WIP - add extractor, generate snippet_data

node_modules/remark-parse/lib/tokenize/link.js

@@ -0,0 +1,381 @@
+'use strict'
+
+var whitespace = require('is-whitespace-character')
+var locate = require('../locate/link')
+
+module.exports = link
+link.locator = locate
+
+var lineFeed = '\n'
+var exclamationMark = '!'
+var quotationMark = '"'
+var apostrophe = "'"
+var leftParenthesis = '('
+var rightParenthesis = ')'
+var lessThan = '<'
+var greaterThan = '>'
+var leftSquareBracket = '['
+var backslash = '\\'
+var rightSquareBracket = ']'
+var graveAccent = '`'
+
+function link(eat, value, silent) {
+  var self = this
+  var subvalue = ''
+  var index = 0
+  var character = value.charAt(0)
+  var pedantic = self.options.pedantic
+  var commonmark = self.options.commonmark
+  var gfm = self.options.gfm
+  var closed
+  var count
+  var opening
+  var beforeURL
+  var beforeTitle
+  var subqueue
+  var hasMarker
+  var isImage
+  var content
+  var marker
+  var length
+  var title
+  var depth
+  var queue
+  var url
+  var now
+  var exit
+  var node
+
+  // Detect whether this is an image.
+  if (character === exclamationMark) {
+    isImage = true
+    subvalue = character
+    character = value.charAt(++index)
+  }
+
+  // Eat the opening.
+  if (character !== leftSquareBracket) {
+    return
+  }
+
+  // Exit when this is a link and we’re already inside a link.
+  if (!isImage && self.inLink) {
+    return
+  }
+
+  subvalue += character
+  queue = ''
+  index++
+
+  // Eat the content.
+  length = value.length
+  now = eat.now()
+  depth = 0
+
+  now.column += index
+  now.offset += index
+
+  while (index < length) {
+    character = value.charAt(index)
+    subqueue = character
+
+    if (character === graveAccent) {
+      // Inline-code in link content.
+      count = 1
+
+      while (value.charAt(index + 1) === graveAccent) {
+        subqueue += character
+        index++
+        count++
+      }
+
+      if (!opening) {
+        opening = count
+      } else if (count >= opening) {
+        opening = 0
+      }
+    } else if (character === backslash) {
+      // Allow brackets to be escaped.
+      index++
+      subqueue += value.charAt(index)
+    } else if ((!opening || gfm) && character === leftSquareBracket) {
+      // In GFM mode, brackets in code still count.  In all other modes,
+      // they don’t.
+      depth++
+    } else if ((!opening || gfm) && character === rightSquareBracket) {
+      if (depth) {
+        depth--
+      } else {
+        // Allow white-space between content and url in GFM mode.
+        if (!pedantic) {
+          while (index < length) {
+            character = value.charAt(index + 1)
+
+            if (!whitespace(character)) {
+              break
+            }
+
+            subqueue += character
+            index++
+          }
+        }
+
+        if (value.charAt(index + 1) !== leftParenthesis) {
+          return
+        }
+
+        subqueue += leftParenthesis
+        closed = true
+        index++
+
+        break
+      }
+    }
+
+    queue += subqueue
+    subqueue = ''
+    index++
+  }
+
+  // Eat the content closing.
+  if (!closed) {
+    return
+  }
+
+  content = queue
+  subvalue += queue + subqueue
+  index++
+
+  // Eat white-space.
+  while (index < length) {
+    character = value.charAt(index)
+
+    if (!whitespace(character)) {
+      break
+    }
+
+    subvalue += character
+    index++
+  }
+
+  // Eat the URL.
+  character = value.charAt(index)
+  queue = ''
+  beforeURL = subvalue
+
+  if (character === lessThan) {
+    index++
+    beforeURL += lessThan
+
+    while (index < length) {
+      character = value.charAt(index)
+
+      if (character === greaterThan) {
+        break
+      }
+
+      if (commonmark && character === lineFeed) {
+        return
+      }
+
+      queue += character
+      index++
+    }
+
+    if (value.charAt(index) !== greaterThan) {
+      return
+    }
+
+    subvalue += lessThan + queue + greaterThan
+    url = queue
+    index++
+  } else {
+    character = null
+    subqueue = ''
+
+    while (index < length) {
+      character = value.charAt(index)
+
+      if (
+        subqueue &&
+        (character === quotationMark ||
+          character === apostrophe ||
+          (commonmark && character === leftParenthesis))
+      ) {
+        break
+      }
+
+      if (whitespace(character)) {
+        if (!pedantic) {
+          break
+        }
+
+        subqueue += character
+      } else {
+        if (character === leftParenthesis) {
+          depth++
+        } else if (character === rightParenthesis) {
+          if (depth === 0) {
+            break
+          }
+
+          depth--
+        }
+
+        queue += subqueue
+        subqueue = ''
+
+        if (character === backslash) {
+          queue += backslash
+          character = value.charAt(++index)
+        }
+
+        queue += character
+      }
+
+      index++
+    }
+
+    subvalue += queue
+    url = queue
+    index = subvalue.length
+  }
+
+  // Eat white-space.
+  queue = ''
+
+  while (index < length) {
+    character = value.charAt(index)
+
+    if (!whitespace(character)) {
+      break
+    }
+
+    queue += character
+    index++
+  }
+
+  character = value.charAt(index)
+  subvalue += queue
+
+  // Eat the title.
+  if (
+    queue &&
+    (character === quotationMark ||
+      character === apostrophe ||
+      (commonmark && character === leftParenthesis))
+  ) {
+    index++
+    subvalue += character
+    queue = ''
+    marker = character === leftParenthesis ? rightParenthesis : character
+    beforeTitle = subvalue
+
+    // In commonmark-mode, things are pretty easy: the marker cannot occur
+    // inside the title.  Non-commonmark does, however, support nested
+    // delimiters.
+    if (commonmark) {
+      while (index < length) {
+        character = value.charAt(index)
+
+        if (character === marker) {
+          break
+        }
+
+        if (character === backslash) {
+          queue += backslash
+          character = value.charAt(++index)
+        }
+
+        index++
+        queue += character
+      }
+
+      character = value.charAt(index)
+
+      if (character !== marker) {
+        return
+      }
+
+      title = queue
+      subvalue += queue + character
+      index++
+
+      while (index < length) {
+        character = value.charAt(index)
+
+        if (!whitespace(character)) {
+          break
+        }
+
+        subvalue += character
+        index++
+      }
+    } else {
+      subqueue = ''
+
+      while (index < length) {
+        character = value.charAt(index)
+
+        if (character === marker) {
+          if (hasMarker) {
+            queue += marker + subqueue
+            subqueue = ''
+          }
+
+          hasMarker = true
+        } else if (!hasMarker) {
+          queue += character
+        } else if (character === rightParenthesis) {
+          subvalue += queue + marker + subqueue
+          title = queue
+          break
+        } else if (whitespace(character)) {
+          subqueue += character
+        } else {
+          queue += marker + subqueue + character
+          subqueue = ''
+          hasMarker = false
+        }
+
+        index++
+      }
+    }
+  }
+
+  if (value.charAt(index) !== rightParenthesis) {
+    return
+  }
+
+  /* istanbul ignore if - never used (yet) */
+  if (silent) {
+    return true
+  }
+
+  subvalue += rightParenthesis
+
+  url = self.decode.raw(self.unescape(url), eat(beforeURL).test().end, {
+    nonTerminated: false
+  })
+
+  if (title) {
+    beforeTitle = eat(beforeTitle).test().end
+    title = self.decode.raw(self.unescape(title), beforeTitle)
+  }
+
+  node = {
+    type: isImage ? 'image' : 'link',
+    title: title || null,
+    url: url
+  }
+
+  if (isImage) {
+    node.alt = self.decode.raw(self.unescape(content), now) || null
+  } else {
+    exit = self.enterLink()
+    node.children = self.tokenizeInline(content, now)
+    exit()
+  }
+
+  return eat(subvalue)(node)
+}