1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
|
package org.jetbrains.dokka.parsers
import com.intellij.psi.PsiElement
import org.jetbrains.dokka.model.doc.*
import org.intellij.markdown.MarkdownElementTypes
import org.intellij.markdown.MarkdownTokenTypes
import org.intellij.markdown.ast.ASTNode
import org.intellij.markdown.ast.CompositeASTNode
import org.intellij.markdown.ast.LeafASTNode
import org.intellij.markdown.ast.impl.ListItemCompositeNode
import org.intellij.markdown.flavours.gfm.GFMElementTypes
import org.intellij.markdown.flavours.gfm.GFMFlavourDescriptor
import org.intellij.markdown.flavours.gfm.GFMTokenTypes
import org.jetbrains.dokka.analysis.DokkaResolutionFacade
import org.jetbrains.dokka.links.DRI
import org.jetbrains.dokka.parsers.factories.DocTagsFromIElementFactory
import org.jetbrains.dokka.utilities.DokkaLogger
import org.jetbrains.kotlin.descriptors.ClassDescriptor
import org.jetbrains.kotlin.descriptors.DeclarationDescriptor
import org.jetbrains.kotlin.idea.kdoc.resolveKDocLink
import org.jetbrains.kotlin.kdoc.parser.KDocKnownTag
import org.jetbrains.kotlin.kdoc.psi.impl.KDocSection
import org.jetbrains.kotlin.kdoc.psi.impl.KDocTag
import java.net.MalformedURLException
import org.intellij.markdown.parser.MarkdownParser as IntellijMarkdownParser
class MarkdownParser(
private val resolutionFacade: DokkaResolutionFacade? = null,
private val declarationDescriptor: DeclarationDescriptor? = null,
private val logger: DokkaLogger
) : Parser() {
inner class MarkdownVisitor(val text: String, val destinationLinksMap: Map<String, String>) {
private fun headersHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
node.type,
visitNode(node.children.find { it.type == MarkdownTokenTypes.ATX_CONTENT }
?: throw IllegalStateException("Wrong AST Tree. ATX Header does not contain expected content")).children
)
private fun horizontalRulesHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(MarkdownTokenTypes.HORIZONTAL_RULE)
private fun emphasisHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
node.type,
children = listOf(visitNode(node.children[node.children.size / 2]))
)
private fun blockquotesHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
node.type, children = node.children
.filterIsInstance<CompositeASTNode>()
.evaluateChildren()
)
private fun listsHandler(node: ASTNode): DocTag {
val children = node.children.filterIsInstance<ListItemCompositeNode>().flatMap {
if (it.children.last().type in listOf(
MarkdownElementTypes.ORDERED_LIST,
MarkdownElementTypes.UNORDERED_LIST
)
) {
val nestedList = it.children.last()
(it.children as MutableList).removeAt(it.children.lastIndex)
listOf(it, nestedList)
} else
listOf(it)
}
return DocTagsFromIElementFactory.getInstance(
node.type,
children =
children
.map {
if (it.type == MarkdownElementTypes.LIST_ITEM)
DocTagsFromIElementFactory.getInstance(
it.type,
children = it
.children
.filterIsInstance<CompositeASTNode>()
.evaluateChildren()
)
else
visitNode(it)
},
params =
if (node.type == MarkdownElementTypes.ORDERED_LIST) {
val listNumberNode = node.children.first().children.first()
mapOf(
"start" to text.substring(
listNumberNode.startOffset,
listNumberNode.endOffset
).trim().dropLast(1)
)
} else
emptyMap()
)
}
private fun resolveDRI(mdLink: String): DRI? =
mdLink
.removePrefix("[")
.removeSuffix("]")
.let { link ->
try {
java.net.URL(link)
null
} catch (e: MalformedURLException) {
try {
if (resolutionFacade != null && declarationDescriptor != null) {
resolveKDocLink(
resolutionFacade.resolveSession.bindingContext,
resolutionFacade,
declarationDescriptor,
null,
link.split('.')
