exstruct 0.2.11__tar.gz → 0.2.21__tar.gz
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.
- {exstruct-0.2.11 → exstruct-0.2.21}/PKG-INFO +43 -38
- {exstruct-0.2.11 → exstruct-0.2.21}/README.md +42 -37
- {exstruct-0.2.11 → exstruct-0.2.21}/pyproject.toml +1 -1
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/__init__.py +2 -1
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/core/cells.py +58 -19
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/core/integrate.py +15 -7
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/engine.py +7 -1
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/models/__init__.py +1 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/LICENSE +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/cli/main.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/core/__init__.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/core/charts.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/core/shapes.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/io/__init__.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/models/maps.py +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/py.typed +0 -0
- {exstruct-0.2.11 → exstruct-0.2.21}/src/exstruct/render/__init__.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.3
|
|
2
2
|
Name: exstruct
|
|
3
|
-
Version: 0.2.
|
|
3
|
+
Version: 0.2.21
|
|
4
4
|
Summary: Excel to structured JSON (tables, shapes, charts) for LLM/RAG pipelines
|
|
5
5
|
Keywords: excel,structure,data,exstruct
|
|
6
6
|
Author: harumiWeb
|
|
@@ -56,16 +56,16 @@ Description-Content-Type: text/markdown
|
|
|
56
56
|
|
|
57
57
|
# ExStruct — Excel Structured Extraction Engine
|
|
58
58
|
|
|
59
|
-
[](https://pypi.org/project/exstruct/) [](https://pepy.tech/projects/exstruct)  [](https://pypi.org/project/exstruct/) [](https://pepy.tech/projects/exstruct)  [](https://github.com/harumiWeb/exstruct/actions/workflows/pytest.yml)
|
|
60
60
|
|
|
61
61
|
|
|
62
62
|
|
|
63
|
-
ExStruct reads Excel workbooks and outputs structured data (tables, shapes, charts) as JSON by default, with optional YAML/TOON formats. It targets both COM/Excel environments (rich extraction) and non-COM environments (cells + table candidates), with tunable detection heuristics and multiple output modes to fit LLM/RAG pipelines.
|
|
63
|
+
ExStruct reads Excel workbooks and outputs structured data (tables, shapes, charts, hyperlinks) as JSON by default, with optional YAML/TOON formats. It targets both COM/Excel environments (rich extraction) and non-COM environments (cells + table candidates), with tunable detection heuristics and multiple output modes to fit LLM/RAG pipelines.
|
|
64
64
|
|
|
65
65
|
## Features
|
|
66
66
|
|
|
67
67
|
- **Excel → Structured JSON**: cells, shapes, charts, and table candidates per sheet.
|
|
68
|
-
- **Output modes**: `light` (cells + table candidates only), `standard` (texted shapes + arrows, charts), `verbose` (all shapes with width/height).
|
|
68
|
+
- **Output modes**: `light` (cells + table candidates only), `standard` (texted shapes + arrows, charts), `verbose` (all shapes with width/height). Verbose also emits cell hyperlinks.
|
|
69
69
|
- **Formats**: JSON (compact by default, `--pretty` available), YAML, TOON (optional dependencies).
|
|
70
70
|
- **Table detection tuning**: adjust heuristics at runtime via API.
|
|
71
71
|
- **CLI rendering** (Excel required): optional PDF and per-sheet PNGs.
|
|
@@ -77,17 +77,17 @@ ExStruct reads Excel workbooks and outputs structured data (tables, shapes, char
|
|
|
77
77
|
pip install exstruct
|
|
78
78
|
```
|
|
79
79
|
|
|
80
|
-
Optional extras:
|
|
81
|
-
|
|
82
|
-
- YAML: `pip install pyyaml`
|
|
83
|
-
- TOON: `pip install python-toon`
|
|
84
|
-
- Rendering (PDF/PNG): Excel + `pip install pypdfium2 pillow`
|
|
85
|
-
- All extras at once: `pip install exstruct[yaml,toon,render]`
|
|
86
|
-
|
|
87
|
-
Platform note:
|
|
88
|
-
- Full extraction (shapes/charts) targets Windows + Excel (COM via xlwings). On other platforms, use `mode=light` to get cells + `table_candidates` safely.
