@feelpp/antora-extensions 1.0.0-rc.2 → 1.0.0-rc.5
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.
- package/README.md +59 -0
- package/package.json +11 -1
- package/src/index.js +4 -0
- package/src/jupyter.js +277 -0
- package/src/listing.js +1 -1
- package/src/lunr.js +145 -0
- package/tests/fixtures/jupyter-contract/content/antora.yml +5 -0
- package/tests/fixtures/jupyter-contract/content/modules/ROOT/nav.adoc +2 -0
- package/tests/fixtures/jupyter-contract/content/modules/ROOT/pages/foundations/sample.adoc +28 -0
- package/tests/fixtures/jupyter-contract/content/modules/ROOT/pages/plain.adoc +3 -0
- package/tests/fixtures/jupyter-contract/expected.json +16 -0
- package/tests/fixtures/jupyter-contract/failures.json +30 -0
- package/tests/jupyter-contract-test.js +152 -0
- package/tests/lunr-test.js +36 -0
package/README.md
CHANGED
|
@@ -1,3 +1,62 @@
|
|
|
1
1
|
# Antora Extensions by Feel++
|
|
2
2
|
|
|
3
|
+

|
|
4
|
+
|
|
3
5
|
A set of Antora extensions.
|
|
6
|
+
|
|
7
|
+
## Extensions
|
|
8
|
+
|
|
9
|
+
### Lunr Search Index
|
|
10
|
+
|
|
11
|
+
Automatically generates a Lunr.js compatible search index during the Antora build process.
|
|
12
|
+
|
|
13
|
+
#### Usage
|
|
14
|
+
|
|
15
|
+
Add the extension to your `site.yml`:
|
|
16
|
+
|
|
17
|
+
```yaml
|
|
18
|
+
antora:
|
|
19
|
+
extensions:
|
|
20
|
+
- '@feelpp/antora-extensions'
|
|
21
|
+
|
|
22
|
+
config:
|
|
23
|
+
lunr:
|
|
24
|
+
indexFile: 'search-index.json' # Output filename (optional)
|
|
25
|
+
maxContentLength: 1000 # Max content per document (optional)
|
|
26
|
+
minContentLength: 50 # Min content to include document (optional)
|
|
27
|
+
debug: false # Enable debug logging (optional)
|
|
28
|
+
```
|
|
29
|
+
|
|
30
|
+
The search index will be automatically generated at `build/site/search-index.json` after the site is built.
|
|
31
|
+
|
|
32
|
+
#### Features
|
|
33
|
+
|
|
34
|
+
- **Zero configuration**: Works out of the box with sensible defaults
|
|
35
|
+
- **Smart content extraction**: Focuses on main content, excludes navigation and footers
|
|
36
|
+
- **Configurable**: Customize content length limits and output location
|
|
37
|
+
- **Performance optimized**: Efficient processing during site generation
|
|
38
|
+
- **Error handling**: Graceful handling of malformed HTML or missing content
|
|
39
|
+
|
|
40
|
+
## Listing
|
|
41
|
+
|
|
42
|
+
### UI assumptions
|
|
43
|
+
|
|
44
|
+
This extension relies on a contract with the UI in order to minimize the configuration the user must perform to get the extension working.
|
|
45
|
+
|
|
46
|
+
#### Environment variable
|
|
47
|
+
|
|
48
|
+
When this extension is enabled, it sets the `SITE_LISTING_EXTENSION_ENABLED` environment variable to the value `true`.
|
|
49
|
+
This variable is available to the UI templates as `env.SITE_LISTING_EXTENSION_ENABLED`.
|
|
50
|
+
The existence of this variable informs the UI template that the listing extension is enabled.
|
|
51
|
+
When this variable is set, the UI is expected to add certain elements to support the extension.
|
|
52
|
+
|
|
53
|
+
#### Listing styles
|
|
54
|
+
|
|
55
|
+
This package provides additional CSS to style the listing results (`data/css/listing.css`).
|
|
56
|
+
The creator of the UI can either bundle those styles or reference them (for instance in `head-styles.hbs`):
|
|
57
|
+
|
|
58
|
+
```hbs
|
|
59
|
+
{{#if env.SITE_LISTING_EXTENSION_ENABLED}}
|
|
60
|
+
<link rel="stylesheet" href="{{{uiRootPath}}}/css/listing.css">
|
|
61
|
+
{{/if}}
|
|
62
|
+
```
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@feelpp/antora-extensions",
|
|
3
3
|
"private": false,
|
|
4
|
-
"version": "1.0.0-rc.
|
|
4
|
+
"version": "1.0.0-rc.5",
|
|
5
5
|
"description": "Antora Extensions by Feel++.",
|
|
6
6
|
"main": "src/index.js",
|
|
7
7
|
"repository": {
|
|
@@ -11,12 +11,22 @@
|
|
|
11
11
|
"keywords": [],
|
|
12
12
|
"author": "Guillaume Grossetie <ggrossetie@yuzutech.fr>",
|
|
13
13
|
"license": "MIT",
|
|
14
|
+
"scripts": {
|
|
15
|
+
"test": "mocha tests/*-test.js"
|
|
16
|
+
},
|
|
14
17
|
"bugs": {
|
|
15
18
|
"url": "https://github.com/feelpp/feelpp-antora-extensions/issues"
|
|
16
19
|
},
|
|
17
20
|
"homepage": "https://github.com/feelpp/feelpp-antora-extensions#readme",
|
|
18
21
|
"devDependencies": {
|
|
22
|
+
"chai": "^4.3.7",
|
|
19
23
|
"libnpmpublish": "^4.0.2",
|
|
24
|
+
"mocha": "^10.2.0",
|
|
20
25
|
"pacote": "^12.0.2"
|
|
26
|
+
},
|
|
27
|
+
"dependencies": {
|
|
28
|
+
"@asciidoctor/core": "2.2.9",
|
|
29
|
+
"asciidoctor-jupyter": "0.7.0",
|
|
30
|
+
"jsdom": "^20.0.3"
|
|
21
31
|
}
|
|
22
32
|
}
|
package/src/index.js
CHANGED
|
@@ -1,7 +1,11 @@
|
|
|
1
1
|
const {register: registerTooboxNavigation} = require('./toolbox-navigation.js')
|
|
2
2
|
const {register: registerListing} = require('./listing.js')
|
|
3
|
+
const {register: registerLunr} = require('./lunr.js')
|
|
4
|
+
const {register: registerJupyter} = require('./jupyter.js')
|
|
3
5
|
|
|
4
6
|
module.exports.register = (thisContext, args) => {
|
|
5
7
|
registerTooboxNavigation.call(thisContext, args)
|
|
6
8
|
registerListing.call(thisContext, args)
|
|
9
|
+
registerLunr.call(thisContext, args)
|
|
10
|
+
registerJupyter.call(thisContext, args)
|
|
7
11
|
}
|
package/src/jupyter.js
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
1
|
+
const crypto = require('node:crypto')
|
|
2
|
+
const path = require('node:path').posix
|
|
3
|
+
|
|
4
|
+
const asciidoctor = require('@asciidoctor/core')()
|
|
5
|
+
const JupyterConverter = require('asciidoctor-jupyter')
|
|
6
|
+
const extensionPackage = require('../package.json')
|
|
7
|
+
const converterPackage = require('asciidoctor-jupyter/package.json')
|
|
8
|
+
|
|
9
|
+
asciidoctor.ConverterFactory.register(JupyterConverter, ['jupyter'])
|
|
10
|
+
|
|
11
|
+
const DEFAULT_FIGURE_CONTRACT = 'alt text, caption, labelled axes and units, and non-colour encoding'
|
|
12
|
+
|
|
13
|
+
function sha256 (value) {
|
|
14
|
+
return crypto.createHash('sha256').update(value).digest('hex')
|
|
15
|
+
}
|
|
16
|
+
|
|
17
|
+
function parseHeaderAttributes (source) {
|
|
18
|
+
const attributes = {}
|
|
19
|
+
const lines = source.split(/\r?\n/)
|
|
20
|
+
let titleSeen = false
|
|
21
|
+
for (const line of lines) {
|
|
22
|
+
if (!titleSeen) {
|
|
23
|
+
if (line.startsWith('= ')) titleSeen = true
|
|
24
|
+
continue
|
|
25
|
+
}
|
|
26
|
+
if (line === '' || line.startsWith('//')) continue
|
|
27
|
+
const match = line.match(/^:([^:]+):(?:\s*(.*))?$/)
|
|
28
|
+
if (!match) break
|
|
29
|
+
attributes[match[1]] = match[2] || ''
|
|
30
|
+
}
|
|
31
|
+
return attributes
|
|
32
|
+
}
|
|
33
|
+
|
|
34
|
+
function listAttribute (attributes, name) {
|
|
35
|
+
return (attributes[name] || '')
|
|
36
|
+
.split(',')
|
|
37
|
+
.map((value) => value.trim())
|
|
38
|
+
.filter(Boolean)
|
|
39
|
+
}
|
|
40
|
+
|
|
41
|
+
function resourceId ({ component, version, module, relative }, family = 'page') {
|
|
42
|
+
const prefix = version ? `${version}@${component}` : component
|
|
43
|
+
const familyMarker = family === 'page' ? '' : `${family}$`
|
|
44
|
+
return `${prefix}:${module}:${familyMarker}${relative}`
|
|
45
|
+
}
|
|
46
|
+
|
|
47
|
+
function matchesInclude (relative, include) {
|
|
48
|
+
if (!include || include.length === 0) return true
|
|
49
|
+
return include.some((pattern) => {
|
|
50
|
+
const escaped = pattern.replace(/[.+?^${}()|[\]\\]/g, '\\$&').replaceAll('*', '.*')
|
|
51
|
+
return new RegExp(`^${escaped}$`).test(relative)
|
|
52
|
+
})
|
|
53
|
+
}
|
|
54
|
+
|
|
55
|
+
function isJupyterPage (page, options = {}) {
|
|
56
|
+
const source = page.contents.toString('utf8')
|
|
57
|
+
const attributes = parseHeaderAttributes(source)
|
|
58
|
+
return Object.hasOwn(attributes, 'page-jupyter') && matchesInclude(page.src.relative, options.include)
|
|
59
|
+
}
|
|
60
|
+
|
|
61
|
+
function notebookMetadata (page, source, attributes, options) {
|
|
62
|
+
const required = [
|
|
63
|
+
'page-course-language',
|
|
64
|
+
'page-course-priority',
|
|
65
|
+
'page-course-difficulty',
|
|
66
|
+
'page-course-duration-minutes',
|
|
67
|
+
'page-course-concepts',
|
|
68
|
+
'page-course-outcomes'
|
|
69
|
+
]
|
|
70
|
+
const missing = required.filter((name) => !attributes[name])
|
|
71
|
+
if (missing.length) {
|
|
72
|
+
throw new Error(`${resourceId(page.src)}: missing notebook metadata: ${missing.join(', ')}`)
|
|
73
|
+
}
|
|
74
|
+
if (attributes['page-course-language'] !== 'en') {
|
|
75
|
+
throw new Error(`${resourceId(page.src)}: generated notebooks must use English`)
