Repository navigation
Conversation
Three drifts, all verified against the code: - Solver selection was documented as the boolean flags `use_indirect`, `gpu`, `mkl`, `cudss`, `dense`. Those were removed in 3.3.0 (78758c8) in favour of `linear_solver=` and the `LinearSolver` enum, and now raise TypeError. - The extension list omitted `_scs_accelerate` (meson.build builds it unconditionally on macOS) and described `_scs_dense` as conditional on use_lapack without noting that use_lapack defaults to true, so it is effectively always built. - Key Files pointed at `scs/include/scsmodule.h` and `scs/include/scsobject.h`; there is no `scs/include/` directory. Added the undocumented `scs/py_ctrlc.c`. Also replaced the vague Solver Variants paragraph with what `_SOLVER_DISPATCH` and `_resolve_auto` actually do, and fixed the scspy.c header comment, which still referred to the front-facing Python code as "scs.py". Closes bodono#237 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Owner
|
Thanks. CLAUDE.md is being removed from version control in #247 so it can no longer drift, and the scspy.c comment fix is included there. Closing. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #237
Three drifts in
CLAUDE.md, each verified against the code rather than inferred:1. Solver selection documented as removed flags. It described selection via
use_indirect,gpu,mkl,cudss,dense. Those were removed in 3.3.0 (78758c8, "Add LinearSolver enum, remove boolean solver-selection flags", #189) in favour oflinear_solver=and theLinearSolverenum. Checked against a built extension — each removed kwarg now raises:2. Incomplete extension list. It omitted
_scs_accelerate, whichmeson.build:424builds unconditionally underhost_machine.system() == 'darwin', and described_scs_denseas conditional onuse_lapackwithout noting thatuse_lapackdefaults to true (meson.options:14), so it is effectively always built.3. Nonexistent paths. Key Files pointed at
scs/include/scsmodule.handscs/include/scsobject.h; there is noscs/include/directory. Also added the previously undocumentedscs/py_ctrlc.c, which carries the process-shared SIGINT state that replacesscs_source/src/ctrlc.cin Python builds.Two smaller things while in there: the vague Solver Variants paragraph now describes what
_SOLVER_DISPATCHand_resolve_autoactually do (deferred imports, so an unbuilt backend raisesModuleNotFoundErrorat construction rather than breakingimport scs), and thescspy.cheader comment no longer refers to the front-facing Python code as"scs.py".Docs and one C comment only — no behaviour change. Suite still green: 397 passed, 67 skipped.
🤖 Generated with Claude Code