|
|
89
|
-
|
|
90
|
-
## Quick Start (CLI)
|
|
80
|
+
Optional extras:
|
|
81
|
+
|
|
82
|
+
- YAML: `pip install pyyaml`
|
|
83
|
+
- TOON: `pip install python-toon`
|
|
84
|
+
- Rendering (PDF/PNG): Excel + `pip install pypdfium2 pillow`
|
|
85
|
+
- All extras at once: `pip install exstruct[yaml,toon,render]`
|
|
86
|
+
|
|
87
|
+
Platform note:
|
|
88
|
+
- Full extraction (shapes/charts) targets Windows + Excel (COM via xlwings). On other platforms, use `mode=light` to get cells + `table_candidates` safely.
|
|
89
|
+
|
|
90
|
+
## Quick Start (CLI)
|
|
91
91
|
|
|
92
92
|
```bash
|
|
93
93
|
exstruct input.xlsx > output.json # compact JSON to stdout (default)
|
|
@@ -120,15 +120,19 @@ wb.save("out.json", pretty=True) # WorkbookData → file (by extension)
|
|
|
120
120
|
first_sheet.save("sheet.json") # SheetData → file (by extension)
|
|
121
121
|
print(first_sheet.to_yaml()) # YAML text (requires pyyaml)
|
|
122
122
|
|
|
123
|
-
# ExStructEngine: per-instance options for extraction/output
|
|
124
|
-
from exstruct import ExStructEngine, StructOptions, OutputOptions
|
|
125
|
-
|
|
126
|
-
engine = ExStructEngine(
|
|
127
|
-
options=StructOptions(mode="
|
|
128
|
-
output=OutputOptions(include_shapes=False, pretty=True),
|
|
129
|
-
)
|
|
130
|
-
wb2 = engine.extract("input.xlsx")
|
|
131
|
-
engine.export(wb2, Path("out_filtered.json")) # drops shapes via OutputOptions
|
|
123
|
+
# ExStructEngine: per-instance options for extraction/output
|
|
124
|
+
from exstruct import ExStructEngine, StructOptions, OutputOptions
|
|
125
|
+
|
|
126
|
+
engine = ExStructEngine(
|
|
127
|
+
options=StructOptions(mode="verbose"), # verbose includes hyperlinks by default
|
|
128
|
+
output=OutputOptions(include_shapes=False, pretty=True),
|
|
129
|
+
)
|
|
130
|
+
wb2 = engine.extract("input.xlsx")
|
|
131
|
+
engine.export(wb2, Path("out_filtered.json")) # drops shapes via OutputOptions
|
|
132
|
+
|
|
133
|
+
# Enable hyperlinks in other modes
|
|
134
|
+
engine_links = ExStructEngine(options=StructOptions(mode="standard", include_cell_links=True))
|
|
135
|
+
with_links = engine_links.extract("input.xlsx")
|
|
132
136
|
```
|
|
133
137
|
|
|
134
138
|
**Note (non-COM environments):** If Excel COM is unavailable, extraction still runs and returns cells + `table_candidates`; `shapes`/`charts` will be empty.
|
|
@@ -148,11 +152,11 @@ set_table_detection_params(
|
|
|
148
152
|
|
|
149
153
|
Use higher thresholds to reduce false positives; lower them if true tables are missed.
|
|
150
154
|
|
|
151
|
-
## Output Modes
|
|
152
|
-
|
|
153
|
-
- **light**: cells + table candidates (no COM needed).
|
|
154
|
-
- **standard**: texted shapes + arrows, charts (COM if available), table candidates.
|
|
155
|
-
- **verbose**: all shapes (with width/height), charts, table candidates.
|
|
155
|
+
## Output Modes
|
|
156
|
+
|
|
157
|
+
- **light**: cells + table candidates (no COM needed).
|
|
158
|
+
- **standard**: texted shapes + arrows, charts (COM if available), table candidates. Hyperlinks are off unless `include_cell_links=True`.
|
|
159
|
+
- **verbose**: all shapes (with width/height), charts, table candidates, and cell hyperlinks.
|
|
156
160
|
|
|
157
161
|
## Error Handling / Fallbacks
|
|
158
162
|
|
|
@@ -376,15 +380,16 @@ BSD-3-Clause. See `LICENSE` for details.
|
|
|
376
380
|
- API Reference (GitHub Pages): https://harumiweb.github.io/exstruct/
|
|
377
381
|
# Engine option cheat sheet
|
|
378
382
|
|
|
379
|
-
| Option class | Field | Meaning |
|
|
380
|
-
| -------------- | ------------------- | ------- |
|
|
381
|
-
| StructOptions | mode | "light"/"standard"/"verbose" |
|
|
382
|
-
| | table_params | Dict passed to `set_table_detection_params` (table_score_threshold, density_min, coverage_min, min_nonempty_cells) |
|
|
383
|
-
|
|
|
384
|
-
|
|
|
385
|
-
| |
|
|
386
|
-
| |
|
|
387
|
-
| |
|
|
383
|
+
| Option class | Field | Meaning |
|
|
384
|
+
| -------------- | ------------------- | ------- |
|
|
385
|
+
| StructOptions | mode | "light"/"standard"/"verbose" |
|
|
386
|
+
| | table_params | Dict passed to `set_table_detection_params` (table_score_threshold, density_min, coverage_min, min_nonempty_cells) |
|
|
387
|
+
| | include_cell_links | Include cell hyperlinks in `rows[*].links` (None -> auto: verbose=True, others=False) |
|
|
388
|
+
| OutputOptions | fmt | Default format ("json"/"yaml"/"yml"/"toon") |
|
|
389
|
+
| | pretty / indent | Pretty-print JSON and control indent |
|
|
390
|
+
| | include_rows | Include rows (False to drop) |
|
|
391
|
+
| | include_shapes | Include shapes |
|
|
392
|
+
| | include_charts | Include charts |
|
|
388
393
|
| | include_tables | Include table_candidates |
|
|
389
394
|
| | sheets_dir | Optional directory for per-sheet exports |
|
|
390
395
|
| | stream | Default stream when output_path is None |
|
|
@@ -1,15 +1,15 @@
|
|
|
1
1
|
# ExStruct — Excel Structured Extraction Engine
|
|
2
2
|
|
|
3
|
-
[](https://pypi.org/project/exstruct/) [](https://pepy.tech/projects/exstruct)  [](https://pypi.org/project/exstruct/) [](https://pepy.tech/projects/exstruct)  [](https://github.com/harumiWeb/exstruct/actions/workflows/pytest.yml)
|
|
4
4
|
|
|
5
5
|
|
|
6
6
|
|
|
7
|
-
ExStruct reads Excel workbooks and outputs structured data (tables, shapes, charts) as JSON by default, with optional YAML/TOON formats. It targets both COM/Excel environments (rich extraction) and non-COM environments (cells + table candidates), with tunable detection heuristics and multiple output modes to fit LLM/RAG pipelines.
|
|
7
|
+
ExStruct reads Excel workbooks and outputs structured data (tables, shapes, charts, hyperlinks) as JSON by default, with optional YAML/TOON formats. It targets both COM/Excel environments (rich extraction) and non-COM environments (cells + table candidates), with tunable detection heuristics and multiple output modes to fit LLM/RAG pipelines.
|
|
8
8
|
|
|
9
9
|
## Features
|
|
10
10
|
|
|
11
11
|
- **Excel → Structured JSON**: cells, shapes, charts, and table candidates per sheet.
|
|
12
|
-
- **Output modes**: `light` (cells + table candidates only), `standard` (texted shapes + arrows, charts), `verbose` (all shapes with width/height).
|
|
12
|
+
- **Output modes**: `light` (cells + table candidates only), `standard` (texted shapes + arrows, charts), `verbose` (all shapes with width/height). Verbose also emits cell hyperlinks.
|
|
13
13
|
- **Formats**: JSON (compact by default, `--pretty` available), YAML, TOON (optional dependencies).
|
|
14
14
|
- **Table detection tuning**: adjust heuristics at runtime via API.
|
|
15
15
|
- **CLI rendering** (Excel required): optional PDF and per-sheet PNGs.