|
|
76
|
+
}
|
|
77
|
+
const duration = Number.parseInt(attributes['page-course-duration-minutes'], 10)
|
|
78
|
+
if (!Number.isInteger(duration) || duration < 1) {
|
|
79
|
+
throw new Error(`${resourceId(page.src)}: invalid page-course-duration-minutes`)
|
|
80
|
+
}
|
|
81
|
+
return {
|
|
82
|
+
schema_version: 1,
|
|
83
|
+
source_kind: 'asciidoc',
|
|
84
|
+
source_resource_id: resourceId(page.src),
|
|
85
|
+
source_path: page.src.relative,
|
|
86
|
+
source_sha256: sha256(source),
|
|
87
|
+
source_page_url: page.pub.url,
|
|
88
|
+
language: 'en',
|
|
89
|
+
priority: attributes['page-course-priority'],
|
|
90
|
+
difficulty: attributes['page-course-difficulty'],
|
|
91
|
+
duration_minutes: duration,
|
|
92
|
+
concept_ids: listAttribute(attributes, 'page-course-concepts'),
|
|
93
|
+
outcomes: listAttribute(attributes, 'page-course-outcomes'),
|
|
94
|
+
execution_profile: attributes['page-course-execution-profile'] || 'fast',
|
|
95
|
+
accessibility: {
|
|
96
|
+
keyboard_only: true,
|
|
97
|
+
colour_alone_forbidden: true,
|
|
98
|
+
required_figure_contract: attributes['page-course-figure-contract'] || DEFAULT_FIGURE_CONTRACT
|
|
99
|
+
},
|
|
100
|
+
generator: {
|
|
101
|
+
asciidoctor_jupyter: converterPackage.version,
|
|
102
|
+
feelpp_antora_extensions: extensionPackage.version,
|
|
103
|
+
feelpp_asciidoctor_extensions: options.asciidoctorExtensionVersion || '1.0.0-rc.17'
|
|
104
|
+
}
|
|
105
|
+
}
|
|
106
|
+
}
|
|
107
|
+
|
|
108
|
+
function normaliseMarkdown (source, page) {
|
|
109
|
+
return source
|
|
110
|
+
.replace(
|
|
111
|
+
/(^|\n)\+\n(\\[\s\S]*?)\n\+(?=\n|$)/g,
|
|
112
|
+
(_match, prefix, latex) => `${prefix}$$\n${latex}\n$$`
|
|
113
|
+
)
|
|
114
|
+
.replace(/\]\(attachment\$([^)]+)\)/g, (_match, target) => {
|
|
115
|
+
const relative = path.relative(path.dirname(page.src.relative), target)
|
|
116
|
+
return `](${relative || path.basename(target)})`
|
|
117
|
+
})
|
|
118
|
+
}
|
|
119
|
+
|
|
120
|
+
function normaliseNotebook (notebook, page, source, attributes, options = {}) {
|
|
121
|
+
notebook.nbformat = 4
|
|
122
|
+
notebook.nbformat_minor = Math.max(notebook.nbformat_minor || 0, 5)
|
|
123
|
+
notebook.metadata = notebook.metadata || {}
|
|
124
|
+
notebook.metadata.kernelspec = notebook.metadata.kernelspec || {}
|
|
125
|
+
notebook.metadata.kernelspec.name = notebook.metadata.kernelspec.name || 'python3'
|
|
126
|
+
notebook.metadata.kernelspec.language = notebook.metadata.kernelspec.language || 'python'
|
|
127
|
+
notebook.metadata.kernelspec.display_name = notebook.metadata.kernelspec.display_name || 'Python 3'
|
|
128
|
+
notebook.metadata.course = notebookMetadata(page, source, attributes, options)
|
|
129
|
+
notebook.cells.forEach((cell, index) => {
|
|
130
|
+
let cellSource = Array.isArray(cell.source) ? cell.source.join('') : cell.source || ''
|
|
131
|
+
if (cell.cell_type === 'markdown') {
|
|
132
|
+
cellSource = normaliseMarkdown(cellSource, page)
|
|
133
|
+
cell.source = cellSource
|
|
134
|
+
}
|
|
135
|
+
cell.id = sha256(`${resourceId(page.src)}\0${index}\0${cell.cell_type}\0${cellSource}`).slice(0, 16)
|
|
136
|
+
if (cell.cell_type === 'code') {
|
|
137
|
+
cell.execution_count = null
|
|
138
|
+
cell.outputs = []
|
|
139
|
+
}
|
|
140
|
+
})
|
|
141
|
+
return notebook
|
|
142
|
+
}
|
|
143
|
+
|
|
144
|
+
function generateNotebook (page, options = {}) {
|
|
145
|
+
const source = page.contents.toString('utf8')
|
|
146
|
+
const attributes = parseHeaderAttributes(source)
|
|
147
|
+
let notebook
|
|
148
|
+
try {
|
|
149
|
+
notebook = JSON.parse(asciidoctor.convert(source, {
|
|
150
|
+
backend: 'jupyter',
|
|
151
|
+
safe: 'safe',
|
|
152
|
+
standalone: true,
|
|
153
|
+
to_file: false,
|
|
154
|
+
attributes: {
|
|
155
|
+
'jupyter-language-name': attributes['jupyter-language-name'] || 'python',
|
|
156
|
+
'jupyter-language-version': attributes['jupyter-language-version'] || '3.12',
|
|
157
|
+
'jupyter-kernel-name': attributes['jupyter-kernel-name'] || 'python3',
|
|
158
|
+
'jupyter-kernel-language': attributes['jupyter-kernel-language'] || 'python'
|
|
159
|
+
}
|
|
160
|
+
}))
|
|
161
|
+
} catch (error) {
|
|
162
|
+
throw new Error(`${resourceId(page.src)}: notebook conversion failed: ${error.message}`)
|
|
163
|
+
}
|
|
164
|
+
normaliseNotebook(notebook, page, source, attributes, options)
|
|
165
|
+
return Buffer.from(`${JSON.stringify(notebook, null, 2)}\n`)
|
|
166
|
+
}
|
|
167
|
+
|
|
168
|
+
function attachmentRelative (page) {
|
|
169
|
+
return page.src.relative.replace(/\.adoc$/, '.ipynb')
|
|
170
|
+
}
|
|
171
|
+
|
|
172
|
+
function manifestEntry (page, attachment, notebook) {
|
|
173
|
+
const metadata = notebook.metadata.course
|
|
174
|
+
const codeBlocks = notebook.cells
|
|
175
|
+
.map((cell, index) => ({ cell, index }))
|
|
176
|
+
.filter(({ cell }) => cell.cell_type === 'code')
|
|
177
|
+
.map(({ cell, index }) => ({
|
|
178
|
+
id: cell.id,
|
|
179
|
+
index,
|
|
180
|
+
source_sha256: sha256(Array.isArray(cell.source) ? cell.source.join('') : cell.source || ''),
|
|
181
|
+
dynamic: false,
|
|
182
|
+
required: true
|
|
183
|
+
}))
|
|
184
|
+
return {
|
|
185
|
+
source_resource_id: metadata.source_resource_id,
|
|
186
|
+
source_path: metadata.source_path,
|
|
187
|
+
source_sha256: metadata.source_sha256,
|
|
188
|
+
page_url: page.pub.url,
|
|
189
|
+
notebook_resource_id: resourceId(attachment.src, 'attachment'),
|
|
190
|
+
notebook_path: attachment.src.relative,
|
|
191
|
+
notebook_sha256: sha256(attachment.contents),
|
|
192
|
+
notebook_url: attachment.pub.url,
|
|
193
|
+
priority: metadata.priority,
|
|
194
|
+
difficulty: metadata.difficulty,
|
|
195
|
+
concept_ids: metadata.concept_ids,
|
|
196
|
+
outcomes: metadata.outcomes,
|
|
197
|
+
code_blocks: codeBlocks,
|
|
198
|
+
warnings: []
|
|
199
|
+
}
|
|
200
|
+
}
|
|
201
|
+
|
|
202
|
+
function buildManifest (entries) {
|
|
203
|
+
return Buffer.from(`${JSON.stringify({
|
|
204
|
+
schema_version: 1,
|
|
205
|
+
artifact: 'asciidoc-notebook-manifest',
|
|
206
|
+
generator: {
|
|
207
|
+
name: extensionPackage.name,
|
|
208
|
+
version: extensionPackage.version
|
|
209
|
+
},
|
|
210
|
+
entries: [...entries].sort((a, b) => a.source_resource_id.localeCompare(b.source_resource_id))
|
|
211
|
+
}, null, 2)}\n`)
|
|
212
|
+
}
|
|
213
|
+
|
|
214
|
+
function register ({ config = {} } = {}) {
|
|
215
|
+
const logger = this.getLogger('jupyter-extension')
|
|
216
|
+
const options = config.jupyter || {}
|
|
217
|
+
const generated = new Map()
|
|
218
|
+
|
|
219
|
+
this.on('contentClassified', ({ contentCatalog }) => {
|
|
220
|
+
const pages = contentCatalog.getPages().filter((page) => isJupyterPage(page, options))
|
|
221
|
+
.sort((a, b) => resourceId(a.src).localeCompare(resourceId(b.src)))
|
|
222
|
+
const entriesByComponent = new Map()
|
|
223
|
+
for (const page of pages) {
|
|
224
|
+
const contents = generateNotebook(page, options)
|
|
225
|
+
const attachment = contentCatalog.addFile({
|
|
226
|
+
contents,
|
|
227
|
+
src: {
|
|
228
|
+
component: page.src.component,
|
|
229
|
+
version: page.src.version,
|
|
230
|
+
module: page.src.module,
|
|
231
|
+
family: 'attachment',
|
|
232
|
+
relative: attachmentRelative(page)
|
|
233
|
+
}
|
|
234
|
+
})
|
|
235
|
+
const notebook = JSON.parse(contents.toString('utf8'))
|
|
236
|
+
const entry = manifestEntry(page, attachment, notebook)
|
|
237
|
+
const key = `${page.src.component}\0${page.src.version || ''}`
|
|
238
|
+
const group = entriesByComponent.get(key) || { page, entries: [] }
|
|
239
|
+
group.entries.push(entry)
|
|
240
|
+
entriesByComponent.set(key, group)
|
|
241
|
+
generated.set(page, attachment)
|
|
242
|
+
logger.info(`generated ${attachment.pub.url} from ${resourceId(page.src)}`)
|
|
243
|
+
}
|
|
244
|
+
for (const { page, entries } of entriesByComponent.values()) {
|
|
245
|
+
contentCatalog.addFile({
|
|
246
|
+
contents: buildManifest(entries),
|
|
247
|
+
src: {
|
|
248
|
+
component: page.src.component,
|
|
249
|
+
version: page.src.version,
|
|
250
|
+
module: 'ROOT',
|
|
251
|
+
family: 'attachment',
|
|
252
|
+
relative: 'generated/asciidoc-notebook-manifest.json'
|
|
253
|
+
}
|
|
254
|
+
})
|
|
255
|
+
}
|
|
256
|
+
})
|
|
257
|
+
|
|
258
|
+
this.on('documentsConverted', () => {
|
|
259
|
+
for (const [page, attachment] of generated) {
|
|
260
|
+
page.asciidoc.attributes['page-jupyter-url'] = attachment.pub.url
|
|
261
|
+
page.asciidoc.attributes['page-jupyter-status'] = 'generated'
|
|
262
|
+
}
|
|
263
|
+
})
|
|
264
|
+
}
|
|
265
|
+
|
|
266
|
+
module.exports = {
|
|
267
|
+
attachmentRelative,
|
|
268
|
+
buildManifest,
|
|
269
|
+
generateNotebook,
|
|
270
|
+
isJupyterPage,
|
|
271
|
+
manifestEntry,
|
|
272
|
+
normaliseMarkdown,
|
|
273
|
+
parseHeaderAttributes,
|
|
274
|
+
register,
|
|
275
|
+
resourceId,
|
|
276
|
+
sha256
|
|
277
|
+
}
|
package/src/listing.js
CHANGED
|
@@ -10,7 +10,7 @@ const template = require('./template')
|
|
|
10
10
|
/**
|
|
11
11
|
* @module listing-extension
|
|
12
12
|
*/
|
|
13
|
-
function register ({ config: { listing, ...unknownOptions } }) {
|
|
13
|
+
function register ({ config: { listing, jupyter: _jupyter, lunr: _lunr, ...unknownOptions } }) {
|
|
14
14
|
const packageName = "@feelpp/antora-listing-extension"
|
|
15
15
|
const logger = this.getLogger(packageName)
|
|
16
16
|
|
package/src/lunr.js
ADDED
|
@@ -0,0 +1,145 @@
|
|
|
1
|
+
/**
|
|
2
|
+
* Antora extension for automatic Lunr.js search index generation
|
|
3
|
+
*
|
|
4
|
+
* This extension automatically generates a Lunr.js compatible search index
|
|
5
|
+
* after the site is published. It extracts content from all HTML files
|
|
6
|
+
* and creates a JSON index that can be used for client-side search.
|
|
7
|
+
*/
|
|
8
|
+
|
|
9
|
+
const fs = require('fs')
|
|
10
|
+
const path = require('path')
|
|
11
|
+
|
|
12
|
+
// Lazy load JSDOM only when needed
|
|
13
|
+
let JSDOM
|
|
14
|
+
function loadJSDOM() {
|
|
15
|
+
if (!JSDOM) {
|
|
16
|
+
try {
|
|
17
|
+
JSDOM = require('jsdom').JSDOM
|
|
18
|
+
} catch (error) {
|
|
19
|
+
throw new Error('JSDOM is required for the Lunr extension. Install it with: npm install jsdom')
|
|
20
|
+
}
|
|
21
|
+
}
|
|
22
|
+
return JSDOM
|
|
23
|
+
}
|
|
24
|
+
|
|
25
|
+
function register({ config, playbook }) {
|
|
26
|
+
const logger = this.getLogger('lunr-extension')
|
|
27
|
+
|
|
28
|
+
this.once('sitePublished', (event) => {
|
|
29
|
+
logger.info('Generating Lunr.js search index...')
|
|
30
|
+
|
|
31
|
+
try {
|
|
32
|
+
// Get output directory from the playbook
|
|
33
|
+
const outputDir = playbook.output?.dir || 'build/site'
|
|
34
|
+
logger.info(`Output directory: ${outputDir}`)
|
|
35
|
+
|
|
36
|
+
generateSearchIndex(outputDir, config.lunr || {}, logger)
|
|
37
|
+
} catch (error) {
|
|
38
|
+
logger.error('Failed to generate search index:', error)
|
|
39
|
+
throw error
|
|
40
|
+
}
|
|
41
|
+
})
|
|
42
|
+
}
|
|
43
|
+
|
|
44
|
+
function generateSearchIndex(outputDir, config = {}, logger) {
|
|
45
|
+
const searchIndexFile = path.join(outputDir, config.indexFile || 'search-index.json')
|
|
46
|
+
const maxContentLength = config.maxContentLength || 1000
|
|
47
|
+
const minContentLength = config.minContentLength || 50
|
|
48
|
+
|
|
49
|
+
// Find all HTML files
|
|
50
|
+
const htmlFiles = walkDirectory(outputDir).filter(file => file.endsWith('.html'))
|
|
51
|
+
const documents = []
|
|
52
|
+
let id = 0
|
|
53
|
+
|
|
54
|
+
htmlFiles.forEach(filePath => {
|
|
55
|
+
try {
|
|
56
|
+
const html = fs.readFileSync(filePath, 'utf8')
|
|
57
|
+
const title = getTitle(html)
|
|
58
|
+
const content = extractTextContent(html)
|
|
59
|
+
|
|
60
|
+
// Skip empty content
|
|
61
|
+
if (!content || content.length < minContentLength) {
|
|
62
|
+
return
|
|
63
|
+
}
|
|
64
|
+
|
|
65
|
+
// Generate relative URL
|
|
66
|
+
const relativePath = path.relative(outputDir, filePath)
|
|
67
|
+
const url = '/' + relativePath.replace(/\\/g, '/').replace(/\/index\.html$/, '/')
|
|
68
|
+
|
|
69
|
+
documents.push({
|
|
70
|
+
id: ++id,
|
|
71
|
+
title: title,
|
|
72
|
+
content: content.substring(0, maxContentLength),
|
|
73
|
+
url: url
|
|
74
|
+
})
|
|
75
|
+
|
|
76
|
+
if (config.debug) {
|
|
77
|
+
logger.debug(`Indexed: ${title} (${url})`)
|
|
78
|
+
}
|
|
79
|
+
} catch (error) {
|
|
80
|
+
logger.warn(`Failed to process ${filePath}:`, error.message)
|
|
81
|
+
}
|
|
82
|
+
})
|
|
83
|
+
|
|
84
|
+
const searchIndex = {
|
|
85
|
+
documents: documents
|
|
86
|
+
}
|
|
87
|
+
|
|
88
|
+
// Write search index
|
|
89
|
+
fs.writeFileSync(searchIndexFile, JSON.stringify(searchIndex, null, 2))
|
|
90
|
+
logger.info(`Search index generated: ${documents.length} documents → ${searchIndexFile}`)
|
|
91
|
+
}
|
|
92
|
+
|
|
93
|
+
function extractTextContent(html) {
|
|
94
|
+
try {
|
|
95
|
+
const JSDOM = loadJSDOM()
|
|
96
|
+
const dom = new JSDOM(html)
|
|
97
|
+
const document = dom.window.document
|
|
98
|
+
|
|
99
|
+
// Remove script and style elements
|
|
100
|
+
const scripts = document.querySelectorAll('script, style, nav, .navbar, .footer')
|
|
101
|
+
scripts.forEach(el => el.remove())
|
|
102
|
+
|
|
103
|
+
// Get main content area
|
|
104
|
+
const main = document.querySelector('main, .main, .content, article')
|
|
105
|
+
const content = main || document.body
|
|
106
|
+
|
|
107
|
+
return content.textContent.trim().replace(/\s+/g, ' ')
|
|
108
|
+
} catch (error) {
|
|
109
|
+
return ''
|
|
110
|
+
}
|
|
111
|
+
}
|
|
112
|
+
|
|
113
|
+
function getTitle(html) {
|
|
114
|
+
try {
|
|
115
|
+
const JSDOM = loadJSDOM()
|
|
116
|
+
const dom = new JSDOM(html)
|
|
117
|
+
const titleEl = dom.window.document.querySelector('title, h1, .page-title')
|
|
