mdld-parse 0.9.1 → 0.9.3

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.
Files changed (59) hide show
  1. package/README.md +34 -1
  2. package/dist/.nojekyll +0 -0
  3. package/dist/CNAME +1 -0
  4. package/dist/README.md +425 -0
  5. package/dist/docs/API.md +397 -0
  6. package/dist/docs/Architecture.md +190 -0
  7. package/dist/docs/Guide.md +322 -0
  8. package/dist/docs/LLM-MDLD-Inducer.md +550 -0
  9. package/dist/docs/Parser.md +654 -0
  10. package/dist/docs/Performance.md +238 -0
  11. package/dist/docs/Spec-render.md +1439 -0
  12. package/dist/docs/Subject.md +180 -0
  13. package/dist/docs/Syntax.md +463 -0
  14. package/dist/docs/TodoList.md +215 -0
  15. package/dist/docs/Use-Cases.md +354 -0
  16. package/dist/docs/index.html +108 -0
  17. package/dist/docs/index.md +56 -0
  18. package/dist/docs/origin.md +459 -0
  19. package/dist/docs/polarity.md +586 -0
  20. package/dist/docs/primary-subject.md +199 -0
  21. package/dist/docs/statements.md +226 -0
  22. package/dist/docs/tokenization-performance-report.md +164 -0
  23. package/dist/examples/Cookbook.md +32 -0
  24. package/dist/examples/Status-SHACL.md +33 -0
  25. package/dist/examples/demo/index.md +35 -0
  26. package/dist/examples/demo/journal.md +18 -0
  27. package/dist/examples/demo/minimal.md +11 -0
  28. package/dist/examples/demo/project.md +15 -0
  29. package/dist/examples/demo/recipe.md +31 -0
  30. package/dist/examples/demo/research.md +46 -0
  31. package/dist/examples/demo/space-mission.md +48 -0
  32. package/dist/examples/dogfood.md +288 -0
  33. package/dist/examples/few-shot.md +309 -0
  34. package/dist/examples/graph/rdf-graph.html +473 -0
  35. package/dist/examples/graph/rdf_force_graph_web_component.html +506 -0
  36. package/dist/examples/index.md +45 -0
  37. package/dist/examples/llm-time-workflow.md +173 -0
  38. package/dist/examples/prov-o.md +2367 -0
  39. package/dist/examples/rdf.md +827 -0
  40. package/dist/examples/shacl.md +3399 -0
  41. package/dist/examples/statements.md +43 -0
  42. package/dist/examples/workflow.md +150 -0
  43. package/dist/index.html +2827 -0
  44. package/dist/index.js +1548 -0
  45. package/dist/logo.svg +7 -0
  46. package/dist/spec/Spec-compact.md +363 -0
  47. package/dist/spec/Spec-ultra.md +221 -0
  48. package/dist/spec/Spec.md +1068 -0
  49. package/dist/spec/index.md +46 -0
  50. package/package.json +13 -8
  51. package/src/constants.js +0 -31
  52. package/src/generate.js +0 -245
  53. package/src/index.js +0 -13
  54. package/src/locate.js +0 -23
  55. package/src/merge.js +0 -157
  56. package/src/parse.js +0 -765
  57. package/src/render.js +0 -576
  58. package/src/shared.js +0 -577
  59. package/src/utils.js +0 -430
@@ -0,0 +1,18 @@
1
+ [my] <tag:alice@example.com,2026:journal:2026-07-18:>
2
+
3
+ ## July 18, 2026 {=my:day .prov:Activity label}
4
+
5
+ Great day! Mood: [happy] {my:mood}, energy: [8/10] {my:energy ^^xsd:integer}.
6
+
7
+ Location: [Central Park] {+my:park ?prov:atLocation .prov:Location label}.
8
+
9
+ Feeling [peaceful] {my:feeling} today. The park was [beautiful] {my:description}.
10
+
11
+ **Morning walk** {=my:walk .prov:Activity label} - 45 minutes, 2.5 miles. Used [running shoes] {+my:shoes ?prov:used}. Generated [heart rate data] {+my:hr ?prov:generated}.
12
+
13
+ **Reading Dune** {=my:reading .prov:Activity label} - page [42] {my:page ^^xsd:integer} of 412. Used [book] {+my:book ?prov:used}.
