Implement JS-esque StateBase.srcCharCodeAt

markdown_it/helpers/parse_link_label.py

@@ -18,7 +18,7 @@ def parseLinkLabel(state: StateInline, start: int, disableNested: bool = False)
     level = 1
 
     while state.pos < state.posMax:
-        marker = state.srcCharCode[state.pos]
+        marker = state.srcCharCodeAt(state.pos)
         if marker == 0x5D:  # /* ] */)
             level -= 1
             if level == 0:

markdown_it/port.yaml

@@ -23,8 +23,9 @@
       to manipulate `Token.attrs`, which have an identical signature to those upstream.
     - Use python version of `charCodeAt`
     - |
-      Reduce use of charCodeAt() by storing char codes in a srcCharCodes attribute for state
-      objects and sharing those whenever possible
+      Reduce use of charCodeAt() by storing char codes in an `_ords` attribute
+      (accessible via `srcCharCodeAt` method) for state objects and sharing
+      those whenever possible.
       This provides a significant performance boost
     - |
       In markdown_it/rules_block/reference.py,

markdown_it/ruler.py

@@ -26,14 +26,16 @@ class Ruler
     TYPE_CHECKING,
     Union,
 )
+import warnings
+
 import attr
 
 if TYPE_CHECKING:
     from markdown_it import MarkdownIt
 
 
 class StateBase:
-    srcCharCode: Tuple[int, ...]
+    _ords: Tuple[int, ...]
 
     def __init__(self, src: str, md: "MarkdownIt", env: MutableMapping):
         self.src = src
@@ -47,7 +49,22 @@ def src(self) -> str:
     @src.setter
     def src(self, value: str) -> None:
         self._src = value
-        self.srcCharCode = tuple(ord(c) for c in self.src)
+        self._ords = tuple(ord(c) for c in self.src)
+
+    @property
+    def srcCharCode(self) -> Tuple[int, ...]:
+        warnings.warn(
+            "`StateBase.srcCharCode` is deprecated. Use `StateBase.srcCharCodeAt`",
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return self._ords
+
+    def srcCharCodeAt(self, idx: int) -> Optional[int]:
+        try:
+            return self._ords[idx]
+        except IndexError:
+            return None
 
 
 # The first positional arg is always a subtype of `StateBase`. Other

markdown_it/rules_block/blockquote.py

@@ -1,6 +1,5 @@
 # Block quotes
 import logging
-from typing import Optional
 
 from .state_block import StateBlock
 from ..common.utils import isSpace
@@ -23,7 +22,7 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool):
         return False
 
     # check the block quote marker
-    if state.srcCharCode[pos] != 0x3E:  # /* > */
+    if state.srcCharCodeAt(pos) != 0x3E:  # /* > */
         return False
     pos += 1
 
@@ -35,21 +34,16 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool):
     # set offset past spaces and ">"
     initial = offset = state.sCount[startLine] + 1
 
-    try:
-        second_char_code: Optional[int] = state.srcCharCode[pos]
-    except IndexError:
-        second_char_code = None
-
     # skip one optional space after '>'
-    if second_char_code == 0x20:  # /* space */
+    if state.srcCharCodeAt(pos) == 0x20:  # /* space */
         # ' >   test '
         #     ^ -- position start of line here:
         pos += 1
         initial += 1
         offset += 1
         adjustTab = False
         spaceAfterMarker = True
-    elif second_char_code == 0x09:  # /* tab */
+    elif state.srcCharCodeAt(pos) == 0x09:  # /* tab */
         spaceAfterMarker = True
 
         if (state.bsCount[startLine] + offset) % 4 == 3:
@@ -72,7 +66,7 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool):
     state.bMarks[startLine] = pos
 
     while pos < max:
-        ch = state.srcCharCode[pos]
+        ch = state.srcCharCodeAt(pos)
 
         if isSpace(ch):
             if ch == 0x09:  # / tab /
@@ -146,29 +140,24 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool):
             # Case 1: line is not inside the blockquote, and this line is empty.
             break
 
-        evaluatesTrue = state.srcCharCode[pos] == 0x3E and not isOutdented  # /* > */
+        evaluatesTrue = state.srcCharCodeAt(pos) == 0x3E and not isOutdented  # /* > */
         pos += 1
         if evaluatesTrue:
             # This line is inside the blockquote.
 