|
|
@@ -21,17 +21,17 @@ ExStruct reads Excel workbooks and outputs structured data (tables, shapes, char
|
|
|
21
21
|
pip install exstruct
|
|
22
22
|
```
|
|
23
23
|
|
|
24
|
-
Optional extras:
|
|
25
|
-
|
|
26
|
-
- YAML: `pip install pyyaml`
|
|
27
|
-
- TOON: `pip install python-toon`
|
|
28
|
-
- Rendering (PDF/PNG): Excel + `pip install pypdfium2 pillow`
|
|
29
|
-
- All extras at once: `pip install exstruct[yaml,toon,render]`
|
|
30
|
-
|
|
31
|
-
Platform note:
|
|
32
|
-
- Full extraction (shapes/charts) targets Windows + Excel (COM via xlwings). On other platforms, use `mode=light` to get cells + `table_candidates` safely.
|
|
33
|
-
|
|
34
|
-
## Quick Start (CLI)
|
|
24
|
+
Optional extras:
|
|
25
|
+
|
|
26
|
+
- YAML: `pip install pyyaml`
|
|
27
|
+
- TOON: `pip install python-toon`
|
|
28
|
+
- Rendering (PDF/PNG): Excel + `pip install pypdfium2 pillow`
|
|
29
|
+
- All extras at once: `pip install exstruct[yaml,toon,render]`
|
|
30
|
+
|
|
31
|
+
Platform note:
|
|
32
|
+
- Full extraction (shapes/charts) targets Windows + Excel (COM via xlwings). On other platforms, use `mode=light` to get cells + `table_candidates` safely.
|
|
33
|
+
|
|
34
|
+
## Quick Start (CLI)
|
|
35
35
|
|
|
36
36
|
```bash
|
|
37
37
|
exstruct input.xlsx > output.json # compact JSON to stdout (default)
|
|
@@ -64,15 +64,19 @@ wb.save("out.json", pretty=True) # WorkbookData → file (by extension)
|
|
|
64
64
|
first_sheet.save("sheet.json") # SheetData → file (by extension)
|
|
65
65
|
print(first_sheet.to_yaml()) # YAML text (requires pyyaml)
|
|
66
66
|
|
|
67
|
-
# ExStructEngine: per-instance options for extraction/output
|
|
68
|
-
from exstruct import ExStructEngine, StructOptions, OutputOptions
|
|
69
|
-
|
|
70
|
-
engine = ExStructEngine(
|
|
71
|
-
options=StructOptions(mode="
|
|
72
|
-
output=OutputOptions(include_shapes=False, pretty=True),
|
|
73
|
-
)
|
|
74
|
-
wb2 = engine.extract("input.xlsx")
|
|
75
|
-
engine.export(wb2, Path("out_filtered.json")) # drops shapes via OutputOptions
|
|
67
|
+
# ExStructEngine: per-instance options for extraction/output
|
|
68
|
+
from exstruct import ExStructEngine, StructOptions, OutputOptions
|
|
69
|
+
|
|
70
|
+
engine = ExStructEngine(
|
|
71
|
+
options=StructOptions(mode="verbose"), # verbose includes hyperlinks by default
|
|
72
|
+
output=OutputOptions(include_shapes=False, pretty=True),
|
|
73
|
+
)
|
|
74
|
+
wb2 = engine.extract("input.xlsx")
|
|
75
|
+
engine.export(wb2, Path("out_filtered.json")) # drops shapes via OutputOptions
|
|
76
|
+
|
|
77
|
+
# Enable hyperlinks in other modes
|
|
78
|
+
engine_links = ExStructEngine(options=StructOptions(mode="standard", include_cell_links=True))
|
|
79
|
+
with_links = engine_links.extract("input.xlsx")
|
|
76
80
|
```
|
|
77
81
|
|
|
78
82
|
**Note (non-COM environments):** If Excel COM is unavailable, extraction still runs and returns cells + `table_candidates`; `shapes`/`charts` will be empty.
|
|
@@ -92,11 +96,11 @@ set_table_detection_params(
|
|
|
92
96
|
|
|
93
97
|
Use higher thresholds to reduce false positives; lower them if true tables are missed.
|
|
94
98
|
|
|
95
|
-
## Output Modes
|
|
96
|
-
|
|
97
|
-
- **light**: cells + table candidates (no COM needed).
|
|
98
|
-
- **standard**: texted shapes + arrows, charts (COM if available), table candidates.
|
|
99
|
-
- **verbose**: all shapes (with width/height), charts, table candidates.
|
|
99
|
+
## Output Modes
|
|
100
|
+
|
|
101
|
+
- **light**: cells + table candidates (no COM needed).