14
+
15
+ Met [Sam] {+my:sam .prov:Person label} for coffee. **Chat** {=my:chat .prov:Activity label} - career changes topic, 30 mins.
16
+
17
+ **Coffee** {=my:coffee .prov:Activity label} at [home] {+my:home ?prov:atLocation}. Black, 12 oz.
18
+
@@ -0,0 +1,11 @@
1
+ [my] <tag:alice@example.com,2026:>
2
+
3
+ # Machine Learning in Healthcare {=my:paper .prov:Entity label}
4
+
5
+ Pages: [24] {my:pages ^^xsd:integer}
6
+ DOI: [10.1234/ml.2024] {+https://doi.org/10.1234/ml.2024 ?my:doi}
7
+
8
+ Created by [Dr. Alice Chen] {+my:alice .prov:Person ?prov:wasAttributedTo label}.
9
+
10
+ **Attribution** {=my:attr .prov:Attribution ?prov:qualifiedAttribution label}: Created by [Dr. Alice Chen] {+my:alice ?prov:agent} as a [Lead Author] {+my:role .prov:Role ?prov:hadRole}.
11
+
@@ -0,0 +1,15 @@
1
+ [proj] <tag:alice@example.com,2026:project/>
2
+
3
+ ## Weekend Project {=proj:weekend .prov:Activity label}
4
+
5
+ Project manager [Alice] {+proj:alice .prov:Person ?prov:wasAssociatedWith}.
6
+
7
+ The project includes [kitchen cleaning] {+proj:clean .proj:Task label} with subtasks: countertops, dishes, and pantry organization. Also includes [bike repair] {+proj:bike .proj:Task label} involving tire pressure, brakes, and chain maintenance. Finally, [reading] {+proj:read .proj:Task label} covers chapters 1-6.
8
+
9
+ **Delegation** {=proj:delegation .prov:Delegation ?prov:qualifiedDelegation}
10
+
11
+ Resources needed: [hardware store] {+proj:hardware .proj:Resource} for screws and bike pump, and [GitHub repo] {+proj:repo .proj:Repository} for source code and documentation.
12
+
13
+ Team: [Alice] {+proj:alice ?prov:agent} as project manager with planning skill, [Bob] {+proj:bob ?prov:agent .prov:Person} as helper with bike repair skill.
14
+
15
+ The project has a [rain forecast] {+proj:rain !proj:blocker .proj:Blocker} blocker, with [indoor work] {+proj:indoor ?proj:alternative} as an alternative.
@@ -0,0 +1,31 @@
1
+ [food] <tag:alice@example.com,2026:food/>
2
+
3
+ ## Apple Pie {=food:pie .prov:Entity label}
4
+
5
+ Grandma Rose's recipe, documented by Alice. Cook time: [45] {food:time ^^xsd:integer} min, serves [8] {food:serves ^^xsd:integer}.
6
+
7
+ **Ingredients**
8
+
9
+ Flour (cups): [2.5] {+#flour ?food:ingredient .food:Ingredient food:ammount ^^food:cup} - used in crust
10
+
11
+ Sugar (cups): `1` {+#sugar ?food:ingredient .food:Ingredient food:ammount ^^food:cup} used in filling
12
+
13
+ Butter (cups): *1* {+#butter ?food:ingredient .food:Ingredient food:ammoun ^^food:cupt} used in crust
14
+
15
+ Apples: **6** {+#apples ?food:ingredient .food:Ingredient food:ammount ^^xsd:integer} used in filling
16
+
17
+ Cinnamon (tsp): __2__ {+food:cinnamon ?food:ingredient .food:Ingredient ^^xsd:tsp} used in filling
18
+
19
+ **Process**
20
+
21
+ **Crust** {=food:crust .prov:Activity} mixed [flour] {+food:pie#flour ?prov:used} and [butter] {+food:pie#butter}, generated [dough] {+food:dough ?prov:generated}. Used [bowl] {+food:bowl ?prov:used}.
22
+
23
+ **Filling** {=food:filling .prov:Activity ?prov:wasInformedBy} prepared apples with sugar and cinnamon, generated [apple mix] {+food:mix ?prov:generated}.
