Repository navigation
Render graphs in parallel - #740
Open
jacobwilliams wants to merge 2 commits into
Open
jacobwilliams wants to merge 2 commits into
jacobwilliams wants to merge 2 commits into
Conversation
Running graphviz was done serially when each graph was constructed, and dominated the "Generating graphs" stage on large projects. Graph construction now only builds the dot source; `FortranGraph.render` runs graphviz, and `GraphManager.graph_all` renders all graphs using a thread pool sized by the `parallel` setting (0 still means serial). Since rendering happens in a subprocess, threads give real parallelism without needing to pickle the graphs. `svg_src` and `scaled` render lazily if accessed before an explicit render. `output_graphs` uses the same thread pool instead of `process_map`. This fixes saving graphs to `graph_dir`, which with the default `parallel` setting failed to pickle the graphs and wrote no files. Procedures' "uses" graphs are now also saved. Also remove the unused `Documentation.parallel` attribute. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
json-fortran, FLAP, FiNeR, and VTKFortran have all moved or removed their Ford project files upstream, causing the corpus regression tests to fail before Ford parsed anything: - json-fortran: `json-fortran.md` -> `ford.md` - FLAP, FiNeR, VTKFortran: `doc/main_page.md` -> `docs/ford.md` Fixes Fortran-FOSS-Programmers#741
Member
Author
|
@Fortran-FOSS-Programmers/contributors any takers for a reviewing this? |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Speed up the "Generating graphs" stage on large projects by running them in parallel.
Disclaimer: this is from Claude Opus 5.5. It seems to work for me, but should be checked. The graph generating was a huge bottleneck for large projects, so this should be a good feature to have.
Fixes #118
Also Fixes #741