             # set offset past spaces and ">"
             initial = offset = state.sCount[nextLine] + 1
 
-            try:
-                next_char: Optional[int] = state.srcCharCode[pos]
-            except IndexError:
-                next_char = None
-
             # skip one optional space after '>'
-            if next_char == 0x20:  # /* space */
+            if state.srcCharCodeAt(pos) == 0x20:  # /* space */
                 # ' >   test '
                 #     ^ -- position start of line here:
                 pos += 1
                 initial += 1
                 offset += 1
                 adjustTab = False
                 spaceAfterMarker = True
-            elif next_char == 0x09:  # /* tab */
+            elif state.srcCharCodeAt(pos) == 0x09:  # /* tab */
                 spaceAfterMarker = True
 
                 if (state.bsCount[nextLine] + offset) % 4 == 3:
@@ -191,7 +180,7 @@ def blockquote(state: StateBlock, startLine: int, endLine: int, silent: bool):
             state.bMarks[nextLine] = pos
 
             while pos < max:
-                ch = state.srcCharCode[pos]
+                ch = state.srcCharCodeAt(pos)
 
                 if isSpace(ch):
                     if ch == 0x09:

markdown_it/rules_block/fence.py

@@ -21,7 +21,7 @@ def fence(state: StateBlock, startLine: int, endLine: int, silent: bool):
     if pos + 3 > maximum:
         return False
 
-    marker = state.srcCharCode[pos]
+    marker = state.srcCharCodeAt(pos)
 
     # /* ~ */  /* ` */
     if marker != 0x7E and marker != 0x60:
@@ -67,7 +67,7 @@ def fence(state: StateBlock, startLine: int, endLine: int, silent: bool):
             #  test
             break
 
-        if state.srcCharCode[pos] != marker:
+        if state.srcCharCodeAt(pos) != marker:
             continue
 
         if state.sCount[nextLine] - state.blkIndent >= 4:

markdown_it/rules_block/heading.py

@@ -1,6 +1,5 @@
 """ Atex heading (#, ##, ...) """
 import logging
-from typing import Optional
 
 from .state_block import StateBlock
 from ..common.utils import isSpace
@@ -19,7 +18,7 @@ def heading(state: StateBlock, startLine: int, endLine: int, silent: bool):
     if state.sCount[startLine] - state.blkIndent >= 4:
         return False
 
-    ch: Optional[int] = state.srcCharCode[pos]
+    ch = state.srcCharCodeAt(pos)
 
     # /* # */
     if ch != 0x23 or pos >= maximum:
@@ -28,18 +27,12 @@ def heading(state: StateBlock, startLine: int, endLine: int, silent: bool):
     # count heading level
     level = 1
     pos += 1
-    try:
-        ch = state.srcCharCode[pos]
-    except IndexError:
-        ch = None
+    ch = state.srcCharCodeAt(pos)
     # /* # */
     while ch == 0x23 and pos < maximum and level <= 6:
         level += 1
         pos += 1
-        try:
-            ch = state.srcCharCode[pos]
-        except IndexError:
-            ch = None
+        ch = state.srcCharCodeAt(pos)
 
     if level > 6 or (pos < maximum and not isSpace(ch)):
         return False
@@ -51,7 +44,7 @@ def heading(state: StateBlock, startLine: int, endLine: int, silent: bool):
 
     maximum = state.skipSpacesBack(maximum, pos)
     tmp = state.skipCharsBack(maximum, 0x23, pos)  # #
-    if tmp > pos and isSpace(state.srcCharCode[tmp - 1]):
+    if tmp > pos and isSpace(state.srcCharCodeAt(tmp - 1)):
         maximum = tmp
 
     state.line = startLine + 1

markdown_it/rules_block/hr.py

@@ -22,7 +22,7 @@ def hr(state: StateBlock, startLine: int, endLine: int, silent: bool):
     if state.sCount[startLine] - state.blkIndent >= 4:
         return False
 
-    marker = state.srcCharCode[pos]
+    marker = state.srcCharCodeAt(pos)
     pos += 1
 
     # Check hr marker: /* * */ /* - */ /* _ */
@@ -33,7 +33,7 @@ def hr(state: StateBlock, startLine: int, endLine: int, silent: bool):
 
     cnt = 1
     while pos < maximum:
-        ch = state.srcCharCode[pos]
+        ch = state.srcCharCodeAt(pos)
         pos += 1
         if ch != marker and not isSpace(ch):
             return False

markdown_it/rules_block/html_block.py

@@ -44,7 +44,7 @@ def html_block(state: StateBlock, startLine: int, endLine: int, silent: bool):
     if not state.md.options.get("html", None):
         return False
 
-    if state.srcCharCode[pos] != 0x3C:  # /* < */
+    if state.srcCharCodeAt(pos) != 0x3C:  # /* < */
         return False
 
     lineText = state.src[pos:maximum]

markdown_it/rules_block/lheading.py

@@ -37,7 +37,7 @@ def lheading(state: StateBlock, startLine: int, endLine: int, silent: bool):
             maximum = state.eMarks[nextLine]
 
             if pos < maximum:
-                marker = state.srcCharCode[pos]
+                marker = state.srcCharCodeAt(pos)
 
