https://bugs.documentfoundation.org/show_bug.cgi?id=150794

Mike Kaganski <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |[email protected]
         Resolution|---                         |NOTABUG

--- Comment #1 from Mike Kaganski <[email protected]> ---
While the general inability of Writer to search across several paragraphs is
itself a real issue, this specific case is not a bug.

1. There is *no* "paragraph character" in Writer at all. Paragraphs are
separate objects, and there is no "character" that separates them. Writer
document is not a plain text with a CR, LF, or other single character having
that special meaning.

2. More importantly: in regular expressions, $ is not a metacharacter to *match
such a end-of-paragraph character*, but a predicate (a simple form of
look-ahead assertion), which itself does not match any single character, only
makes the adjacent expression only match when it's in the end (of either a
paragraph, or a whole text). It should *not* "select" the something that
separates paragraphs, so Kate behaves not "correctly" in that regards, but it
uses some extension of the convention.

See the regular expression syntax documentation [1] that is used in
LibreOffice, which explicitly tells:

> $     Match at the end of a line. Line terminating characters are \u000a,
>       \u000b, \u000c, \u000d, \u0085, \u2028, \u2029 and the sequence \u000d
>       \u000a.

Note the "match /at/ the end ...", with the "at" meaning that not the "end"
itself is matched, but characters that appear at that end.

You may also test regexes and results of replacement of "1$" with anything at
resources like https://regex101.com/.

[1] https://unicode-org.github.io/icu/userguide/strings/regexp.html

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to