Speedup line_offset property

* Replace dynamic regex with string find operation * Add cache of where each line starts so we don't have quadratic behavior identifying line numbers when importing large chunks of html

markdown/htmlparser.py

@@ -83,6 +83,9 @@ def __init__(self, md, *args, **kwargs):
  # Block tags that should contain no content (self closing)
  self.empty_tags = set(['hr'])
 
+ self.lineno_start_cache = [0]
+
+
  # This calls self.reset
  super().__init__(*args, **kwargs)
  self.md = md
@@ -94,6 +97,8 @@ def reset(self):
  self.stack = [] # When `inraw==True`, stack contains a list of tags
  self._cache = []
  self.cleandoc = []
+ self.lineno_start_cache = [0]
+
  super().reset()
 
  def close(self):
@@ -114,6 +119,15 @@ def close(self):
  @property
  def line_offset(self) -> int:
  """Returns char index in `self.rawdata` for the start of the current line. """
+ for ii in range(len(self.lineno_start_cache)-1, self.lineno-1):
+ last_line_start_pos = self.lineno_start_cache[ii]
+ lf_pos = self.rawdata.find('\n', last_line_start_pos)
+ if lf_pos == -1:
+ # No more newlines found. Use end of rawdata.
+ lf_pos = len(self.rawdata)
+ self.lineno_start_cache.append(lf_pos+1)
+
+ return self.lineno_start_cache[self.lineno-1]
  if self.lineno > 1 and '\n' in self.rawdata:
  m = re.match(r'([^\n]*\n){{{}}}'.format(self.lineno-1), self.rawdata)
  if m: