[gpad] Fix crash in streaming web canvas after TColor::DefinedColors(1)#22603
Merged
Conversation
TColor::DefinedColors(1) sets the global color bookkeeping to the sticky "always store colors" mode (gLastDefinedColors = -1). When a canvas is then serialized to JSON for the web display - as JupyROOT does to show a canvas inline in a notebook - TWebCanvas::CreatePadSnapshot temporarily detaches the pad primitives (fPrimitives = nullptr) and calls the no-arg TColor::DefinedColors() to clear the change flag so the canvas streamer skips color storage. In sticky mode that call returns kTRUE early without resetting the flag, so TCanvas::Streamer entered the color-storing branch and dereferenced the null fPrimitives, crashing with a segmentation violation. Guard the color-storing branch with a null check on fPrimitives. Colors and the palette are delivered separately by TWebCanvas::AddColorsPalette in that path, and normal file saving (where fPrimitives is never null) is unaffected. Add a regression test that reproduces the original report by drawing a canvas and calling TWebCanvas::CreateCanvasJSON after TColor::DefinedColors(1); it segfaults without this fix and passes with it. The JupyROOT notebook test is registered without output comparison (it draws a canvas, so its output is not reproducible), matching Cpp_IMT_Canvas.ipynb. Closes root-project#20018. 🤖 Done with the help of [Claude Code](https://claude.com/claude-code) (Claude Opus 4.8)
guitargeek
requested review from
bellenot,
couet,
dpiparo,
linev and
vepadulano
as code owners
2 tasks
Test Results 20 files 20 suites 2d 23h 4m 31s ⏱️ Results for commit 7f070cf. ♻️ This comment has been updated with latest results. |
Contributor
Author
|
/backport to 6.40 |
|
Preparing to backport PR #22603 to branch 6.40 requested by guitargeek |
|
Something went wrong when assigning the PR or setting labels @guitargeek please see the logs |
|
This PR has been backported to branch 6.40: #22609 |
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
4 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.
TColor::DefinedColors(1) sets the global color bookkeeping to the sticky "always store colors" mode (gLastDefinedColors = -1). When a canvas is then serialized to JSON for the web display - as JupyROOT does to show a canvas inline in a notebook - TWebCanvas::CreatePadSnapshot temporarily detaches the pad primitives (fPrimitives = nullptr) and calls the no-arg TColor::DefinedColors() to clear the change flag so the canvas streamer skips color storage. In sticky mode that call returns kTRUE early without resetting the flag, so TCanvas::Streamer entered the color-storing branch and dereferenced the null fPrimitives, crashing with a segmentation violation.
Guard the color-storing branch with a null check on fPrimitives. Colors and the palette are delivered separately by TWebCanvas::AddColorsPalette in that path, and normal file saving (where fPrimitives is never null) is unaffected.
Add a regression test that reproduces the original report by drawing a canvas and calling TWebCanvas::CreateCanvasJSON after TColor::DefinedColors(1); it segfaults without this fix and passes with it. The JupyROOT notebook test is registered without output comparison (it draws a canvas, so its output is not reproducible), matching Cpp_IMT_Canvas.ipynb.
Closes #20018.
🤖 Done with the help of Claude Code (Claude Opus 4.8)
Since the added notebook tests are a bit abstract without visual feedback, I attach here the screenshots of running the original reproducer from #20018 in a local notebook.
Before this PR:
After this PR:
