Modification to truncate_utf8() to support word boundary-aware chopping. The added parameter is a flag that defaults to FALSE.

An Example:

I wish I had sextants to use to navigate this sea

(Currently, you don't want to chop to 16 chars)

with word boundary-aware truncate, this will truncate to the nearest word boundary after the length:

I wish I had

This is also helpful when people post URLs as content. (otherwise we'd end up with wrong URLs)

CommentFileSizeAuthor
common.inc.patch.txt946 bytesarnabdotorg

Comments

arnabdotorg’s picture

arnabdotorg commented
Steven’s picture

Steven commented

- The code does not conform to the coding standards.
- It's not very flexible. A word boundary is more than just a space, it can be tons of characters.
- A string without spaces gets chopped to nothing. I'm not sure if this is desirable.

PS: Set the issue status correctly next time please.

arnabdotorg’s picture

arnabdotorg commented

Committed to HEAD. [this is why this is fixed]

arnabdotorg’s picture

arnabdotorg commented

- coding standards were fixed in the patch applied.
- yes, we could have a whole bunch of things for boundaries, but a space is the most "resilient", imho: (e.g. URLS, etc invalidate the use of ".", ":","/" as boundaries).
- I'd rather have correct information than mangled text.

But you're right, (2) and (3) are pretty subjective issues.

Anonymous’s picture

(not verified) commented