👌 Throw an error if 3rd party plugin doesn't increment or counters (previously, markdown-it would likely go into infinite loop instead)

Author: hukkin

markdown_it/common/utils.py

@@ -91,25 +91,24 @@ def fromCodePoint(c: int) -> str:
     r'\\([!"#$%&\'()*+,\-.\/:;<=>?@[\\\]^_`{|}~])' + "|" + r"&([a-z#][a-z0-9]{1,31});",
     re.IGNORECASE,
 )
-DIGITAL_ENTITY_BASE10_RE = re.compile(r"#([0-9]{1,8})")
-DIGITAL_ENTITY_BASE16_RE = re.compile(r"#x([a-f0-9]{1,8})", re.IGNORECASE)
+DIGITAL_ENTITY_TEST_RE = re.compile(
+    r"^#((?:x[a-f0-9]{1,8}|[0-9]{1,8}))$", re.IGNORECASE
+)
 
 
 def replaceEntityPattern(match: str, name: str) -> str:
     """Convert HTML entity patterns,
     see https://spec.commonmark.org/0.30/#entity-references
     """
+    code = 0
+
     if name in entities:
         return entities[name]
 
-    code: None | int = None
-    if pat := DIGITAL_ENTITY_BASE10_RE.fullmatch(name):
-        code = int(pat.group(1), 10)
-    elif pat := DIGITAL_ENTITY_BASE16_RE.fullmatch(name):
-        code = int(pat.group(1), 16)
-
-    if code is not None and isValidEntityCode(code):
-        return fromCodePoint(code)
+    if name[0] == "#" and DIGITAL_ENTITY_TEST_RE.search(name):
+        code = int(name[2:], 16) if name[1].lower() == "x" else int(name[1:], 10)
+        if isValidEntityCode(code):
+            return fromCodePoint(code)
 
     return match
 

markdown_it/helpers/parse_link_destination.py

@@ -15,9 +15,8 @@ def __init__(self) -> None:
         self.str = ""
 
 
-def parseLinkDestination(string: str, pos: int, maximum: int) -> _Result:
-    lines = 0
-    start = pos
+def parseLinkDestination(string: str, start: int, maximum: int) -> _Result:
+    pos = start
     result = _Result()
 
     if charCodeAt(string, pos) == 0x3C:  # /* < */
@@ -80,7 +79,6 @@ def parseLinkDestination(string: str, pos: int, maximum: int) -> _Result:
         return result
 
     result.str = unescapeAll(string[start:pos])
-    result.lines = lines
     result.pos = pos
     result.ok = True
     return result

markdown_it/helpers/parse_link_title.py

@@ -16,9 +16,9 @@ def __str__(self) -> str:
         return self.str
 
 
-def parseLinkTitle(string: str, pos: int, maximum: int) -> _Result:
+def parseLinkTitle(string: str, start: int, maximum: int) -> _Result:
     lines = 0
-    start = pos
+    pos = start
     result = _Result()
 
     if pos >= maximum:

markdown_it/parser_block.py

@@ -58,6 +58,7 @@ def __init__(self) -> None:
 
     def tokenize(self, state: StateBlock, startLine: int, endLine: int) -> None:
         """Generate tokens for input range."""
+        ok = False
         rules = self.ruler.getRules("")
         line = startLine
         maxNesting = state.md.options.maxNesting
@@ -82,10 +83,19 @@ def tokenize(self, state: StateBlock, startLine: int, endLine: int) -> None:
             # - update `state.line`
             # - update `state.tokens`
             # - return True
+            prevLine = state.line
+
             for rule in rules:
-                if rule(state, line, endLine, False):
+                ok = rule(state, line, endLine, False)
+                if ok:
+                    if prevLine >= state.line:
+                        raise Exception("block rule didn't increment state.line")
                     break
 
+            # this can only happen if user disables paragraph rule
+            if not ok:
+                raise Exception("none of the block rules matched")
+
             # set state.tight if we had an empty line before current tag
             # i.e. latest empty line should not count
             state.tight = not hasEmptyLines

markdown_it/parser_inline.py

@@ -84,6 +84,8 @@ def skipToken(self, state: StateInline) -> None:
                 ok = rule(state, True)
                 state.level -= 1
                 if ok:
+                    if pos >= state.pos:
+                        raise Exception("inline rule didn't increment state.pos")
                     break
         else:
             # Too much nesting, just skip until the end of the paragraph.
@@ -117,11 +119,14 @@ def tokenize(self, state: StateInline) -> None:
             # - update `state.pos`
             # - update `state.tokens`
             # - return true
+            prevPos = state.pos
 
             if state.level < maxNesting:
                 for rule in rules:
                     ok = rule(state, False)
                     if ok:
+                        if prevPos >= state.pos:
+                            raise Exception("inline rule didn't increment state.pos")
                         break
 
             if ok:

markdown_it/port.yaml

@@ -1,7 +1,7 @@
 - package: markdown-it/markdown-it
   version: 13.0.1
-  commit: e843acc9edad115cbf8cf85e676443f01658be08
-  date: May 3, 2022
+  commit: 49ca65bbef067c7dba63468a48c4aee3048607dc
+  date: Sep 26, 2023
   notes:
     - Rename variables that use python built-in names, e.g.
       - `max` -> `maximum`

markdown_it/renderer.py

@@ -77,11 +77,12 @@ def render(
         result = ""
 
         for i, token in enumerate(tokens):
-            if token.type == "inline":
+            type = token.type
+            if type == "inline":
                 if token.children:
                     result += self.renderInline(token.children, options, env)
-            elif token.type in self.rules:
-                result += self.rules[token.type](tokens, i, options, env)
+            elif type in self.rules:
+                result += self.rules[type](tokens, i, options, env)
             else:
                 result += self.renderToken(tokens, i, options, env)
 
@@ -217,7 +218,7 @@ def code_inline(
             "<code"
             + self.renderAttrs(token)
             + ">"
-            + escapeHtml(tokens[idx].content)
+            + escapeHtml(token.content)
             + "</code>"
         )
 

markdown_it/rules_block/blockquote.py

@@ -27,81 +27,16 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool) ->
             return False
     except IndexError:
         return False
-    pos += 1
 
     # we know that it's going to be a valid blockquote,
     # so no point trying to find the end of it in silent mode
     if silent:
         return True
 
-    # set offset past spaces and ">"
-    initial = offset = state.sCount[startLine] + 1
-
-    try:
-        second_char: str | None = state.src[pos]
-    except IndexError:
-        second_char = None
-
-    # skip one optional space after '>'
-    if second_char == " ":
-        # ' >   test '
-        #     ^ -- position start of line here:
-        pos += 1
-        initial += 1
-        offset += 1
-        adjustTab = False
-        spaceAfterMarker = True
-    elif second_char == "\t":
-        spaceAfterMarker = True
-
-        if (state.bsCount[startLine] + offset) % 4 == 3:
-            # '  >\t  test '
-            #       ^ -- position start of line here (tab has width==1)
-            pos += 1
-            initial += 1
-            offset += 1
-            adjustTab = False
-        else:
-            # ' >\t  test '
-            #    ^ -- position start of line here + shift bsCount slightly
-            #         to make extra space appear
-            adjustTab = True
-
-    else:
-        spaceAfterMarker = False
-
-    oldBMarks = [state.bMarks[startLine]]
-    state.bMarks[startLine] = pos
-
-    while pos < max:
-        ch = state.src[pos]
-
-        if isStrSpace(ch):
-            if ch == "\t":
-                offset += (
-                    4
-                    - (offset + state.bsCount[startLine] + (1 if adjustTab else 0)) % 4
-                )
-            else:
-                offset += 1
-
-        else:
-            break
-
-        pos += 1
-
-    oldBSCount = [state.bsCount[startLine]]
-    state.bsCount[startLine] = (
-        state.sCount[startLine] + 1 + (1 if spaceAfterMarker else 0)
-    )
-
-    lastLineEmpty = pos >= max
-
-    oldSCount = [state.sCount[startLine]]
-    state.sCount[startLine] = offset - initial
-
-    oldTShift = [state.tShift[startLine]]
-    state.tShift[startLine] = pos - state.bMarks[startLine]
+    oldBMarks = []
+    oldBSCount = []
+    oldSCount = []
+    oldTShift = []
 
     terminatorRules = state.md.block.ruler.getRules("blockquote")
 
@@ -127,8 +62,8 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool) ->
     #      - - -
     #     ```
 
-    # for (nextLine = startLine + 1; nextLine < endLine; nextLine++) {
-    nextLine = startLine + 1
+    # for (nextLine = startLine; nextLine < endLine; nextLine++) {
+    nextLine = startLine
     while nextLine < endLine:
         # check if it's outdented, i.e. it's inside list item and indented
         # less than said list item:
@@ -153,7 +88,7 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool) ->
             # This line is inside the blockquote.
 
             # set offset past spaces and ">"
-            initial = offset = state.sCount[nextLine] + 1
+            initial = state.sCount[nextLine] + 1
 
             try:
                 next_char: str | None = state.src[pos]
@@ -166,18 +101,16 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool) ->
                 #     ^ -- position start of line here:
                 pos += 1
                 initial += 1
-                offset += 1
                 adjustTab = False
                 spaceAfterMarker = True
             elif next_char == "\t":
                 spaceAfterMarker = True
 
-                if (state.bsCount[nextLine] + offset) % 4 == 3:
+                if (state.bsCount[nextLine] + initial) % 4 == 3:
                     # '  >\t  test '
                     #       ^ -- position start of line here (tab has width==1)
                     pos += 1
                     initial += 1
-                    offset += 1
                     adjustTab = False
                 else:
                     # ' >\t  test '
@@ -188,6 +121,7 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool) ->
             else:
                 spaceAfterMarker = False
 
+            offset = initial
             oldBMarks.append(state.bMarks[nextLine])
             state.bMarks[nextLine] = pos