118
|
+
return titleEl ? titleEl.textContent.trim() : 'Untitled'
|
|
119
|
+
} catch (error) {
|
|
120
|
+
return 'Untitled'
|
|
121
|
+
}
|
|
122
|
+
}
|
|
123
|
+
|
|
124
|
+
function walkDirectory(dir, fileList = []) {
|
|
125
|
+
if (!fs.existsSync(dir)) {
|
|
126
|
+
return fileList
|
|
127
|
+
}
|
|
128
|
+
|
|
129
|
+
const files = fs.readdirSync(dir)
|
|
130
|
+
|
|
131
|
+
files.forEach(file => {
|
|
132
|
+
const filePath = path.join(dir, file)
|
|
133
|
+
const stat = fs.statSync(filePath)
|
|
134
|
+
|
|
135
|
+
if (stat.isDirectory()) {
|
|
136
|
+
walkDirectory(filePath, fileList)
|
|
137
|
+
} else {
|
|
138
|
+
fileList.push(filePath)
|
|
139
|
+
}
|
|
140
|
+
})
|
|
141
|
+
|
|
142
|
+
return fileList
|
|
143
|
+
}
|
|
144
|
+
|
|
145
|
+
module.exports = { register }
|
|
@@ -0,0 +1,28 @@
|
|
|
1
|
+
= Generated notebook sample
|
|
2
|
+
:page-jupyter:
|
|
3
|
+
:page-course-language: en
|
|
4
|
+
:page-course-concepts: DP-FND-01, DP-REP-01
|
|
5
|
+
:page-course-priority: P0
|
|
6
|
+
:page-course-difficulty: D1
|
|
7
|
+
:page-course-duration-minutes: 20
|
|
8
|
+
:page-course-outcomes: LO1, LO9
|
|
9
|
+
:page-course-execution-profile: fast
|
|
10
|
+
:jupyter-language-name: python
|
|
11
|
+
:jupyter-language-version: 3.12
|
|
12
|
+
:jupyter-kernel-name: python3
|
|
13
|
+
:jupyter-kernel-language: python
|
|
14
|
+
|
|
15
|
+
The page is the canonical explanation. Its generated notebook preserves this prose and the executable evidence below.
|
|
16
|
+
|
|
17
|
+
[stem]
|
|
18
|
+
++++
|
|
19
|
+
\widehat\mu = \frac{1}{n}\sum_{i=1}^{n}X_i.
|
|
20
|
+
++++
|
|
21
|
+
|
|
22
|
+
[source,python]
|
|
23
|
+
----
|
|
24
|
+
value = 6 * 7
|
|
25
|
+
print(value)
|
|
26
|
+
----
|
|
27
|
+
|
|
28
|
+
xref:attachment$foundations/sample.ipynb[Download the generated notebook].
|
|
@@ -0,0 +1,16 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schema_version": 1,
|
|
3
|
+
"component": "executable-contract",
|
|
4
|
+
"version": "",
|
|
5
|
+
"module": "ROOT",
|
|
6
|
+
"source_relative_path": "foundations/sample.adoc",
|
|
7
|
+
"attachment_relative_path": "foundations/sample.ipynb",
|
|
8
|
+
"page_url": "/executable-contract/foundations/sample.html",
|
|
9
|
+
"attachment_url": "/executable-contract/_attachments/foundations/sample.ipynb",
|
|
10
|
+
"plain_page_generates_notebook": false,
|
|
11
|
+
"ui_contract": {
|
|
12
|
+
"requested_attribute": "jupyter",
|
|
13
|
+
"resolved_url_attribute": "jupyter-url",
|
|
14
|
+
"status_attribute": "jupyter-status"
|
|
15
|
+
}
|
|
16
|
+
}
|
|
@@ -0,0 +1,30 @@
|
|
|
1
|
+
{
|
|
2
|
+
"schema_version": 1,
|
|
3
|
+
"cases": [
|
|
4
|
+
{
|
|
5
|
+
"id": "duplicate-attachment",
|
|
6
|
+
"importance": "P0",
|
|
7
|
+
"expected": "reject two generated resources with the same Antora attachment ID"
|
|
8
|
+
},
|
|
9
|
+
{
|
|
10
|
+
"id": "converter-warning",
|
|
11
|
+
"importance": "P0",
|
|
12
|
+
"expected": "reject a required page when the converter reports an unsupported node"
|
|
13
|
+
},
|
|
14
|
+
{
|
|
15
|
+
"id": "missing-converter",
|
|
16
|
+
"importance": "P0",
|
|
17
|
+
"expected": "fail with the package name and requested page resource ID"
|
|
18
|
+
},
|
|
19
|
+
{
|
|
20
|
+
"id": "unresolved-url",
|
|
21
|
+
"importance": "P0",
|
|
22
|
+
"expected": "do not expose a notebook URL or generated status"
|
|
23
|
+
},
|
|
24
|
+
{
|
|
25
|
+
"id": "nondeterministic-manifest",
|
|
26
|
+
"importance": "P1",
|
|
27
|
+
"expected": "reject timestamps, unstable ordering, or missing source/output digests"
|
|
28
|
+
}
|
|
29
|
+
]
|
|
30
|
+
}
|
|
@@ -0,0 +1,152 @@
|
|
|
1
|
+
const fs = require('node:fs')
|
|
2
|
+
const path = require('node:path')
|
|
3
|
+
const chai = require('chai')
|
|
4
|
+
const expect = chai.expect
|
|
5
|
+
const jupyter = require('../src/jupyter.js')
|
|
6
|
+
|
|
7
|
+
const fixtureDir = path.join(__dirname, 'fixtures', 'jupyter-contract')
|
|
8
|
+
|
|
9
|
+
describe('Jupyter generation contract fixture', () => {
|
|
10
|
+
it('defines a page-relative attachment and confirmed UI URL contract', () => {
|
|
11
|
+
const expected = JSON.parse(fs.readFileSync(path.join(fixtureDir, 'expected.json'), 'utf8'))
|
|
12
|
+
|
|
13
|
+
expect(expected.source_relative_path).to.equal('foundations/sample.adoc')
|
|
14
|
+
expect(expected.attachment_relative_path).to.equal('foundations/sample.ipynb')
|
|
15
|
+
expect(expected.attachment_url).to.equal(
|
|
16
|
+
'/executable-contract/_attachments/foundations/sample.ipynb'
|
|
17
|
+
)
|
|
18
|
+
expect(expected.plain_page_generates_notebook).to.equal(false)
|
|
19
|
+
expect(expected.ui_contract.resolved_url_attribute).to.equal('jupyter-url')
|
|
20
|
+
expect(expected.ui_contract.status_attribute).to.equal('jupyter-status')
|
|
21
|
+
})
|
|
22
|
+
|
|
23
|
+
it('contains marked and unmarked representative Antora pages', () => {
|
|
24
|
+
const pages = path.join(fixtureDir, 'content', 'modules', 'ROOT', 'pages')
|
|
25
|
+
const marked = fs.readFileSync(path.join(pages, 'foundations', 'sample.adoc'), 'utf8')
|
|
26
|
+
const plain = fs.readFileSync(path.join(pages, 'plain.adoc'), 'utf8')
|
|
27
|
+
|
|
28
|
+
expect(marked).to.contain(':page-jupyter:')
|
|
29
|
+
expect(marked).to.contain('[source,python]')
|
|
30
|
+
expect(plain).not.to.contain(':page-jupyter:')
|
|
31
|
+
})
|
|
32
|
+
|
|
33
|
+
it('inventories required generation failures before implementation', () => {
|
|
34
|
+
const failures = JSON.parse(fs.readFileSync(path.join(fixtureDir, 'failures.json'), 'utf8'))
|
|
35
|
+
const ids = failures.cases.map((item) => item.id)
|
|
36
|
+
|
|
37
|
+
expect(ids).to.deep.equal([
|
|
38
|
+
'duplicate-attachment',
|
|
39
|
+
'converter-warning',
|
|
40
|
+
'missing-converter',
|
|
41
|
+
'unresolved-url',
|
|
42
|
+
'nondeterministic-manifest'
|
|
43
|
+
])
|
|
44
|
+
expect(failures.cases.filter((item) => item.importance === 'P0')).to.have.length(4)
|
|
45
|
+
})
|
|
46
|
+
|
|
47
|
+
it('converts only marked pages and normalises notebook metadata deterministically', () => {
|
|
48
|
+
const pagesRoot = path.join(fixtureDir, 'content', 'modules', 'ROOT', 'pages')
|
|
49
|
+
const page = fixturePage('foundations/sample.adoc', pagesRoot)
|
|
50
|
+
const plain = fixturePage('plain.adoc', pagesRoot)
|
|
51
|
+
|
|
52
|
+
expect(jupyter.isJupyterPage(page)).to.equal(true)
|
|
53
|
+
expect(jupyter.isJupyterPage(plain)).to.equal(false)
|
|
54
|
+
const first = jupyter.generateNotebook(page)
|
|
55
|
+
const second = jupyter.generateNotebook(page)
|
|
56
|
+
expect(first.equals(second)).to.equal(true)
|
|
57
|
+
|
|
58
|
+
const notebook = JSON.parse(first.toString('utf8'))
|
|
59
|
+
expect(notebook.nbformat).to.equal(4)
|
|
60
|
+
expect(notebook.nbformat_minor).to.equal(5)
|
|
61
|
+
expect(notebook.metadata.kernelspec).to.include({
|
|
62
|
+
name: 'python3',
|
|
63
|
+
language: 'python',
|
|
64
|
+
display_name: 'Python 3'
|
|
65
|
+
})
|
|
66
|
+
expect(notebook.metadata.course).to.include({
|
|
67
|
+
source_kind: 'asciidoc',
|
|
68
|
+
source_path: 'foundations/sample.adoc',
|
|
69
|
+
priority: 'P0',
|
|
70
|
+
difficulty: 'D1',
|
|
71
|
+
language: 'en',
|
|
72
|
+
execution_profile: 'fast'
|
|
73
|
+
})
|
|
74
|
+
expect(notebook.metadata.course.concept_ids).to.deep.equal(['DP-FND-01', 'DP-REP-01'])
|
|
75
|
+
expect(notebook.cells.every((cell) => /^[0-9a-f]{16}$/.test(cell.id))).to.equal(true)
|
|
76
|
+
const markdown = notebook.cells
|
|
77
|
+
.filter((cell) => cell.cell_type === 'markdown')
|
|
78
|
+
.map((cell) => Array.isArray(cell.source) ? cell.source.join('') : cell.source)
|
|
79
|
+
.join('\n')
|
|
80
|
+
expect(markdown).to.contain('$$\n\\widehat\\mu')
|
|
81
|
+
expect(markdown).to.contain('](sample.ipynb)')
|
|
82
|
+
expect(markdown).not.to.match(/\n\+\n\\widehat/)
|
|
83
|
+
expect(markdown).not.to.contain('attachment$')
|
|
84
|
+
expect(notebook.cells.filter((cell) => cell.cell_type === 'code'))
|
|
85
|
+
