From: Benjamin Mako Hill Date: Fri, 28 Apr 2023 21:30:42 +0000 (-0700) Subject: make sure that content is defined before testing for search patterns X-Git-Url: https://code.communitydata.science/mediawiki_dump_tools.git/commitdiff_plain/7e6cd5b38610e0c1df5c8764e2527135a7ede38f?ds=inline make sure that content is defined before testing for search patterns This appears to have been causing a bug with comments/text that were deleted. Kaylea fixed and I adapated the code. --- diff --git a/wikiq b/wikiq index 4a5c129..2c1ef38 100755 --- a/wikiq +++ b/wikiq @@ -146,7 +146,7 @@ class RegexPair(object): if self.has_groups: # if there are matches of some sort in this revision content, fill the lists for each cap_group - if self.pattern.search(content) is not None: + if content is not None and self.pattern.search(content) is not None: m = self.pattern.finditer(content) matchobjects = list(m) @@ -174,7 +174,7 @@ class RegexPair(object): # there are no capture groups, we just search for all the matches of the regex else: #given that there are matches to be made - if self.pattern.search(content) is not None: + if content is not None and self.pattern.search(content) is not None: m = self.pattern.findall(content) temp_dict[self.label] = ', '.join(m) else: