jetbrains-jdk: fix with structured attrs#425529
Merged
fabianhjr merged 1 commit intoNixOS:masterfrom Jul 24, 2025
Merged
Conversation
13 tasks
nixpkgs-ci
bot
added
10.rebuild-linux: 11-100
symphorien
force-pushed
the
jdk_structured_attrs
branch
2 times, most recently
from
9d1c078 to
98b27c2
Compare
Contributor
|
Thanks for quick response! It might be issue with my setup, have you tried running some IDE with these changes? |
Member
Author
|
I don't know how to use java at all |
symphorien
changed the title
Contributor
I guess quick hack like this can also be applied here for now (both set to false) |
Member
Author
did you confirm that is solves your font issue ? I had a look at a diffoscope between this PR and this pr plus disabling structured attrs and separatedebuginfo and the diff does has no significant matches for "font". Sometimes mixing font cache from stable with fontconfig from unstable fails, could it be the issue here ? |
Contributor
Yep, basically rolling back to how it was makes the JDK usable (aka no changes from this PR at all):
I only use nixos-unstable flake, so mixing channels is not possible. LLM mentions that the I don't have any mental energy to check if it makes sense right now |
Contributor
It seems I am not alone with this issue.
So this is the only working solution for now Edit: alternative PR with some jdk fix #419026 |
the fontconfig addition solves the following runtime error: Exception in thread "main" java.lang.RuntimeException: Fontconfig head is null, check your fonts or fonts configuration which surfaced during review. How is it caused by the switch to __structiredAttrs ? Good question.
3 tasks
Contributor
|
13 tasks
Member
Author
symphorien
force-pushed
the
jdk_structured_attrs
branch
from
98b27c2 to
4cfb945
Compare
Member
Author
|
I pushed something that appears to fix the font issue. please review. |
symphorien
force-pushed
the
jdk_structured_attrs
branch
from
4cfb945 to
0bfaccd
Compare
Contributor
PyCharm builds and works, and jcef does not segfault in markdown! (unlike the fix from the alternative PR) However, there is a cursor theme bug (no theme), like in the days when this stuff was not built from source. I never figured root cause of it. This feels so hackish overall :c |
Member
|
@symphorien Thanks for the patch. @jys1670 Thanks for testing PyCharm! |
aoli-al
approved these changes
Jul 16, 2025
aoli-al
added
12.approvals: 1
prehonor
reviewed
Jul 17, 2025
Contributor
|
Member
Author
|
@jys1670 sorry I'm a bit unfamiliar with the matters so maybe all this should be obvious, but please be more precise in your comments.
what alternative PR. What steps do you use to trigger the segfault ?
what does that mean ? steps to reproduce ? Screenshot ?
why ? |
Member
Author
|
the aarch64 failures are due to libxml2 having changed its soname from .so.2 to .so.16 so I won't fix those. I think this PR is ready. |
Contributor
|
Sorry, perhaps my English is lacking.
This PR #419026 fixes the JDK build too, but takes a slightly different approach.
This means that inside PyCharm you get the default X11 mouse pointer/cursor, while all other applications use whatever the user has set globally. To reproduce: just build an IDE for your favorite language and open a project with it. Look around source files/markdown readmes and notice weird cursor.
Again, to fix the build in master without any side effects or problems, I just added two lines above this line: And it builds and works (just like K900 did with openjdk8 to fix build) My point is: it feels like what you're fixing here should normally be fixed in the original OpenJDK derivation first. For example, I don't understand why fontconfig is suddenly explicitly needed. It's in the buildInputs of the original OpenJDK and used to work, but now suddenly doesn't? It's just weird.
This is fixed in the alternative PR. In any case, I don't want to stop/nitpick anything. If people support merging either of two PRs, let it be so. |
13 tasks
Contributor
See my PR - but in short: It was always explicit, moving to structured attrs had an unintended side effect though. The OpenJDK package explicitly sets As for the reason why its there in the first place: OpenJDK seems to use |
Contributor
|
Member
|
I got the following stack trace while trying to open idea-community-bin with option This can be fixed with the following patch: diff --git a/pkgs/development/compilers/jetbrains-jdk/default.nix b/pkgs/development/compilers/jetbrains-jdk/default.nix
index c8ee60822c7d..2729ffdd12ed 100644
--- a/pkgs/development/compilers/jetbrains-jdk/default.nix
+++ b/pkgs/development/compilers/jetbrains-jdk/default.nix
@@ -20,6 +20,7 @@
libX11,
libXext,
libxcb,
+ libxkbcommon,
nss,
nspr,
libdrm,
@@ -141,6 +142,7 @@ jdk.overrideAttrs (oldAttrs: rec {
libX11
libXext
libxcb
+ libxkbcommon
nss
nspr
libdrm
|
Contributor
|
I'm waiting for a fix to land in nix-25.05. Does anybody have a workaround that I can apply without making my own fork of nixpkgs? The solutions that use an override for the jdk don't seem to work reliably. |
Member
|
NixOS 25.05 shouldn't be affected, afaik this only affects unstable? Or did something change? I pull Jetbrains from unstable and can't test stable right now, so I have no clue. |
Member
|
fabianhjr
approved these changes
Jul 24, 2025
fabianhjr
added
12.approvals: 2
Contributor
|
@theCapypara not at my computer right now, but AFAICT this jetbrains jdk doesn't build on stable right now. I was assuming that the change that broke it had propagated from unstable to stable. |
13 tasks
3 tasks
pbek
added a commit
to pbek/nixcfg
that referenced
this pull request
Jul 26, 2025
Signed-off-by: Patrizio Bekerle <[email protected]>
13 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
9 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)Add a 👍 reaction to pull requests you find important.