Markdown parser updates #18
Conversation
|
It shouldn't break in pure Python, I did test it in the console to make sure it was matching as expected and had no issues.
Maybe escape the single quote too rather than remove it? It is a valid URL character after all, and not that unusual since it's often used to enclose values in parameters
|
|
Break may have been the wrong word. What's your expected input? >>> import re
>>> def _parse_links(input_str: str) -> str:
... input_str = re.sub(
... r"\[([\w\W]+?)\]\(([\w\W]+?)\)",
... r'<a href="\2">\1</a>',
... input_str,
... flags=re.MULTILINE,
... )
... return re.sub(
... r"(?<!href=\")(https?:\/\/[-a-zA-Z0-9._~:/?#@!$&()*+,;=%]+')",
... r'<a href="\1">\1</a>',
... input_str,
... flags=re.MULTILINE,
... )
...
>>> test_str = """
... https://example.com
... [example](https://example.com)
... https://example.com'
... """
>>> print(_parse_links(test_str))
https://example.com
<a href="https://example.com">example</a>
<a href="https://example.com'">https://example.com'</a>Did you intend to only support links that were suffixed with |
That... was an oversight between converting the expression enclosure from single to double-quotes in the browser editor 😅 |
|
Added some other minor changes, doesn't seem like I broke anything 😅 |
|
Anything else holding this and #20 ? |
|
No blockers at a glance -- sorry for the slow review, I've just been incredibly busy with irl. I'll block out some time this weekend to take a closer look and/or merge. |
* Marker matches on both sides * Opener and closer is not escaped * First character inside is not a space
|
Should have addressed all notes 🤞 |
|
|
||
|
|
||
| def _parse_strong_emphasis(input_str: str) -> str: | ||
| return "\n".join( | ||
| re.sub( | ||
| r"(?:\*\*|\_\_)([\w]+?[\w\W]+?[\w]+?)(?:\*\*|\_\_)", | ||
| r"(?<!\\)\*\*(\w.*?)(?<!\\)\*\*|(?<!\\)__(\w.*?)(?<!\\)__", |
There was a problem hiding this comment.
Nice, I'm a fan of also handling the escaped character.
I think you're handling the first character being a space, but not the last?
>>> text = """
... **asdf**
... ** asdf**
... **asdf **
... ** asdf **
... """
>>> print(_parse_strong_emphasis(text))
<strong>asdf</strong>
** asdf**
<strong>asdf </strong>
** asdf **It's fine though, I'm not going to block on this (especially since the original impl was flawed as it only allowed \w which ignores symbols as valid characters).
|
I couldn't figure an elegant way to filter trailing spaces as well. While another \w at the end might handle it, it would also make it impossible to format single characters like **A** letter or _1_ digit.
Turns out a space with the {0} quantifier doesn't work as a negative 🙁
|
* Parse standalone URLs too * move single-quote back into the set * Tighten URL characters for CFM * Improve headers formatting * Simplify expressions * Update example * Parse URLs regardless of casing * open links in new tabs * Put atx-style back into docstring * Absorb more spaces and optional closing on headers * Refine bold and italics expressions by ensuring: * Marker matches on both sides * Opener and closer is not escaped * First character inside is not a space
Parsed after the markdown format since it has the added benefit of avoiding its closing brackets at the end of the URL