24
+
25
+ **Baking** {=food:bake .prov:Activity ?prov:wasInformedBy} baked at 375°F for 45 min. Used [oven] {+food:oven ?prov:used}. Generated [baked pie] {+food:baked ?prov:generated}.
26
+
27
+ We got the **Pie** {=food:pie}
28
+
29
+ The final pie was derived from [dough] {+food:dough ?prov:wasDerivedFrom} and [apple mix] {+food:mix ?prov:wasDerivedFrom}.
30
+
31
+ **Related**: Apple Crumble {+food:crumble !food:variation .food:Recipe}, Peach Pie {+food:peach !food:variation .food:Recipe}
@@ -0,0 +1,46 @@
1
+ [research] <https://research.example.org/>
2
+ [data] <https://research.example.org/datasets/>
3
+
4
+ # Machine Learning in Healthcare {=research:paper .prov:Entity label}
5
+
6
+ **Authors**: Dr. Sarah Martinez and Dr. James Liu, MIT CS Lab.
7
+
8
+ **Abstract**
9
+
10
+ This paper analyzes attention-based neural networks for medical imaging, citing prior work on deep learning in radiology.
11
+
12
+ **Methodology**
13
+
14
+ **Dataset** {=data:dataset .research:Dataset}
15
+ Source: GitHub medical imaging repository
16
+ Training: [80k] {+data:train ?research:detail .research:DatasetDetail} DICOM images, 15 classes
17
+ Validation: [10k] {+data:val ?research:detail .research:DatasetDetail} DICOM images
18
+ Test: [10k] {+data:test ?research:detail .research:DatasetDetail} DICOM images
19
+
20
+ **Kaggle Dataset** {=data:kaggle .research:Dataset} derived from primary dataset, featuring 512x512 grayscale images with patient metadata.
21
+
22
+ **Model Training** {=research:training .prov:Activity}
23
+ Duration: Feb 1-15, 2024
24
+ Association: Dr. James Liu as ML Engineer
25
+ Plan: Training protocol v2.0
26
+ Used: ML Dataset and Kaggle Dataset
27
+ Generated: [trained model] {+research:model ?prov:generated}
28
+
29
+ **Results**
30
+
31
+ Statistical significance: p < 0.05, sample size: [1000] {research:n ^^xsd:integer} patients.
32
+
33
+ **Accuracy Metrics** {=research:metrics .prov:Entity}
34
+ Training: [94.5%] {research:trainAcc}
35
+ Validation: [92.3%] {research:valAcc}
36
+ Test: [91.8%] {research:testAcc}
37
+ F1: [0.89] {research:f1 ^^xsd:decimal}
38
+ AUC-ROC: [0.93] {research:auc ^^xsd:decimal}
39
+
40
+ **SHACL Validation** {=research:shape .sh:NodeShape}
41
+ Target: Scientific Paper
42
+ Title: exactly 1 required
43
+ Author: at least 1 required
44
+ Citations: at least 1 required
45
+
46
+ **Funding**: NASA Space Medicine Research Initiative, grant NRA-2024-12345.
@@ -0,0 +1,48 @@
1
+ [nasa] <http://nasa.gov/>
2
+
3
+ # Apollo 11 Mission {=nasa:apollo11 .prov:Activity label}
4
+
5
+ ## METADATA
6
+
7
+ Launch: [1969-07-16T13:32:00Z] {prov:startedAtTime ^^xsd:dateTime}
8
+ Landing: [1969-07-20T20:17:00Z] {nasa:landing ^^xsd:dateTime}
9
+ Duration (days): [8] {nasa:durationDays ^^xsd:integer}
10
+ Budget (B USD): [2.5] {nasa:budget ^^xsd:decimal}
11
+ Operator: [NASA] {+nasa:agency ?prov:wasAssociatedWith .prov:Organization}
12
+
13
+ ## CREW
14
+
15
+ **Neil Armstrong** {+nasa:armstrong .prov:Person ?nasa:crew} - Commander, 1930
16
+ **Buzz Aldrin** {+nasa:aldrin .prov:Person ?nasa:crew} - LMP, 1930
17
+ **Michael Collins** {+nasa:collins .prov:Person ?nasa:crew} - CMP, 1930
18
+
19
+ ## EQUIPMENT
20
+
21
+ **Saturn V** {+nasa:saturnv .nasa:Rocket ?nasa:equipment} - Boeing, 363ft
22
+ **LM Eagle** {+nasa:eagle .nasa:Module ?nasa:equipment} - Grumman
23
+ **CM Columbia** {+nasa:columbia .nasa:Module ?nasa:equipment} - NAA
24
+
25
+ ## ACTIVITIES
26
+
27
+ **Launch** {+nasa:launch .prov:Activity} - Used Saturn V, generated orbit trajectory
28
+
29
+ **Lunar Landing** {+nasa:landing .prov:Activity} - Used LM Eagle, generated lunar presence and samples
30
+
31
+ **EVA** {+nasa:eva .prov:Activity} - 2h 18m, 250m, 21.5kg samples collected
32
+
33
+ ## OBJECTIVES {=nasa:appollo11}
34
+
35
+ **Land on Moon** {+nasa:goal1 .nasa:Goal} - ACHIEVED 1969-07-20
36
+ **Return safely** {+nasa:goal2 .nasa:Goal} - ACHIEVED 1969-07-24
37
+ **Collect samples** {+nasa:goal3 .nasa:Goal} - ACHIEVED 21.5kg
38
+
39
+ ## RELATIONSHIPS
40
+
41
+ Preceded by: [Apollo 10] {+nasa:apollo10 !nasa:precededBy .nasa:SpaceMission}, [Gemini 8] {+nasa:gemini8 !nasa:precededBy .nasa:SpaceMission}
42
+
43
+ Succeeded by: [Apollo 12] {+nasa:apollo12 ?nasa:succeededBy .nasa:SpaceMission}
44
+
45
+ **Delegation** {=nasa:delegation .prov:Delegation ?prov:qualifiedDelegation}
46
+
47
+ Delegate: [Mission Control] {+nasa:mcc ?prov:agent .prov:Organization}
48
+ Responsible: [NASA Admin] {+nasa:admin ?prov:agent .prov:Person}
@@ -0,0 +1,288 @@
1
+ [mdld] <https://mdld.js.org/>
2
+ [xsd] <http://www.w3.org/2001/XMLSchema#>
3
+ [@vocab] <http://www.w3.org/2000/01/rdf-schema#>
4
+ [my] <tag:mdld@example.com,2026:>
5
+ [ex] <my:example:>
6
+
7
+ # MD-LD: Complete Syntax Specification {=mdld:dogfood .Class label}
8
+
9
+ > MD-LD is a **semantic annotation layer** for CommonMark Markdown that creates RDF knowledge graphs from explicit `{...}` annotations. {mdld:dogfood comment}
10
+
11
+ ## Default Context {=mdld:default-context .Class label}
12
+
13
+ MD-LD provides these built-in prefixes and vocabulary:
14
+
15
+ Default vocabulary and prefixes: {mdld:includes .Class label}
16
+ - `@vocab`: `http://www.w3.org/2000/01/rdf-schema#` {=mdld:vocab}
17
+ - `rdf`: `http://www.w3.org/1999/02/22-rdf-syntax-ns#` {=mdld:rdf-prefix}
18
+ - `rdfs`: `http://www.w3.org/2000/01/rdf-schema#` {=mdld:rdfs-prefix}
19
+ - `xsd`: `http://www.w3.org/2001/XMLSchema#` {=mdld:xsd-prefix}
20
+ - `sh`: `http://www.w3.org/ns/shacl#` {=mdld:sh-prefix}
21
+ - `prov`: `http://www.w3.org/ns/prov#` {=mdld:prov-prefix}
22
+
23
+ ## Prefix Folding {=mdld:prefix-folding .Container label}
24
+
25
+ MD-LD supports hierarchical namespace building through prefix folding:
26
+
27
+ Base namespace declaration:
28
+ - `[my] <tag:mdld@example.com,2026:>` {+mdld:base-namespace ?member}
29
+
30
+ Derived namespaces using CURIE syntax:
31
+ - `[ex] <my:example:>` {+mdld:derived-namespace ?member}
32
+
33
+ This resolves to: `tag:mdld@example.com,2026:example:`
34
+
35