|
|
102
|
+
- **standard**: texted shapes + arrows, charts (COM if available), table candidates. Hyperlinks are off unless `include_cell_links=True`.
|
|
103
|
+
- **verbose**: all shapes (with width/height), charts, table candidates, and cell hyperlinks.
|
|
100
104
|
|
|
101
105
|
## Error Handling / Fallbacks
|
|
102
106
|
|
|
@@ -320,15 +324,16 @@ BSD-3-Clause. See `LICENSE` for details.
|
|
|
320
324
|
- API Reference (GitHub Pages): https://harumiweb.github.io/exstruct/
|
|
321
325
|
# Engine option cheat sheet
|
|
322
326
|
|
|
323
|
-
| Option class | Field | Meaning |
|
|
324
|
-
| -------------- | ------------------- | ------- |
|
|
325
|
-
| StructOptions | mode | "light"/"standard"/"verbose" |
|
|
326
|
-
| | table_params | Dict passed to `set_table_detection_params` (table_score_threshold, density_min, coverage_min, min_nonempty_cells) |
|
|
327
|
-
|
|
|
328
|
-
|
|
|
329
|
-
| |
|
|
330
|
-
| |
|
|
331
|
-
| |
|
|
327
|
+
| Option class | Field | Meaning |
|
|
328
|
+
| -------------- | ------------------- | ------- |
|
|
329
|
+
| StructOptions | mode | "light"/"standard"/"verbose" |
|
|
330
|
+
| | table_params | Dict passed to `set_table_detection_params` (table_score_threshold, density_min, coverage_min, min_nonempty_cells) |
|
|
331
|
+
| | include_cell_links | Include cell hyperlinks in `rows[*].links` (None -> auto: verbose=True, others=False) |
|
|
332
|
+
| OutputOptions | fmt | Default format ("json"/"yaml"/"yml"/"toon") |
|
|
333
|
+
| | pretty / indent | Pretty-print JSON and control indent |
|
|
334
|
+
| | include_rows | Include rows (False to drop) |
|
|
335
|
+
| | include_shapes | Include shapes |
|
|
336
|
+
| | include_charts | Include charts |
|
|
332
337
|
| | include_tables | Include table_candidates |
|
|
333
338
|
| | sheets_dir | Optional directory for per-sheet exports |
|
|
334
339
|
| | stream | Default stream when output_path is None |
|
|
@@ -36,7 +36,8 @@ ExtractionMode = Literal["light", "standard", "verbose"]
|
|
|
36
36
|
|
|
37
37
|
def extract(file_path: str | Path, mode: ExtractionMode = "standard") -> WorkbookData:
|
|
38
38
|
"""Extract workbook semantic structure and return WorkbookData."""
|
|
39
|
-
|
|
39
|
+
include_links = True if mode == "verbose" else False
|
|
40
|
+
engine = ExStructEngine(options=StructOptions(mode=mode, include_cell_links=include_links))
|
|
40
41
|
return engine.extract(file_path, mode=mode)
|
|
41
42
|
|
|
42
43
|
|
|
@@ -33,25 +33,64 @@ def warn_once(key: str, message: str) -> None:
|
|
|
33
33
|
_warned_keys.add(key)
|
|
34
34
|
|
|
35
35
|
|
|
36
|
-
def extract_sheet_cells(file_path: Path) -> Dict[str, List[CellRow]]:
|
|
37
|
-
"""Read all sheets via pandas and convert to CellRow list while skipping empty cells."""
|
|
38
|
-
dfs = pd.read_excel(file_path, header=None, sheet_name=None, dtype=str)
|
|
39
|
-
result: Dict[str, List[CellRow]] = {}
|
|
40
|
-
for sheet_name, df in dfs.items():
|
|
41
|
-
df = df.fillna("")
|
|
42
|
-
rows: List[CellRow] = []
|
|
43
|
-
for excel_row, row in enumerate(df.itertuples(index=False, name=None), start=1):
|
|
44
|
-
filtered: Dict[str, int | float | str] = {}
|
|
45
|
-
for j, v in enumerate(row):
|
|
46
|
-
s = "" if v is None else str(v)
|
|
47
|
-
if s.strip() == "":
|
|
48
|
-
continue
|
|
49
|
-
filtered[str(j)] = _coerce_numeric_preserve_format(s)
|
|
50
|
-
if not filtered:
|
|
51
|
-
continue
|
|
52
|
-
rows.append(CellRow(r=excel_row, c=filtered))
|
|
53
|
-
result[sheet_name] = rows
|
|
54
|
-
return result
|
|
36
|
+
def extract_sheet_cells(file_path: Path) -> Dict[str, List[CellRow]]:
|
|
37
|
+
"""Read all sheets via pandas and convert to CellRow list while skipping empty cells."""
