|
|
3866
|
+
"table_caption",
|
|
3867
|
+
"image_caption",
|
|
3868
|
+
"chart_caption",
|
|
3869
|
+
"note_caption",
|
|
3870
|
+
"filling",
|
|
3871
|
+
"uppercase",
|
|
3872
|
+
"new_line"
|
|
3873
|
+
],
|
|
3874
|
+
"text_line_flag": "",
|
|
3875
|
+
"text_line_flag<desc>": "Line-level layout hints: hyphenation, indent, terminal punctuation, drop cap, filling, etc.",
|
|
3876
|
+
"text_line_flag<use>": "Tune text_line_update or neighbours rules when line breaks carry semantic meaning.",
|
|
3877
|
+
"text_line_flag<type>": "string",
|
|
3878
|
+
"text_line_flag<value>": [
|
|
3879
|
+
"hyphen",
|
|
3880
|
+
"new_line",
|
|
3881
|
+
"indent",
|
|
3882
|
+
"terminal",
|
|
3883
|
+
"drop_cap",
|
|
3884
|
+
"filling",
|
|
3885
|
+
"uppercase",
|
|
3886
|
+
"no_new_line"
|
|
3887
|
+
],
|
|
3888
|
+
"text_state_flag": "",
|
|
3889
|
+
"text_state_flag<desc>": "Graphics-state text attributes on a text run: underline, strikeout, highlight, sub/superscript, unicode hints.",
|
|
3890
|
+
"text_state_flag<use>": "Select runs for special tagging (e.g. underline as Link) in text_run_update.",
|
|
3891
|
+
"text_state_flag<type>": "string",
|
|
3892
|
+
"text_state_flag<value>": [
|
|
3893
|
+
"underline",
|
|
3894
|
+
"strikeout",
|
|
3895
|
+
"highlight",
|
|
3896
|
+
"subscript",
|
|
3897
|
+
"superscript",
|
|
3898
|
+
"no_unicode",
|
|
3899
|
+
"white_space",
|
|
3900
|
+
"unicode"
|
|
3901
|
+
],
|
|
3902
|
+
"word_flag": "",
|
|
3903
|
+
"word_flag<desc>": "Word-level classification flag (hyphen, bullet, numbering family, etc.) aligned with pagemap_regex-driven markers.",
|
|
3904
|
+
"word_flag<use>": "Target word_update rules at bullets, ordinals, or page numbers without matching raw Unicode.",
|
|
3905
|
+
"word_flag<type>": "string",
|
|
3906
|
+
"word_flag<value>": [
|
|
3907
|
+
"hyphen",
|
|
3908
|
+
"bullet",
|
|
3909
|
+
"colon",
|
|
3910
|
+
"number",
|
|
3911
|
+
"subscript",
|
|
3912
|
+
"superscript",
|
|
3913
|
+
"terminal",
|
|
3914
|
+
"capital",
|
|
3915
|
+
"image",
|
|
3916
|
+
"decimal_num",
|
|
3917
|
+
"roman_num",
|
|
3918
|
+
"letter_num",
|
|
3919
|
+
"page_num",
|
|
3920
|
+
"filling",
|
|
3921
|
+
"uppercase",
|
|
3922
|
+
"comma",
|
|
3923
|
+
"no_unicode"
|
|
3924
|
+
]
|
|
3925
|
+
},
|
|
3926
|
+
"suffix<desc>": "Meta-keys (<desc>, <type>, <link>, …) that annotate schema entries for editors and docs, not runtime JSON children.",
|
|
3927
|
+
"suffix<rule>": "A name suffix marks the preceding token as the real key; the suffix segment is documentation only and is not loaded as a child property.",
|
|
3928
|
+
"suffix": {
|
|
3929
|
+
"<desc>": "Human-readable explanation of the key or node.",
|
|
3930
|
+
"<use>": "Guidance for template authors (when to set the value, typical ranges).",
|
|
3931
|
+
"<type>": "JSON or UI type of the value (string, bool, int, …).",
|
|
3932
|
+
"<types>": "Allowed types when a node accepts more than one representation.",
|
|
3933
|
+
"<value>": "Enumeration of legal string tokens or literals for pickers.",
|
|
3934
|
+
"<default>": "Fallback value when the author omits the field.",
|
|
3935
|
+
"<rule>": "Extra constraints or interaction between sibling keys.",
|
|
3936
|
+
"<num_values>": "How many operands this operator or field expects.",
|
|
3937
|
+
"<link>": "Pointer to another schema path (schema:…) for shared definitions.",
|
|
3938
|
+
"<insert>": "Default tree-node payload used when inserting this schema shape in UI editors."
|
|
3939
|
+
},
|
|
3940
|
+
"condition<desc>": "How a field is compared: fixed value (condition_value), one operator object (comparison), or an array of operators (comparison_array).",
|
|
3941
|
+
"condition<use>": "Reference param indices like 0_width with $lt / $regex / $in under query logical groups.",
|
|
3942
|
+
"condition": {
|
|
3943
|
+
"condition_value<desc>": "Example: {\"0_width\": 100} — parameter 0 width must equal 100.",
|
|
3944
|
+
"condition_value": 0,
|
|
3945
|
+
"comparison<desc>": "Example: {\"0_width\": {\"$lt\": 100}} — one comparison operator per field.",
|
|
3946
|
+
"comparison": {
|
|
3947
|
+
"$eq": 0
|
|
3948
|
+
},
|
|
3949
|
+
"comparison_array<desc>": "Example: {\"0_width\": [{\"$lt\": 10}, {\"$gt\": 5}]} — multiple operator objects chained on the same indexed field.",
|
|
3950
|
+
"comparison_array": {
|
|
3951
|
+
"$gt": 0,
|
|
3952
|
+
"$lt": 0
|
|
3953
|
+
}
|
|
3954
|
+
},
|
|
3955
|
+
"keywords<desc>": "Reserved template keywords: query, param, statement, disable, and related control fields shared by function nodes.",
|
|
3956
|
+
"keywords": {
|
|
3957
|
+
"general": "",
|
|
3958
|
+
"general<desc>": "Metadata bucket (version, dates, id, SDK) at template root.",
|
|
3959
|
+
"template": "",
|
|
3960
|
+
"template<desc>": "Root object that holds pagemap, element_create, *_update functions, and tagging hooks.",
|
|
3961
|
+
"query": "",
|
|
3962
|
+
"query<desc>": "Present on every function node: declares which object types are evaluated and optional $and/$or condition trees.",
|
|
3963
|
+
"param": "",
|
|
3964
|
+
"param<desc>": "Nested arrays of type names (e.g. pde_text, pde_table) aligned with the function’s insert signature.",
|
|
3965
|
+
"param<type>": "array_param",
|
|
3966
|
+
"statement": "$if",
|
|
3967
|
+
"statement<desc>": "$if / $elif / $else on a function node: evaluation stops at the first matching branch with a passing query.",
|
|
3968
|
+
"statement<type>": "string",
|
|
3969
|
+
"statement<value>": [
|
|
3970
|
+
"$if",
|
|
3971
|
+
"$elif",
|
|
3972
|
+
"$else"
|
|
3973
|
+
],
|
|
3974
|
+
"disable": false,
|
|
3975
|
+
"disable<desc>": "If true, the entire function node is skipped. Default false.",
|
|
3976
|
+
"disable<type>": "bool",
|
|
3977
|
+
"disable<value>": [
|
|
3978
|
+
"true",
|
|
3979
|
+
"false"
|
|
3980
|
+
],
|
|
3981
|
+
"purpose": "",
|
|
3982
|
+
"purpose<desc>": "Optional free-text note for authors; not interpreted by the engine.",
|
|
3983
|
+
"purpose<type>": "string"
|
|
3984
|
+
},
|
|
3985
|
+
"insert<desc>": "Factory snippets the schema UI uses when inserting new template fragments.",
|
|
3986
|
+
"insert<use>": "Not merged into runtime templates automatically; editors copy these shapes into element_create or cell grids.",
|
|
3987
|
+
"insert": {
|
|
3988
|
+
"elements_add_value": "",
|
|
3989
|
+
"elements_add_value<desc>": "Minimal pde_text initial element with zero bbox (replace with real coordinates).",
|
|
3990
|
+
"elements_add_value<insert>": {
|
|
3991
|
+
"type": "pde_text",
|
|
3992
|
+
"bbox": {
|
|
3993
|
+
"left": "0",
|
|
3994
|
+
"bottom": "0",
|
|
3995
|
+
"right": "0",
|
|
3996
|
+
"top": "0"
|
|
3997
|
+
}
|
|
3998
|
+
},
|
|
3999
|
+
"elements_add_cell": "",
|
|
4000
|
+
"elements_add_cell<desc>": "Starter pde_cell with unit span and default scope for table shells.",
|
|
4001
|
+
"elements_add_cell<insert>": {
|
|
4002
|
+
"bbox": [
|
|
4003
|
+
0,
|
|
4004
|
+
0,
|
|
4005
|
+
0,
|
|
4006
|
+
0
|
|
4007
|
+
],
|
|
4008
|
+
"cell_column": 1,
|
|
4009
|
+
"cell_column_span": 1,
|
|
4010
|
+
"cell_header": false,
|
|
4011
|
+
"cell_row": 1,
|
|
4012
|
+
"cell_row_span": 1,
|
|
4013
|
+
"cell_scope": "column",
|
|
4014
|
+
"type": "pde_cell"
|
|
4015
|
+
}
|
|
4016
|
+
},
|
|
4017
|
+
"math_expressions": "SUM",
|
|
4018
|
+
"math_expressions<desc>": "Built-in math helpers (SUM, MIN, MAX, …) for macro or variable definitions in advanced templates.",
|
|
4019
|
+
"math_expressions<use>": "Compose bbox or threshold expressions that reference multiple anchor variables.",
|
|
4020
|
+
"math_expressions<type>": "string",
|
|
4021
|
+
"math_expressions<value>": [
|
|
4022
|
+
"SUM()",
|
|
4023
|
+
"MINUS()",
|
|
4024
|
+
"ABS()",
|
|
4025
|
+
"MULTIPLY()",
|
|
4026
|
+
"DIVIDE()",
|
|
4027
|
+
"MIN()",
|
|
4028
|
+
"MAX()",
|
|
4029
|
+
"MOD()",
|
|
4030
|
+
"FLOOR()",
|
|
4031
|
+
"CEILING()"
|
|
4032
|
+
]
|
|
4033
|
+
}
|
|
4034
|
+
}
|