+ **Resolution Rules:**
36
+ - Prefixes must be declared before they can be referenced (forward-reference only)
37
+ - Circular references are treated as literal strings
38
+ - Later declarations override earlier ones
39
+
40
+ **Examples:**
41
+ - `ex:doc` → `tag:mdld@example.com,2026:example:doc`
42
+ - `my:journal:2026-01-27` → `tag:mdld@example.com,2026:journal:2026-01-27`
43
+
44
+ ### Prefix Folding Examples
45
+ Demonstrated namespaces:
46
+ - Document: [Example doc] {+ex:doc .Class label ?member}
47
+ - Feature: [Example feature] {+ex:feature .Class label ?member}
48
+ - Parser: [Example parser] {+ex:parser .Class label ?member}
49
+
50
+ ## Angle-Bracket URLs {=mdld:angle-urls .Container label}
51
+
52
+ MD-LD supports angle-bracket URLs as value carriers with soft subject behavior:
53
+
54
+ ### URL Value Carriers
55
+ External resources as soft subjects:
56
+ - <https://www.w3.org/TR/rdf11-concepts/> {.Specification label}
57
+ - <https://github.com/mdld-js/mdld-parse> {?mdld:implementation .Repository label}
58
+ - <https://arxiv.org/abs/2301.07041> {!mdld:cites .Paper label}
59
+
60
+ ### URL Behavior Rules
61
+ Angle-bracket URLs follow these rules:
62
+ - Type declarations use URL as subject: `<URL> {.Type}`
63
+ - Object predicates use current subject: `<URL> {?predicate}`
64
+ - Reverse predicates use URL as subject: `<URL> {!predicate}`
65
+ - Literal predicates are ignored: `<URL> {literal}` (no output)
66
+
67
+ ## Core Classes {=mdld:classes .Container label}
68
+
69
+ All classes used in this document:
70
+
71
+ - Document {+mdld:Document ?member .Class label}
72
+ - Concept {+mdld:Concept ?member .Class label}
73
+ - Resource {+mdld:Resource ?member .Class label}
74
+ - Principle {+mdld:Principle ?member .Class label}
75
+ - Feature {+mdld:Feature ?member .Class label}
76
+ - Layer {+mdld:Layer ?member .Class label}
77
+ - Component {+mdld:Component ?member .Class label}
78
+ - Software {+mdld:Software ?member .Class label}
79
+
80
+ ## Parser Architecture {=mdld:parser-arch .Container label}
81
+
82
+ ### Tokenization Phase
83
+ The parser processes documents in a single pass with these token types: {mdld:includes .Class label}
84
+
85
+ Token types created during scanning:
86
+ - `prefix` - Context declarations `[name] <IRI>`
87
+ - `heading` - Markdown headings `# text {attrs}`
88
+ - `list` - List items `- text {attrs}` or `1. text {attrs}`
89
+ - `blockquote` - Blockquotes `> text {attrs}`
90
+ - `code` - Code blocks with annotations
91
+ - `para` - Paragraphs with potential inline carriers
92
+
93
+ ### Processing Pipeline
94
+ Document processing follows this pipeline:
95
+ 1. **Line-by-line scanning** {+#line-by-line .Class label ?member}
96
+ 2. **Context resolution** {+#context-resolution .Class label ?member}
97
+ 3. **Subject tracking** {+#subject-tracking .Class label ?member}
98
+ 4. **Annotation processing** {+#annotation-processing .Class label ?member}
99
+ 5. **Quad emission** {+#quad-emission .Class label ?member}
100
+
101
+ ### Semantic Block Processing
102
+ Annotations are parsed using cached semantic blocks: {mdld:includes .Class label}
103
+ - Parse `{...}` blocks into structured semantic objects
104
+ - Cache frequently used patterns for performance
105
+ - Extract subjects, predicates, types, and modifiers
106
+ - Handle all three predicate forms (`p`, `?p`, `!p`)
107
+
108
+ ## Context Declarations {=mdld:context .Container label}
109
+
110
+ MD-LD supports prefix and vocabulary declarations following JSON-LD patterns:
111
+
112
+ Prefix declarations:
113
+ - `[mdld] <https://mdld.js.org/>` {+mdld:prefix .Class label ?member}
114
+ - `[xsd] <http://www.w3.org/2001/XMLSchema#>` {+mdld:xsd .Class label ?member}
115
+ - `[@vocab] <http://www.w3.org/2000/01/rdf-schema#>` {+mdld:vocab .Class label ?member}
116
+
117
+ ## Core Principles {=mdld:principles .Container label}
118
+
119
+ MD-LD follows strict rules:
120
+
121
+ **Markdown-preserving** {+#markdown-preserving ?member label} - Remove `{...}` → valid Markdown
122
+ **Explicit only** {+#explicit-only ?member label} - No implicit semantics or guessing
123
+ **Single-pass** {+#single-pass ?member label} - Streaming-friendly processing
124
+ **Deterministic** {+#deterministic ?member label} - Same input → same output
125
+ **Traceable** {+#traceable ?member label} - Every fact traces to its source
126
+
127
+ ## Subject Management {=mdld:subjects .Container label}
128
+
129
+ ### Full IRI Subjects
130
+ [Document with full IRI] {+mdld:doc-full ?member .Class label} demonstrates persistent subjects
131
+
132
+ ### Fragment Subjects
133
+ [Section with fragment] {+#fragment-example ?member .Class label} creates relative identifiers
134
+
135
+ ### Subject Reset
136
+ [{=}] {+mdld:reset ?member label} shows subject clearing
137
+
138
+ ### Temporary Objects
139
+ [Related concept] {+mdld:temporary ?member label} demonstrates block-scoped objects
140
+
141
+ ### Soft Fragments
142
+ [Soft fragment example] {+#soft-fragment ?member label} demonstrates temporary fragments
143
+
144
+ **Use cases:**
145
+ - `[Related Item] {+ex:related ?schema:isRelatedTo}` - Links to a local fragment
146
+ - `[Parent] {+ex:parent !schema:hasPart}` - Reverse relationship
147
+ - `[Section] {+#section1 name ?hasPart}` - Object property with fragment
148
+
149
+ The soft IRI only exists within the current annotation block.
150
+
151
+ ## Value Carriers {=mdld:carriers .Container label}
152
+
153
+ ### Inline Elements
154
+ **Inline elements:** [*italic*] {+mdld:italic ?member} [**bold**] {+mdld:bold ?member} [`code`] {+mdld:code ?member}
155
+
156
+ ### Block Elements
157
+ **Block elements:** # [Headings] {+mdld:heading ?member} > [Quotes] {+mdld:quote ?member}
158
+
159
+ ### Links and Media
160
+ **Links:** [MD-LD Website](https://mdld.js.org) {?member}
161
+ **Images:** ![Logo](https://mdld.js.org/logo.png) {?member}
162
+
163
+ ## Predicate Forms {=mdld:predicates .Class label}
164
+
165
+ Three predicate directions create different relationships:
166
+
167
+ - **Subject→Literal**: [MD-LD] {label} creates `mdld:doc label "MD-LD"`
168
+ - **Subject→Object**: [RDF](https://www.w3.org/RDF) {mdld:seeAlso} creates `mdld:doc rdfs:seeAlso <https://www.w3.org/RDF>`
169
+ - **Object→Subject**: [Example] {mdld:isDefinedBy} creates `mdld:example mdld:isDefinedBy mdld:doc`
170
+
171
+ ## Type Declarations {=mdld:types .Container label}
172
+
173
+ ### Single Types
174
+ [Single type example] {+mdld:single ?member .Class label}
175
+
176
+ ### Multiple Types
177
+ [Multiple types example] {+mdld:multiple ?member .Class .mdld:Class1 label}
178
+
179
+ ## Literal Modifiers {=mdld:literals .Container label}
180
+
181
+ ### Typed Literals
182
+ [2024] {mdld:year ^^xsd:gYear}
183
+ [3.14159] {mdld:pi ^^xsd:decimal}
184
+ [true] {mdld:bool ^^xsd:boolean}
185
+
186
+ ### Language-Tagged Literals
187
+ [Hello] {mdld:english @en}
188
+ [Bonjour] {mdld:french @fr}
189
+ [Hola] {mdld:spanish @es}
190
+
191
+ ## Lists and Nesting {=mdld:lists .Container label}
192
+
193
+ Lists are pure Markdown structure with **no semantic scope**. Each list item requires explicit annotations.
194
+
195
+ ### List Item Requirements
196
+
197
+ **Critical Rule**: All list items must have explicit subject (`{=iri}` or `{+iri}`) to emit semantics.
198
+
199
+ ### Simple Lists
200
+ Research team members:
201
+
202
+ - **Principal investigator** {+#pi ?member .mdld:Role label}
203
+ - **Research assistant** {+#assistant ?member .mdld:Role label}
204
+ - **Lab technician** {+#tech ?member .mdld:Role label}
205
+
206
+
207
+ ## Code Blocks {=mdld:code-blocks .Container label}
208
+
209
+ ### Annotated Code Blocks
210
+ ```javascript {=mdld:example-code .Class text}
211
+ console.log("MD-LD creates RDF from Markdown");
212
+ ```
213
+
214
+ ### Parser Implementation Examples
215
+ ```javascript {=mdld:parser-code .Class text}
216
+ // Token scanning implementation
217
+ function scanTokens(text) {
218
+ const tokens = [];
219
+ const lines = text.split('\n');
220
+ // Process each line for tokens...
221
+ }
222
+ ```
223
+
224
+ ### EBNF Grammar Examples
225
+ ```txt {=mdld:ebnf-grammar .Class text}
226
+ attrsBlock = "{" , whitespace* , attrsTokens? , whitespace* , "}" ;
227
+ attrsTokens = attrsToken , { whitespace+ , attrsToken } ;
228
+ ```
229
+
230
+ This *JavaScript* {=mdld:js label} code shows annotation syntax.
231
+
232
+ ## Blockquotes {=mdld:blockquote .Class label}
233
+
234
+ > Blockquotes serve as carriers for normative statements and comments. {mdld:blockquote comment}
235
+
236
+ > **Important**: All semantics must be explicit - no implicit inference. {mdld:normative comment}
237
+
238
+ > **Parser Note**: The implementation uses token-based scanning, not AST building. {mdld:parser-note comment}
239
+
240
+ ## Advanced Features {=mdld:advanced .Class label}
241
+
242
+ ### Soft Fragments
243
+ [Soft fragment example] {+#soft-fragment mdld:seeAlso label} demonstrates temporary fragments
244
+
245
+ ### Complex Annotations
246
+ [Complex example] {=mdld:complex .Class label comment mdld:seeAlso} shows multiple predicates
247
+
248
+ ### Reverse Relationships
249
+ [Target entity] {!mdld:hasPart label} demonstrates reverse predicate form
250
+
251
+ ## Processing Model {=mdld:processing .Class label}
252
+
253
+ ### Streaming Processing
254
+ MD-LD processes documents in a single pass: {?mdld:includes .Class label}
255
+ - Line-by-line tokenization {=#line-by-line}
256
+ - Subject tracking {=#subject-tracking}
257
+ - Quad emission {=#quad-emission}
258
+
259
+ ### Origin Tracking
260
+ Every quad maintains origin information for round-trip serialization:
261
+ - Block identification
262
+ - Position tracking
263
+ - Slot indexing
264
+
265
+ ### Semantic Block Caching
266
+ The parser uses caching for performance: {mdld:includes .Class label}
267
+ - Parse semantic blocks once and cache results
268
+ - Avoid repeated parsing of identical annotations
269
+ - Maintain deterministic output while improving speed
270
+
271
+ ## Error Handling {=mdld:errors .Class label}
272
+
273
+ ### Graceful Degradation
274
+ MD-LD recovers from malformed annotations:
275
+ - Unknown prefixes → skip annotation
276
+ - Malformed IRIs → skip annotation
277
+ - Unbalanced braces → fatal error
278
+ - Invalid carriers → skip annotation
279
+
280
+ ## Connected Knowledge Graph {=mdld:graph .Class label}
281
+
282
+ This document demonstrates how MD-LD creates a **self-describing knowledge graph** where:
283
+ - The document explains MD-LD using MD-LD
284
+ - Every concept is explicitly typed and related
285
+ - The resulting RDF graph mirrors the document structure
286
+ - All features connect into a coherent semantic model
287
+
288
+ > The document itself becomes both specification and proof of concept. { comment}