I want to remove HTML comments from an html text
<h1>heading</h1> <!-- comment-with-hyphen --> some text <-- con --> more text <hello></hello> more text
should result in:
<h1>heading</h1> some text <-- con --> more text <hello></hello> more text
-
Using regular expressions on a limited, known set of HTML may be appropriate. However, you should be aware that there are countless cases where it will break and it is generally not advised.grc– grc2015-01-29 06:38:49 +00:00Commented Jan 29, 2015 at 6:38
-
Related: stackoverflow.com/a/1732454/3001761jonrsharpe– jonrsharpe2015-01-29 07:57:56 +00:00Commented Jan 29, 2015 at 7:57
-
Why the downvotes on the question? If you are working on a "known set of HTML" this was a legit question.Rushabh Mehta– Rushabh Mehta2015-01-30 07:24:59 +00:00Commented Jan 30, 2015 at 7:24
-
Consider using a HTML specific library like Beatiful Soup, like this other question-solutions suggests: stackoverflow.com/questions/23299557/…hectorcanto– hectorcanto2020-04-22 00:39:48 +00:00Commented Apr 22, 2020 at 0:39
Add a comment
|
6 Answers
You shouldn't ignore Carriage return.
re.sub("(<!--.*?-->)", "", s, flags=re.DOTALL)
3 Comments
Ethan
Why shouldn't we remove the carriage returns as well?
Greg Lindahl
huazhihao's answer matches comments that have carriage returns within the comment. One of the other answers lacks flags=re.MULTILINE
fjsj
actually should be
re.DOTALL, not re.MULTILINE. It's re.DOTALL who matches \n on .html = re.sub(r"<!--(.|\s|\n)*?-->", "", html)
re.sub basically find the matching instance and replace with the second arguments. For this case, <!--(.|\s|\n)*?--> matches anything start with <!-- and end with -->. The dot and ? means anything, and the \s and \n add the cases of muti line comment.
1 Comment
jpaugh
Welcome to Stack Overflow! If the OP could understand your code by itself, he probably would not be asking. Please explain what it does, so that it provides value for those who would need to look up a regex.
Finally came up with this option:
re.sub("(<!--.*?-->)", "", t)
Adding the ? makes the search non-greedy and does not combine multiple comment tags.
Comments
Don't use regex. Use an XML parser instead, the one in the standard library is more than sufficient.
from xml.etree import ElementTree as ET
html = ET.parse("comments.html")
ET.dump(html) # Dumps to stdout
ET.write("no-comments.html", method="html") # Write to a file
1 Comment
Greg Lindahl
While this is good advice, the performance of XML parsers is much, much, much slower than this sort of regex.
re.sub("(?s)<!--.+?-->", "", s)
or
re.sub("<!--.+?-->", "", s, flags=re.DOTALL)
Comments
You could try this regex <![^<]*>
3 Comments
Greg Lindahl
Your regex matches too much -- note that the question has an example "<-- con -->", which is not an HTML comment.
dragon2fly
@GregLindahl this regex didn't match "<-- con -->" and returned the result as the OP expected.
k-den
This won't match a comment with an HTML tag inside of it, like <!-- <br/> -->