Skip to content

0.6.3: arraybridge 0.5.0 - #42

Merged
trissim merged 1 commit into
mainfrom
refactor/arraybridge-0.5
Oct 12, 2026
Merged

trissim merged 1 commit into
mainfrom
refactor/arraybridge-0.5

Conversation

@trissim

@trissim trissim commented Oct 12, 2026

Copy link
Copy Markdown
Contributor

PolyStore 0.6.2 cannot be imported with arraybridge 0.5.0 (polystore.constants imports removed names) and every non-NumPy write would raise (gpu_id=0 to a NumPy target).

  • array_payload.storage_numpy_array: ArrayLibrary.of(data).to_numpy(data). The library that made the array downloads it; no device is named. Parameter annotated object instead of Any.
  • constants.py: the re-exports of CPU_MEMORY_TYPES, GPU_MEMORY_TYPES, SUPPORTED_MEMORY_TYPES are deleted. Nothing in PolyStore, OpenHCS or pyqt-reactive read them, so there is no question to redirect to the library family.
  • polystore.streaming.handlers deleted (327 lines): it imports metaclass_registry.lazy, absent from the already-required metaclass-registry 0.4, so it was unimportable and unreferenced. Found by the new import test.
  • tests/test_every_module_imports.py: imports every module in a child interpreter; any first-party import error fails.
  • arraybridge>=0.5.0,<0.6; uv.lock regenerated (uv lock --check clean).

Local, fresh Python 3.12 venv from PyPI (arraybridge 0.5.0): 600 passed, 5 skipped; with the torch CPU wheel added: 601 passed, 4 skipped. Baseline on main with arraybridge 0.4.1: 598 passed, 5 skipped.

🤖 Generated with Claude Code

storage_numpy_array asks the library that made the array to download it
(ArrayLibrary.of(data).to_numpy(data)); it no longer names a device, and
the removed gpu_id keyword is gone.

polystore.constants no longer re-exports CPU_MEMORY_TYPES, GPU_MEMORY_TYPES
and SUPPORTED_MEMORY_TYPES, which arraybridge 0.5.0 removed; nothing read
them. polystore.MemoryType stays.

polystore.streaming.handlers is deleted: it imported metaclass_registry.lazy,
which metaclass-registry 0.4 (already required) does not have, so the package
could not be imported and nothing imported it.

A new test imports every module in a child interpreter and fails on any
first-party import error.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Copilot AI balanced review requested due to automatic review settings October 12, 2026 19:29

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

@trissim
trissim merged commit 5fdb617 into main Oct 12, 2026
11 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants