jpoelen added a comment.
I agree that some rules are excessive or even false, but the last time I
watched (it was several months ago now), some regexes suggested to be
DetectString seemed to have writing errors (mainly `\\` count as 2 characters
in xml) and some more or less useful propositi
dhaumann added a comment.
Herald added a project: Kate.
Herald edited subscribers, added: kde-frameworks-devel, kwrite-devel; removed:
Frameworks.
What do we do with this?
I think some of the checks are a bit too aggressive, such as folding multiple
regexps into a single one: This makes
jpoelen updated this revision to Diff 30448.
jpoelen added a comment.
New suggestions:
- Regex "^xyz\b" -> WordDetect with column=0
- Regex "^xyz" with xyz a string of more than 2 characters -> StringDetext
with column=0
Output:
- Double quote are no longer escaped
- "^\\s*"
jpoelen updated this revision to Diff 29869.
jpoelen added a comment.
More suggestions and fixes some false positives.
REPOSITORY
R216 Syntax Highlighting
CHANGES SINCE LAST UPDATE
https://phabricator.kde.org/D10621?vs=27439&id=29869
BRANCH
suggest
REVISION DETAIL
https://phabricato
jpoelen added a comment.
I just tried with a format containing 5000 `` rules. This gives 2000 to 6000 more
allocation than with WordDetect. Same for StringDetect vs Detect2Chars.
At the speed level, for a single rule and a file with 13 * 8000 `aa `,
WordDetect is 10% faster, but there is
dhaumann added a comment.
@jpoelen What would be interesting is to check which optimizations are really
an improvement. Because we should either get a significant speed boost (e.g.
RegExpr -> WordDetect), or at least reduce memory allocations (possibly
StringDetect -> Detect2Chars and Detect
jpoelen added a comment.
I've already planned to fix all the suggestions, but I lost a lot of time
updating the `sql*.xml` files and came across a very strange bug related to
QRegularExpression by adding some tests. At the moment, I do not know if it's
my tool to test, my version of Qt that
vkrause added a comment.
Very nice idea! RegExp rules are the by far biggest cost factor, so every
single one we can get rid of is good :)
REPOSITORY
R216 Syntax Highlighting
REVISION DETAIL
https://phabricator.kde.org/D10621
To: jpoelen, dhaumann
Cc: vkrause, #framework_syntax_highligh
dhaumann added a comment.
Just for info:
https://kate-editor.org/2018/03/10/improving-syntax-highlighting-files/ tries
to reach a broader audience for getting patches.
REPOSITORY
R216 Syntax Highlighting
REVISION DETAIL
https://phabricator.kde.org/D10621
To: jpoelen, dhaumann
Cc: #fram
dhaumann requested changes to this revision.
dhaumann added a comment.
This revision now requires changes to proceed.
Btw, current generated list finds 1658 issues: https://paste.kde.org/p7iareuxc
Not all of them are correct, for instance
zsh.xml" line 919 RegExpr candidate for "De
dhaumann added a comment.
As quick comment: I like the idea of programmatically suggesting better usage
of rules. I did not yet fine the time to look into this, so I do not know how
much output this currently generates.
In any case, for the highlighting files we ship with the syntax high
jpoelen added a project: Framework: Syntax Highlighting.
jpoelen added a subscriber: Framework: Syntax Highlighting.
REPOSITORY
R216 Syntax Highlighting
REVISION DETAIL
https://phabricator.kde.org/D10621
To: jpoelen, dhaumann
Cc: #framework_syntax_highlighting, dhaumann, #frameworks, michael
jpoelen created this revision.
jpoelen added a reviewer: dhaumann.
Restricted Application added a project: Frameworks.
Restricted Application added a subscriber: Frameworks.
jpoelen requested review of this revision.
REVISION SUMMARY
Proposes mergers of rules and the replacement of:
- RegEx
13 matches
Mail list logo
