fix: correct errors found while testing the interpretations by ariostas · Pull Request #1824 · scikit-hep/uproot

ariostas · 2026-11-06T14:42:41Z

🤖 AI text below 🤖

Part of the split of #1814 (the list is in #1834). These bugs were found while writing the coverage tests of #1814, and tests/test_1824_fix_interpretation.py holds their regression tests: every test in it fails on main (for parametrized tests, at least one of their cases).

Containers
- Objectwise std::maps read their string keys and values without a header, and an empty map has the same dtypes as the others.
- A C array (AsArray) whose header has no byte count raises a DeserializationError instead of reading a wrong length. AwkwardForth finds the end of an array from its byte count rather than the end of the entry, and arrays of a NumPy dtype fall back from AwkwardForth.
- AsArray and AsDynamic get __eq__ and __hash__, and AsDynamic.typename works for any model.
- AsFIXME errors carry the chunk, cursor and file.
- Values that are str subclasses print as strings.
Object and strided arrays
- ObjectArray supports negative indexes, slices (step 1) and iteration from an offset.
- StridedObjectArray items are recognized as numpy.void.
- Strided objects with headers come out in native byte order, with their members in offset order.
- When an object branch falls back from AwkwardForth, its machine is reset.
- Pandas output of object branches gets the right interpretation.
Strings
- The AwkwardForth reader of AsStrings reads big-endian lengths and accumulates offsets.
- Baskets that mix StringArrays and Awkward arrays are concatenated.
Libraries
- Awkward jagged and string arrays with 32-bit offsets no longer use ListOffsetArray32/ListOffsetArrayU32, which Awkward 2 doesn't have.
- how="zip" with no arrays works.
- The Pandas how docs and error say what a string how actually does.
Identification
- Typename parsing doesn't index past the end of the tokens.
- A Float16/Double32 title whose brackets don't hold a valid range is treated like one without brackets, instead of raising UnboundLocalError.
- Leaf-lists with a count leaf are rejected before building their dtype.
- awkward_form raises a TypeError for a dtype it can't represent.
- awkward_form_of_iter keeps the fields of records and tuples, and recursively_fix_awkward_form_of_iter (which coffea calls) no longer calls awkward_can_optimize, which doesn't exist.
- _content_cls_from_name strips suffixes from the right end.
Misc
- JaggedArray.parents_localindex used uproot._util.win, which doesn't exist, so it always raised AttributeError.
- TLeafG is registered.
- Version 3 TStreamerElements are read with the range (fXmin, fXmax, fFactor) that only that version writes; reading one used to fail.
- The str of an AwkwardForth node indents all of its children alike.
- TRef keeps its reference number in its members, like other models.
- TBasket.array passes final_array a dict, as every interpretation expects.
- Unpickling a dynamic model doesn't skip attributes that a base class has.
- reset_classes also reloads TClonesArray.
- The TruncatedNumerical and language.python error messages no longer fail while being formatted.
- CustomInterpretation and AsBinary compare by identity.

The reset_classes fixture in tests/conftest.py now puts the original model classes back afterward. Some tests here need that, because they run in the same process (with pytest-xdist) as a test that reloads the classes. #1823 makes the same change.

codecov · 2026-11-06T14:57:27Z

Codecov Report

❌ Patch coverage is 98.70130% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 80.06%. Comparing base (39ff721) to head (f179574).
✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
src/uproot/interpretation/objects.py	90.90%	1 Missing ⚠️

Additional details and impacted files

Files with missing lines	Coverage Δ
src/uproot/_awkwardforth.py	`93.67% <100.00%> (+2.29%)`	⬆️
src/uproot/_util.py	`78.09% <100.00%> (+10.36%)`	⬆️
src/uproot/containers.py	`83.61% <100.00%> (+15.38%)`	⬆️
src/uproot/interpretation/custom.py	`88.75% <100.00%> (+3.22%)`	⬆️
src/uproot/interpretation/identify.py	`90.70% <100.00%> (+15.78%)`	⬆️
src/uproot/interpretation/jagged.py	`96.17% <ø> (+5.90%)`	⬆️
src/uproot/interpretation/library.py	`74.13% <100.00%> (+4.31%)`	⬆️
src/uproot/interpretation/numerical.py	`85.49% <ø> (+1.96%)`	⬆️
src/uproot/interpretation/strings.py	`90.65% <100.00%> (+17.16%)`	⬆️
src/uproot/language/python.py	`73.21% <ø> (+0.89%)`	⬆️
... and 6 more

... and 3 files with indirect coverage changes

The check that a sequence isn't equal to a string passed for anything but True. The test of Node.__str__ expected its wrong indentation, which scikit-hep#1824 fixes; it is now there, with the right one. The test of Library equality in test_0016 is deleted here, as test_names_and_equality covers it. Assisted-by: Claude Code:claude-opus-5-5

The edges of the variable-width axis spanned exactly fXmin to fXmax, so the test couldn't tell the mean of the widths from (fXmax - fXmin) / fNbins. The files that Rewritable opens are now closed after each test. The test of a version 3 TStreamerElement leaves for scikit-hep#1824, which reads its range. Assisted-by: Claude Code:claude-opus-5-5

These were found by the tests of scikit-hep#1814, which are the regression tests here. The fixes are to containers (std::map, C arrays, AsDynamic and AsArray equality and hashing), strided and object arrays, the Awkward and Pandas libraries, string baskets, typename parsing, Float16/Double32 titles, TLeafG, TRef, TBasket.array, and model pickling. The reset_classes fixture now puts the original model classes back after the test, so that tests running after it in the same process (with pytest-xdist) compare against the classes they looked up. Assisted-by: Claude Code:claude-opus-5-5

Version 3 of TStreamerElement writes fXmin, fXmax and fFactor after its type name (later versions take the range from the title instead), and reading one stopped before them, so a file with version 3 elements failed to deserialize. The format for them was defined but unused. Assisted-by: Claude Code:claude-opus-5-5

Children after the first were indented two spaces less than the first. Assisted-by: Claude Code:claude-opus-5-5

Assisted-by: Claude Code:claude-opus-5-5

ariostas · 2026-11-10T19:25:28Z

pre-commit.ci run

ianna

@ariostas - Thanks! LGTM.

* test: cover the interpretations New tests from #1814, apart from the regression tests of its fixes. Assisted-by: Claude Code:claude-opus-5-5 * test: tighten an equality check and drop the test of Node.__str__ The check that a sequence isn't equal to a string passed for anything but True. The test of Node.__str__ expected its wrong indentation, which #1824 fixes; it is now there, with the right one. The test of Library equality in test_0016 is deleted here, as test_names_and_equality covers it. Assisted-by: Claude Code:claude-opus-5-5

The edges of the variable-width axis spanned exactly fXmin to fXmax, so the test couldn't tell the mean of the widths from (fXmax - fXmin) / fNbins. The files that Rewritable opens are now closed after each test. The test of a version 3 TStreamerElement leaves for scikit-hep#1824, which reads its range. Assisted-by: Claude Code:claude-opus-5-5

The check that a sequence isn't equal to a string passed for anything but True. The test of Node.__str__ expected its wrong indentation, which scikit-hep#1824 fixes; it is now there, with the right one. The test of Library equality in test_0016 is deleted here, as test_names_and_equality covers it. Assisted-by: Claude Code:claude-opus-5-5

* test: cover the interpretations New tests from #1814, apart from the regression tests of its fixes. Assisted-by: Claude Code:claude-opus-5-5 * test: tighten an equality check and drop the test of Node.__str__ The check that a sequence isn't equal to a string passed for anything but True. The test of Node.__str__ expected its wrong indentation, which #1824 fixes; it is now there, with the right one. The test of Library equality in test_0016 is deleted here, as test_names_and_equality covers it. Assisted-by: Claude Code:claude-opus-5-5

The edges of the variable-width axis spanned exactly fXmin to fXmax, so the test couldn't tell the mean of the widths from (fXmax - fXmin) / fNbins. The files that Rewritable opens are now closed after each test. The test of a version 3 TStreamerElement leaves for scikit-hep#1824, which reads its range. Assisted-by: Claude Code:claude-opus-5-5

github-actions Bot added the type/fix PR title type: fix (set automatically) label Nov 6, 2026

This was referenced Nov 6, 2026

test: tighten the existing tests and make their fixtures robust #1823

Merged

test: enforce 100% line and branch coverage on Codecov #1834

Draft

test: reach 100% testing coverage and enforce it on Codecov #1814

Closed

ariostas marked this pull request as ready for review November 6, 2026 14:47

ariostas force-pushed the fix/interpretation branch from aee6d7c to 92957c6 Compare November 6, 2026 15:09

ariostas added 4 commits November 10, 2026 11:24

fix: indent every child of an AwkwardForth node in its str

ef5ca5d

Children after the first were indented two spaces less than the first. Assisted-by: Claude Code:claude-opus-5-5

test: close the files of the strided-object tests with headers

f179574

Assisted-by: Claude Code:claude-opus-5-5

ariostas force-pushed the fix/interpretation branch from 5601fc1 to f179574 Compare November 10, 2026 16:25

ianna approved these changes Nov 10, 2026

View reviewed changes

ariostas merged commit f74da82 into scikit-hep:main Nov 10, 2026
21 checks passed

Navigation Menu

Repository navigation

fix: correct errors found while testing the interpretations - #1824

fix: correct errors found while testing the interpretations#1824
ariostas merged 4 commits into
scikit-hep:mainfrom
ariostas:fix/interpretation

ariostas commented Nov 6, 2026 •

edited

Loading

Uh oh!

codecov Bot commented Nov 6, 2026 •

edited

Loading

Uh oh!

ariostas commented Nov 10, 2026

Uh oh!

ianna left a comment

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

ariostas commented Nov 6, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov Bot commented Nov 6, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

ariostas commented Nov 10, 2026

Uh oh!

ianna left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

ariostas commented Nov 6, 2026 •

edited

Loading

codecov Bot commented Nov 6, 2026 •

edited

Loading