).minBy { it is ClassDescriptor }?.let { DRI.from(it) }
} else null
} catch (e1: IllegalArgumentException) {
null
}
}
}
private fun referenceLinksHandler(node: ASTNode): DocTag {
val linkLabel = node.children.find { it.type == MarkdownElementTypes.LINK_LABEL }
?: throw IllegalStateException("Wrong AST Tree. Reference link does not contain expected content")
val linkText = node.children.findLast { it.type == MarkdownElementTypes.LINK_TEXT } ?: linkLabel
val linkKey = text.substring(linkLabel.startOffset, linkLabel.endOffset)
val link = destinationLinksMap[linkKey.toLowerCase()] ?: linkKey
return linksHandler(linkText, link)
}
private fun inlineLinksHandler(node: ASTNode): DocTag {
val linkText = node.children.find { it.type == MarkdownElementTypes.LINK_TEXT }
?: throw IllegalStateException("Wrong AST Tree. Inline link does not contain expected content")
val linkDestination = node.children.find { it.type == MarkdownElementTypes.LINK_DESTINATION }
?: throw IllegalStateException("Wrong AST Tree. Inline link does not contain expected content")
val linkTitle = node.children.find { it.type == MarkdownElementTypes.LINK_TITLE }
val link = text.substring(linkDestination.startOffset, linkDestination.endOffset)
return linksHandler(linkText, link, linkTitle)
}
private fun autoLinksHandler(node: ASTNode): DocTag {
val link = text.substring(node.startOffset + 1, node.endOffset - 1)
return linksHandler(node, link)
}
private fun linksHandler(linkText: ASTNode, link: String, linkTitle: ASTNode? = null): DocTag {
val dri: DRI? = resolveDRI(link)
val params = if (linkTitle == null)
mapOf("href" to link)
else
mapOf("href" to link, "title" to text.substring(linkTitle.startOffset + 1, linkTitle.endOffset - 1))
return DocTagsFromIElementFactory.getInstance(
MarkdownElementTypes.INLINE_LINK,
params = params,
children = linkText.children.drop(1).dropLast(1).evaluateChildren(),
dri = dri
)
}
private fun imagesHandler(node: ASTNode): DocTag {
val linkNode =
node.children.last().children.find { it.type == MarkdownElementTypes.LINK_LABEL }!!.children[1]
val link = text.substring(linkNode.startOffset, linkNode.endOffset)
val src = mapOf("src" to link)
return DocTagsFromIElementFactory.getInstance(
node.type,
params = src,
children = listOf(visitNode(node.children.last().children.find { it.type == MarkdownElementTypes.LINK_TEXT }!!))
)
}
private fun codeSpansHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
node.type,
children = listOf(
DocTagsFromIElementFactory.getInstance(
MarkdownTokenTypes.TEXT,
body = text.substring(node.startOffset + 1, node.endOffset - 1).replace('\n', ' ').trimIndent()
)
)
)
private fun codeFencesHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
node.type,
children = node
.children
.dropWhile { it.type != MarkdownTokenTypes.CODE_FENCE_CONTENT }
.dropLastWhile { it.type != MarkdownTokenTypes.CODE_FENCE_CONTENT }
.map {
if (it.type == MarkdownTokenTypes.EOL)
LeafASTNode(MarkdownTokenTypes.HARD_LINE_BREAK, 0, 0)
else
it
}.evaluateChildren(),
params = node
.children
.find { it.type == MarkdownTokenTypes.FENCE_LANG }
?.let { mapOf("lang" to text.substring(it.startOffset, it.endOffset)) }
?: emptyMap()
)
private fun codeBlocksHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(node.type, children = node.children.evaluateChildren())
private fun defaultHandler(node: ASTNode): DocTag =
DocTagsFromIElementFactory.getInstance(
MarkdownElementTypes.PARAGRAPH,
children = node.children.evaluateChildren()
)
fun visitNode(node: ASTNode): DocTag =
when (node.type) {
MarkdownElementTypes.ATX_1,
MarkdownElementTypes.ATX_2,
MarkdownElementTypes.ATX_3,
MarkdownElementTypes.ATX_4,
MarkdownElementTypes.ATX_5,
MarkdownElementTypes.ATX_6 -> headersHandler(node)
MarkdownTokenTypes.HORIZONTAL_RULE -> horizontalRulesHandler(node)
MarkdownElementTypes.STRONG,
MarkdownElementTypes.EMPH -> emphasisHandler(node)
MarkdownElementTypes.FULL_REFERENCE_LINK,
MarkdownElementTypes.SHORT_REFERENCE_LINK -> referenceLinksHandler(node)
MarkdownElementTypes.INLINE_LINK -> inlineLinksHandler(node)
MarkdownElementTypes.AUTOLINK -> autoLinksHandler(node)
MarkdownElementTypes.BLOCK_QUOTE -> blockquotesHandler(node)
MarkdownElementTypes.UNORDERED_LIST,
MarkdownElementTypes.ORDERED_LIST -> listsHandler(node)
MarkdownElementTypes.CODE_BLOCK -> codeBlocksHandler(node)
MarkdownElementTypes.CODE_FENCE -> codeFencesHandler(node)
MarkdownElementTypes.CODE_SPAN -> codeSpansHandler(node)
MarkdownElementTypes.IMAGE -> imagesHandler(node)
MarkdownTokenTypes.HARD_LINE_BREAK -> DocTagsFromIElementFactory.getInstance(node.type)
MarkdownTokenTypes.CODE_FENCE_CONTENT,
MarkdownTokenTypes.CODE_LINE,
MarkdownTokenTypes.TEXT -> DocTagsFromIElementFactory.getInstance(
MarkdownTokenTypes.TEXT,
body = text
.substring(node.startOffset, node.endOffset).transform()
)
MarkdownElementTypes.MARKDOWN_FILE -> if (node.children.size == 1) visitNode(node.children.first()) else defaultHandler(
node
)
GFMElementTypes.STRIKETHROUGH -> DocTagsFromIElementFactory.getInstance(
GFMElementTypes.STRIKETHROUGH,
body = text
.substring(node.startOffset, node.endOffset).transform()
)
GFMElementTypes.TABLE -> DocTagsFromIElementFactory.getInstance(
GFMElementTypes.TABLE,
children = node.children.filterTabSeparators().evaluateChildren()
)
GFMElementTypes.HEADER -> DocTagsFromIElementFactory.getInstance(
GFMElementTypes.HEADER,
children = node.children.filterTabSeparators().evaluateChildren()
)
GFMElementTypes.ROW -> DocTagsFromIElementFactory.getInstance(
GFMElementTypes.ROW,
children = node.children.filterTabSeparators().evaluateChildren()
)
else -> defaultHandler(node)
}
private fun List<ASTNode>.filterTabSeparators() =
this.filterNot { it.type == GFMTokenTypes.TABLE_SEPARATOR }
private fun List<ASTNode>.evaluateChildren(): List<DocTag> =
this.removeUselessTokens().mergeLeafASTNodes().map { visitNode(it) }
private fun List<ASTNode>.removeUselessTokens(): List<ASTNode> =
this.filterIndexed { index, node ->
!(node.type == MarkdownElementTypes.LINK_DEFINITION || (
node.type == MarkdownTokenTypes.EOL &&
this.getOrNull(index - 1)?.type == MarkdownTokenTypes.HARD_LINE_BREAK
))
}
private val notLeafNodes = listOf(MarkdownTokenTypes.HORIZONTAL_RULE, MarkdownTokenTypes.HARD_LINE_BREAK)
private fun List<ASTNode>.isNotLeaf(index: Int): Boolean =
if (index in 0..this.lastIndex)
(this[index] is CompositeASTNode) || this[index].type in notLeafNodes
else
false
private fun List<ASTNode>.mergeLeafASTNodes(): List<ASTNode> {
val children: MutableList<ASTNode> = mutableListOf()
var index = 0
while (index <= this.lastIndex) {
if (this.isNotLeaf(index)) {
children += this[index]
} else {
val startOffset = this[index].startOffset
while (index < this.lastIndex) {
if (this.isNotLeaf(index + 1) || this[index + 1].startOffset != this[index].endOffset) {
val endOffset = this[index].endOffset
if (text.substring(startOffset, endOffset).transform().trim().isNotEmpty())
children += LeafASTNode(MarkdownTokenTypes.TEXT, startOffset, endOffset)
break
}
index++
}
if (index == this.lastIndex) {
val endOffset = this[index].endOffset
if (text.substring(startOffset, endOffset).transform().trim().isNotEmpty())
children += LeafASTNode(MarkdownTokenTypes.TEXT, startOffset, endOffset)
}
}
index++
}
return children
}
private fun String.transform() = this
.replace(Regex("\n\n+"), "") // Squashing new lines between paragraphs
.replace(Regex("\n"), " ")
.replace(Regex(" >+ +"), " ") // Replacement used in blockquotes, get rid of garbage
}
private fun getAllDestinationLinks(text: String, node: ASTNode): List<Pair<String, String>> =
node.children
.filter { it.type == MarkdownElementTypes.LINK_DEFINITION }
.map {
text.substring(it.children[0].startOffset, it.children[0].endOffset).toLowerCase() to
text.substring(it.children[2].startOffset, it.children[2].endOffset)
} +
node.children.filterIsInstance<CompositeASTNode>().flatMap { getAllDestinationLinks(text, it) }
private fun markdownToDocNode(text: String): DocTag {
val flavourDescriptor = GFMFlavourDescriptor()
val markdownAstRoot: ASTNode = IntellijMarkdownParser(flavourDescriptor).buildMarkdownTreeFromString(text)
return MarkdownVisitor(text, getAllDestinationLinks(text, markdownAstRoot).toMap()).visitNode(markdownAstRoot)
}
override fun parseStringToDocNode(extractedString: String) = markdownToDocNode(extractedString)
override fun preparse(text: String) = text
private fun findParent(kDoc: PsiElement): PsiElement =
if (kDoc is KDocSection) findParent(kDoc.parent) else kDoc
private fun getAllKDocTags(kDocImpl: PsiElement): List<KDocTag> =
kDocImpl.children.filterIsInstance<KDocTag>().filterNot { it is KDocSection } + kDocImpl.children.flatMap {
getAllKDocTags(
it
)
}
fun parseFromKDocTag(kDocTag: KDocTag?): DocumentationNode {
return if (kDocTag == null)
DocumentationNode(emptyList())
else
DocumentationNode(
(listOf(kDocTag) + getAllKDocTags(findParent(kDocTag))).map {
when (it.knownTag) {
null -> if (it.name == null) Description(parseStringToDocNode(it.getContent())) else CustomWrapperTag(
parseStringToDocNode(it.getContent()),
it.name!!
)
KDocKnownTag.AUTHOR -> Author(parseStringToDocNode(it.getContent()))
KDocKnownTag.THROWS -> Throws(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty()
)
KDocKnownTag.EXCEPTION -> Throws(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty()
)
KDocKnownTag.PARAM -> Param(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty()
)
KDocKnownTag.RECEIVER -> Receiver(parseStringToDocNode(it.getContent()))
KDocKnownTag.RETURN -> Return(parseStringToDocNode(it.getContent()))
KDocKnownTag.SEE -> See(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty(),
parseStringToDocNode("[${it.getSubjectName()}]")
.let {
val link = it.children[0]
if (link is DocumentationLink) link.dri
else null
}
)
KDocKnownTag.SINCE -> Since(parseStringToDocNode(it.getContent()))
KDocKnownTag.CONSTRUCTOR -> Constructor(parseStringToDocNode(it.getContent()))
KDocKnownTag.PROPERTY -> Property(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty()
)
KDocKnownTag.SAMPLE -> Sample(
parseStringToDocNode(it.getContent()),
it.getSubjectName().orEmpty()
)
KDocKnownTag.SUPPRESS -> Suppress(parseStringToDocNode(it.getContent()))
}
}
)
}
}
|