|
|
38
|
+
dfs = pd.read_excel(file_path, header=None, sheet_name=None, dtype=str)
|
|
39
|
+
result: Dict[str, List[CellRow]] = {}
|
|
40
|
+
for sheet_name, df in dfs.items():
|
|
41
|
+
df = df.fillna("")
|
|
42
|
+
rows: List[CellRow] = []
|
|
43
|
+
for excel_row, row in enumerate(df.itertuples(index=False, name=None), start=1):
|
|
44
|
+
filtered: Dict[str, int | float | str] = {}
|
|
45
|
+
for j, v in enumerate(row):
|
|
46
|
+
s = "" if v is None else str(v)
|
|
47
|
+
if s.strip() == "":
|
|
48
|
+
continue
|
|
49
|
+
filtered[str(j)] = _coerce_numeric_preserve_format(s)
|
|
50
|
+
if not filtered:
|
|
51
|
+
continue
|
|
52
|
+
rows.append(CellRow(r=excel_row, c=filtered))
|
|
53
|
+
result[sheet_name] = rows
|
|
54
|
+
return result
|
|
55
|
+
|
|
56
|
+
|
|
57
|
+
def extract_sheet_cells_with_links(file_path: Path) -> Dict[str, List[CellRow]]:
|
|
58
|
+
"""
|
|
59
|
+
Extract cells and hyperlinks per sheet.
|
|
60
|
+
|
|
61
|
+
Returns:
|
|
62
|
+
{sheet_name: [CellRow(r=..., c=..., links={"col_index": url, ...}), ...]}
|
|
63
|
+
|
|
64
|
+
Notes:
|
|
65
|
+
- Uses pandas extraction for values (same filtering as extract_sheet_cells).
|
|
66
|
+
- Collects hyperlinks via openpyxl (requires read_only=False because border maps/hyperlinks need full objects).
|
|
67
|
+
- Links are mapped by column index string (e.g., "0") to hyperlink.target.
|
|
68
|
+
"""
|
|
69
|
+
cell_rows = extract_sheet_cells(file_path)
|
|
70
|
+
wb = load_workbook(file_path, data_only=True, read_only=False)
|
|
71
|
+
links_by_sheet: Dict[str, Dict[int, Dict[str, str]]] = {}
|
|
72
|
+
for ws in wb.worksheets:
|
|
73
|
+
sheet_links: Dict[int, Dict[str, str]] = {}
|
|
74
|
+
for row in ws.iter_rows():
|
|
75
|
+
for cell in row:
|
|
76
|
+
link = getattr(cell, "hyperlink", None)
|
|
77
|
+
target = getattr(link, "target", None) if link else None
|
|
78
|
+
if not target:
|
|
79
|
+
continue
|
|
80
|
+
col_str = str(cell.col_idx - 1) # zero-based to align with extract_sheet_cells
|
|
81
|
+
sheet_links.setdefault(cell.row, {})[col_str] = target
|
|
82
|
+
links_by_sheet[ws.title] = sheet_links
|
|
83
|
+
|
|
84
|
+
merged: Dict[str, List[CellRow]] = {}
|
|
85
|
+
for sheet_name, rows in cell_rows.items():
|
|
86
|
+
sheet_links = links_by_sheet.get(sheet_name, {})
|
|
87
|
+
merged_rows: List[CellRow] = []
|
|
88
|
+
for row in rows:
|
|
89
|
+
links = sheet_links.get(row.r, {})
|
|
90
|
+
merged_rows.append(CellRow(r=row.r, c=row.c, links=links or None))
|
|
91
|
+
merged[sheet_name] = merged_rows
|
|
92
|
+
wb.close()
|
|
93
|
+
return merged
|
|
55
94
|
|
|
56
95
|
|
|
57
96
|
def shrink_to_content(
|
|
@@ -4,15 +4,17 @@ from pathlib import Path
|
|
|
4
4
|
from typing import Dict, List, Literal
|
|
5
5
|
|
|
6
6
|
import logging
|
|
7
|
+
import os
|
|
7
8
|
|
|
8
9
|
import xlwings as xw
|
|
9
10
|
|
|
10
11
|
