Correctly match newlines whitin htmltags. #122
Conversation
lib/Caxy/HtmlDiff/AbstractDiff.php
Outdated
| @@ -433,7 +433,7 @@ protected function convertHtmlToListOfWords(string $text) : array | |||
| // Normalize no-break-spaces to regular spaces | |||
| $text = str_replace("\xc2\xa0", ' ', $text); | |||
|
|
|||
| preg_match_all('/<.+?>|[^<]+/mu', $text, $sentencesAndTags, PREG_SPLIT_NO_EMPTY); | |||
| preg_match_all('/<[^>]+?>|[^<]+/mu', $text, $sentencesAndTags, PREG_SPLIT_NO_EMPTY); | |||
There was a problem hiding this comment.
Alternatively we could add the s (PCRE_DOTALL) modifier so that the dot would match newlines as well.
However, this might seem more clear? Are there situations where this change would actually affect the outcome in a negative way?
I think it's technically allowed to have > within an attribute value of an HTML tag...
<div data-some-value="1 > 0">that's technically valid I think?
So are we better off adding the s modifier to solve this instead?
There was a problem hiding this comment.
I've tested all regexes (current, mine, yours) with the example you provided and they all break the tag in the data-some-value contents. So this is currently a problem and i dont know how to fix this particular issue.
However, i do agree the dotall modifier is more clear, i will change the merge request to use that.
When using newlines within htmltags the regular expression used doesn't account for newlines, this pull request aims to fix this issue.