                 # /* - */  /* = */
                 if marker == 0x2D or marker == 0x3D:
@@ -74,6 +74,7 @@ def lheading(state: StateBlock, startLine: int, endLine: int, silent: bool):
     state.line = nextLine + 1
 
     token = state.push("heading_open", "h" + str(level), 1)
+    assert marker is not None
     token.markup = chr(marker)
     token.map = [startLine, state.line]
 

markdown_it/rules_block/list.py

@@ -14,14 +14,14 @@ def skipBulletListMarker(state: StateBlock, startLine: int):
     pos = state.bMarks[startLine] + state.tShift[startLine]
     maximum = state.eMarks[startLine]
 
-    marker = state.srcCharCode[pos]
+    marker = state.srcCharCodeAt(pos)
     pos += 1
     # Check bullet /* * */ /* - */ /* + */
     if marker != 0x2A and marker != 0x2D and marker != 0x2B:
         return -1
 
     if pos < maximum:
-        ch = state.srcCharCode[pos]
+        ch = state.srcCharCodeAt(pos)
 
         if not isSpace(ch):
             # " -test " - is not a list item
@@ -42,7 +42,8 @@ def skipOrderedListMarker(state: StateBlock, startLine: int):
     if pos + 1 >= maximum:
         return -1
 
-    ch = state.srcCharCode[pos]
+    ch = state.srcCharCodeAt(pos)
+    assert ch is not None
     pos += 1
 
     # /* 0 */  /* 9 */
@@ -54,7 +55,8 @@ def skipOrderedListMarker(state: StateBlock, startLine: int):
         if pos >= maximum:
             return -1
 
-        ch = state.srcCharCode[pos]
+        ch = state.srcCharCodeAt(pos)
+        assert ch is not None
         pos += 1
 
         # /* 0 */  /* 9 */
@@ -74,7 +76,7 @@ def skipOrderedListMarker(state: StateBlock, startLine: int):
         return -1
 
     if pos < maximum:
-        ch = state.srcCharCode[pos]
+        ch = state.srcCharCodeAt(pos)
 
         if not isSpace(ch):
             # " 1.test " - is not a list item
@@ -156,7 +158,8 @@ def list_block(state: StateBlock, startLine: int, endLine: int, silent: bool):
             return False
 
     # We should terminate list on style change. Remember first one to compare.
-    markerCharCode = state.srcCharCode[posAfterMarker - 1]
+    markerCharCode = state.srcCharCodeAt(posAfterMarker - 1)
+    assert markerCharCode is not None
 
     # For validation mode we can terminate immediately
     if silent:
@@ -198,7 +201,7 @@ def list_block(state: StateBlock, startLine: int, endLine: int, silent: bool):
         )
 
         while pos < maximum:
-            ch = state.srcCharCode[pos]
+            ch = state.srcCharCodeAt(pos)
 
             if ch == 0x09:  # \t
                 offset += 4 - (offset + state.bsCount[nextLine]) % 4
@@ -318,7 +321,7 @@ def list_block(state: StateBlock, startLine: int, endLine: int, silent: bool):
             if posAfterMarker < 0:
                 break
 
-        if markerCharCode != state.srcCharCode[posAfterMarker - 1]:
+        if markerCharCode != state.srcCharCodeAt(posAfterMarker - 1):
             break
 
     # Finalize list

markdown_it/rules_block/reference.py

@@ -22,17 +22,17 @@ def reference(state: StateBlock, startLine, _endLine, silent):
     if state.sCount[startLine] - state.blkIndent >= 4:
         return False
 
-    if state.srcCharCode[pos] != 0x5B:  # /* [ */
+    if state.srcCharCodeAt(pos) != 0x5B:  # /* [ */
         return False
 
     # Simple check to quickly interrupt scan on [link](url) at the start of line.
     # Can be useful on practice: https:#github.com/markdown-it/markdown-it/issues/54
     while pos < maximum:
         # /* ] */  /* \ */  /* : */
-        if state.srcCharCode[pos] == 0x5D and state.srcCharCode[pos - 1] != 0x5C:
+        if state.srcCharCodeAt(pos) == 0x5D and state.srcCharCodeAt(pos - 1) != 0x5C:
             if pos + 1 == maximum:
                 return False
-            if state.srcCharCode[pos + 1] != 0x3A:
+            if state.srcCharCodeAt(pos + 1) != 0x3A:
                 return False
             break
         pos += 1