from ..models import CellRow, Shape, SheetData, WorkbookData
|
|
11
|
-
from .cells import (
|
|
12
|
-
detect_tables,
|
|
13
|
-
detect_tables_openpyxl,
|
|
14
|
-
extract_sheet_cells,
|
|
15
|
-
|
|
12
|
+
from .cells import (
|
|
13
|
+
detect_tables,
|
|
14
|
+
detect_tables_openpyxl,
|
|
15
|
+
extract_sheet_cells,
|
|
16
|
+
extract_sheet_cells_with_links,
|
|
17
|
+
)
|
|
16
18
|
from .charts import get_charts
|
|
17
19
|
from .shapes import get_shapes_with_position
|
|
18
20
|
|
|
@@ -74,13 +76,16 @@ def integrate_sheet_content(
|
|
|
74
76
|
|
|
75
77
|
|
|
76
78
|
def extract_workbook(
|
|
77
|
-
file_path: Path,
|
|
79
|
+
file_path: Path,
|
|
80
|
+
mode: Literal["light", "standard", "verbose"] = "standard",
|
|
81
|
+
*,
|
|
82
|
+
include_cell_links: bool = False,
|
|
78
83
|
) -> WorkbookData:
|
|
79
84
|
"""Extract workbook and return WorkbookData; fallback to cells+tables if Excel COM is unavailable."""
|
|
80
85
|
if mode not in _ALLOWED_MODES:
|
|
81
86
|
raise ValueError(f"Unsupported mode: {mode}")
|
|
82
87
|
|
|
83
|
-
cell_data = extract_sheet_cells(file_path)
|
|
88
|
+
cell_data = extract_sheet_cells_with_links(file_path) if include_cell_links else extract_sheet_cells(file_path)
|
|
84
89
|
|
|
85
90
|
def _cells_and_tables_only(reason: str) -> WorkbookData:
|
|
86
91
|
sheets: Dict[str, SheetData] = {}
|
|
@@ -104,6 +109,9 @@ def extract_workbook(
|
|
|
104
109
|
if mode == "light":
|
|
105
110
|
return _cells_and_tables_only("Light mode selected.")
|
|
106
111
|
|
|
112
|
+
if os.getenv("SKIP_COM_TESTS"):
|
|
113
|
+
return _cells_and_tables_only("SKIP_COM_TESTS is set; skipping COM/xlwings access.")
|
|
114
|
+
|
|
107
115
|
try:
|
|
108
116
|
wb, close_app = _open_workbook(file_path)
|
|
109
117
|
except Exception as e:
|
|
@@ -32,6 +32,7 @@ class StructOptions:
|
|
|
32
32
|
|
|
33
33
|
mode: ExtractionMode = "standard"
|
|
34
34
|
table_params: Optional[dict] = None # forwarded to set_table_detection_params if provided
|
|
35
|
+
include_cell_links: Optional[bool] = None # None → auto: verbose=True, others=False
|
|
35
36
|
|
|
36
37
|
|
|
37
38
|
@dataclass(frozen=True)
|
|
@@ -133,8 +134,13 @@ class ExStructEngine:
|
|
|
133
134
|
chosen_mode = mode or self.options.mode
|
|
134
135
|
if chosen_mode not in ("light", "standard", "verbose"):
|
|
135
136
|
raise ValueError(f"Unsupported mode: {chosen_mode}")
|
|
137
|
+
include_links = (
|
|
138
|
+
self.options.include_cell_links
|
|
139
|
+
if self.options.include_cell_links is not None
|
|
140
|
+
else chosen_mode == "verbose"
|
|
141
|
+
)
|
|
136
142
|
with self._table_params_scope():
|
|
137
|
-
return extract_workbook(Path(file_path), mode=chosen_mode)
|
|
143
|
+
return extract_workbook(Path(file_path), mode=chosen_mode, include_cell_links=include_links)
|
|
138
144
|
|
|
139
145
|
def serialize(
|
|
140
146
|
self,
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|