-
Notifications
You must be signed in to change notification settings - Fork 256
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Better asm syntax highlighting #538
Conversation
e4e6820
to
29eddc8
Compare
TODO:
|
5a909fb
to
7e878ff
Compare
I'm a bit confused, why does "Syntax Highlighting: None" have any color? But nice to see that those issues are tackled! |
someone didn't implement synchronisation yet |
As this auto-drops the highlighting on the hex values and also changes the code involved here, I highly suggest to implement #527 (moving the hex parts out as requested by Milian) first. Could be done in this PR, too. |
81a7e60
to
fbfbec9
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the cleanup/fixup patches should go into a separate MR because they can land directly. the rest of this patch needs some cleanup
b4a785e
to
5273ad4
Compare
9dd780e
to
5b67e29
Compare
8241728
to
565a132
Compare
Just wanted to share that #561 seems to be fixed with the last commit (only checked two different examples that did not work before) - if it should be then please add its reference to the commits / this PR. #527 is definitely solved by the "add option to hide" commit and should be referenced there. One downside: it seems that the syntax highlighting in the Disassembly does not work (AppImage).
|
@GitMensch can check you the most recent appimage? I notices that QTextLayout is very slow (multiple seconds for 50k+ lines) and added a cache. You complained about slow performance so maybe this finally fixes it. |
That is most likely, because |
... and/or because of the following?
|
Ohh, right, in that case it should fallback to using |
8963c85
to
23cc2fb
Compare
@GitMensch I looked into it. Currently I only copy the font color from |
Would there be any "thickness" it the objdump colored output that may should be copied over? Note: in any case I'd "expect" to have the objdump coloring as an extra entry of the highlighting (create a manual list including all syntax highlighters that have "asm" in them, the add "objdump" on top, if it is available). |
9ec76f9
to
5445957
Compare
5445957
to
341eed8
Compare
If it is more likely to get included faster, then we may split this PR in 4:
|
code lgtm, probably needs a rebase now but otherwise this should be ready to go in IIUC? @GitMensch is your highlighting bug resolved now in the latest iteration of this pull request? |
No, sadly #561 is not yet solved. Just tested with objdump.log, generated by "GNU objdump (GNU Binutils for Debian) 2.35.2", which doesn't support I guess that the actual highlighting is not the main problem but the splitting of the jump-visualisation and hex-values. I'm keen to recheck after a rebase (shouldn't make a big difference, I guess) and even more after a fix to this. Just a debugging note (you're likely aware of this): you may add an "objdump" script that only does a |
Thanks, I finally understand your problem and managed to reproduce it. |
this allows for a simpler feature detection in objdump since we can now fetch the help output once and then query it multiple times for different arguments
this patch adds a syntax highlighter that decoded ansi escape sequences and colors the text accordingly
QTextLayout::setFormats is really slow so this patch introduces a cache named HighlightedLine which will only call QTextLayout::setFormats if necessary
341eed8
to
953d3e0
Compare
@GitMensch It seems that #566 fixed the issue, can you check again? At least I can't reproduce it anymore after that pr was merged. I checked your files too and they worked. |
Yes, it is and the current appimage from this PR works fine there, too. As Milian is also fine with this PR it seems its ready for merge :-) |
@milianw can you merge that in, please? |
No description provided.