.to.satisfy((cells) => cells.length === 1 && cells.every((cell) => cell.outputs.length === 0))
|
|
86
|
+
})
|
|
87
|
+
|
|
88
|
+
it('registers page-relative attachments, a manifest, and confirmed UI attributes', () => {
|
|
89
|
+
const pagesRoot = path.join(fixtureDir, 'content', 'modules', 'ROOT', 'pages')
|
|
90
|
+
const page = fixturePage('foundations/sample.adoc', pagesRoot)
|
|
91
|
+
const plain = fixturePage('plain.adoc', pagesRoot)
|
|
92
|
+
const files = []
|
|
93
|
+
const contentCatalog = {
|
|
94
|
+
getPages: () => [plain, page],
|
|
95
|
+
addFile: (file) => {
|
|
96
|
+
if (files.some((candidate) => JSON.stringify(candidate.src) === JSON.stringify(file.src))) {
|
|
97
|
+
throw new Error(`duplicate attachment: ${file.src.relative}`)
|
|
98
|
+
}
|
|
99
|
+
file.pub = {
|
|
100
|
+
url: `/executable-contract/_attachments/${file.src.relative}`
|
|
101
|
+
}
|
|
102
|
+
files.push(file)
|
|
103
|
+
return file
|
|
104
|
+
}
|
|
105
|
+
}
|
|
106
|
+
const handlers = {}
|
|
107
|
+
const context = {
|
|
108
|
+
getLogger: () => ({ info: () => {} }),
|
|
109
|
+
on: (event, handler) => { handlers[event] = handler }
|
|
110
|
+
}
|
|
111
|
+
|
|
112
|
+
jupyter.register.call(context, { config: {} })
|
|
113
|
+
handlers.contentClassified({ contentCatalog })
|
|
114
|
+
page.asciidoc = { attributes: {} }
|
|
115
|
+
plain.asciidoc = { attributes: {} }
|
|
116
|
+
handlers.documentsConverted()
|
|
117
|
+
|
|
118
|
+
expect(files.map((file) => file.src.relative)).to.deep.equal([
|
|
119
|
+
'foundations/sample.ipynb',
|
|
120
|
+
'generated/asciidoc-notebook-manifest.json'
|
|
121
|
+
])
|
|
122
|
+
expect(page.asciidoc.attributes).to.include({
|
|
123
|
+
'page-jupyter-status': 'generated',
|
|
124
|
+
'page-jupyter-url': '/executable-contract/_attachments/foundations/sample.ipynb'
|
|
125
|
+
})
|
|
126
|
+
expect(plain.asciidoc.attributes).to.deep.equal({})
|
|
127
|
+
|
|
128
|
+
const manifest = JSON.parse(files[1].contents.toString('utf8'))
|
|
129
|
+
expect(manifest.entries).to.have.length(1)
|
|
130
|
+
expect(manifest.entries[0].notebook_path).to.equal('foundations/sample.ipynb')
|
|
131
|
+
expect(manifest.entries[0].notebook_url).to.equal(
|
|
132
|
+
'/executable-contract/_attachments/foundations/sample.ipynb'
|
|
133
|
+
)
|
|
134
|
+
expect(manifest).not.to.have.property('generated_at')
|
|
135
|
+
})
|
|
136
|
+
})
|
|
137
|
+
|
|
138
|
+
function fixturePage (relative, root) {
|
|
139
|
+
return {
|
|
140
|
+
contents: fs.readFileSync(path.join(root, relative)),
|
|
141
|
+
src: {
|
|
142
|
+
component: 'executable-contract',
|
|
143
|
+
version: '',
|
|
144
|
+
module: 'ROOT',
|
|
145
|
+
family: 'page',
|
|
146
|
+
relative
|
|
147
|
+
},
|
|
148
|
+
pub: {
|
|
149
|
+
url: `/executable-contract/${relative.replace(/\.adoc$/, '.html')}`
|
|
150
|
+
}
|
|
151
|
+
}
|
|
152
|
+
}
|
|
@@ -0,0 +1,36 @@
|
|
|
1
|
+
const { describe, it } = require('mocha')
|
|
2
|
+
const { expect } = require('chai')
|
|
3
|
+
const fs = require('fs')
|
|
4
|
+
const path = require('path')
|
|
5
|
+
|
|
6
|
+
describe('Lunr extension', () => {
|
|
7
|
+
const { register } = require('../src/lunr.js')
|
|
8
|
+
|
|
9
|
+
it('should export a register function', () => {
|
|
10
|
+
expect(register).to.be.a('function')
|
|
11
|
+
})
|
|
12
|
+
|
|
13
|
+
it('should register with Antora context', () => {
|
|
14
|
+
let eventRegistered = false
|
|
15
|
+
const mockContext = {
|
|
16
|
+
getLogger: () => ({
|
|
17
|
+
info: () => {},
|
|
18
|
+
error: () => {},
|
|
19
|
+
debug: () => {},
|
|
20
|
+
warn: () => {}
|
|
21
|
+
}),
|
|
22
|
+
once: (event, handler) => {
|
|
23
|
+
if (event === 'sitePublished') {
|
|
24
|
+
eventRegistered = true
|
|
25
|
+
}
|
|
26
|
+
}
|
|
27
|
+
}
|
|
28
|
+
|
|
29
|
+
const config = {}
|
|
30
|
+
const playbook = { output: { dir: 'build/site' } }
|
|
31
|
+
|
|
32
|
+
register.call(mockContext, { config, playbook })
|
|
33
|
+
|
|
34
|
+
expect(eventRegistered).to.be.true
|
|
35
|
+
})
|
|
36
|
+
})
|