textpattern.github.io
textpattern.github.io copied to clipboard
textpattern
page: /tags/excerpt
URLs
- Back: {repo URL here (.md file)}
- Front: {website URL here}
Explanation of immediate task
The trim regex given as an example to trim body for excerpt seems to trip up if textile is enable in the body field (prepending the truncated text with p>), no matter what the escape attributes. I have better luck with <txp:body escape="tags" trim="/^(<[^>]*>)*((?:\w+\W+){35})\w.*$/s" replace="$2…" />, which excludes initial tags.
Resources to support effort
- https://regex101.com/r/XBYWrs/1
Current to-dos
- Maybe update example with note for textile users
Thank you for the report, but I'm not sure textile interferes here. Could you provide more details re the article body content?
Hi, I've just reinstated the trim regex from the TXP docs and made a test article using a couple of paragraphs of plain text. Screenshots attached of front and back end, firstly with Textile on then with it off.
I've only seen this happen on a couple of older sites, so may be something in the config or hosting conditions or a plugin.
I've tried the regex on the regex101 site and it removes the opening bracket from an opening p tag, so I'm guessing the regex trim happens before the transform / escape?
LEAVE TEXT UNTOUCHED
TEXTILE
Thank you. Have you also set escape="tags" attribute before trim? Unless php strip_tags() function is somehow disabled on your server (see diagnostics), it should have removed all html tags prior to trimming.
This said, the example regex might be a bit loose, indeed.