morluto/rea

Fork 0

mirror of https://github.com/morluto/rea.git synced 2026-11-05 07:47:29 +00:00

[PR #6710] [CLOSED] fix(model): read null safetensors metadata as absent #7395

New issue

Closed

opened 2026-11-01 21:47:33 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-11-01 21:47:33 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/6710
Author: @Sriram-PR
Created: 10/23/2026
Status: ❌ Closed

Base: main ← Head: fix/safetensors-null-metadata

📝 Commits (1)

58117be fix(model): read null safetensors metadata as absent

📊 Changes

3 files changed (+46 additions, -3 deletions)

View changed files

📝 docs/model-artifacts.md (+2 -1)
📝 src/model/SafetensorsHeader.ts (+13 -2)
📝 tests/boundary/model/modelArtifact.test.ts (+31 -0)

📄 Description

Summary

inspect_model_artifact rejects a safetensors file whose header has "__metadata__": null with file_format_rejected ("Safetensors header requires unsigned integer declarations"). The reference reader declares that field as Option<HashMap<String, String>> (HashMetadata in safetensors/src/tensor.rs), so null is valid there and the same bytes load fine with safetensors itself.

Fixes #6656

Problem and expected behavior

Smallest trigger: the issue's 82-byte file, one U8 tensor x with shape [1] and "__metadata__": null. On main a281455f, ModelArtifactService.inspect with detail: "complete" fails as above; the same file with __metadata__ left out or set to {"a":"b"} passes. The header JSON reader in SafetensorsHeader.ts has no case for null, so it falls into the integer branch. Expected: no metadata rows and the tensor reported as usual.

Change and scope

HeaderJson now parses the null literal.
A root __metadata__ of null is skipped like a missing one.
objectOf rejects null explicitly (typeof null is "object"), so null anywhere else still fails like it does in the reference reader: a tensor entry, a metadata value, a shape dimension, offsets or dtype.
docs/model-artifacts.md notes that null __metadata__ reads as absent.
Not included: other JSON literals (true, false, floats). The reference reader rejects those in every header position too.

Contribution scope: ordinary fix within existing boundaries; no scope decision needed.

Contract and boundary impact

Semantic owner and earliest changed stage: HeaderJson and readSafetensorsHeader in src/model/SafetensorsHeader.ts.
Contribution scope fit and linked maintainer decision: not applicable (bug fix, #6656).
CLI and MCP/tool-catalog contract: none; a previously rejected valid file is now inspected.
Provider, bridge, target-format, or platform compatibility: safetensors headers with null __metadata__ are now accepted.
Evidence, artifact, provenance, or reconstruction contract: none.
Target selection, authority, upstream behavior, and cleanup details: not applicable.
Generated metadata, package, or installation impact: none.

Expected effects (check known effects; use documentation-only on its own).
Unchecked items do not establish absence; leave uncertainties visible.

Read existing files or captured data.
Launch or attach to a process.
Inject instrumentation or run scripts/code.
Change process memory, registers, files, or configuration.
Contact network services.
Recover protected code or work around protections.
Documentation or usability only; no target interaction.
Not sure yet; investigation or maintainer guidance needed.

Evidence and regression coverage

Tests added or updated: tests/boundary/model/modelArtifact.test.ts adds reads null safetensors __metadata__ as absent, as the reference reader does (no metadata rows, tensor x U8 [1]) and three rows in the file_format_rejected table: null tensor entry, null metadata value, null shape dimension.

Base reproduction: the issue's file fails on main a281455f through ModelArtifactService.inspect. The new test fails on main's source with:

Error: Cannot open artifact: Safetensors header requires unsigned integer declarations
Caused by: ModelFormatFailure: Safetensors header requires unsigned integer declarations

User-visible CLI/MCP output: the issue's file now reports tensor x (U8, shape ["1"], byte length 1) and an empty metadata list.
Remaining proof gaps: I checked the reference behavior from the safetensors source (Option field) and didn't run the Python package here.

For evidence-bearing changes:

Observed, derived, and inferred claims remain distinguishable.
Artifact identity, source provenance, and failed attempts remain preserved.
Unsupported, incomplete, unavailable, or uncertain outcomes remain visible.

Validation performed

npm run test:focused -- tests/boundary/model/modelArtifact.test.ts: 87 passed (the new positive test fails on main's source)
npm run test:focused -- tests/acceptance/modelArtifactCli.test.ts tests/process-global/modelHostReadFailure.test.ts: 12 passed
npm run check:fast: typecheck and lint passed

Compatibility, safety, and release

Breaking changes or migration steps: none.
Real Hopper/Ghidra, browser, or OS coverage: Linux x64, Node 24.18.0.
Package or release metadata impact: none.
Security, privacy, process, or containment review: none; the parser still rejects null everywhere the reference does.

Author readiness

Human owner: @Sriram-PR

I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work.
I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps.
I declare this PR ready for initial maintainer review.

Before merge

This PR addresses a concrete problem or an agreed enhancement.
Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff.
The PR has one focused outcome and the title follows type(scope): outcome.
Related issue is linked, or the reason for not linking one is stated above.
Tests cover changed observable behavior and meaningful failure paths.
Owning docs, contracts, and generated metadata are updated where needed.
User-visible CLI/MCP changes include representative output.
I checked the final diff for secrets, unrelated cleanup, and unsupported claims.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/morluto/rea/pull/6710 **Author:** [@Sriram-PR](https://github.com/Sriram-PR) **Created:** 10/23/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `fix/safetensors-null-metadata` --- ### 📝 Commits (1) - [`58117be`](https://github.com/morluto/rea/commit/58117be4f99332a009192d527adbf4fb6e39bdef) fix(model): read null safetensors __metadata__ as absent ### 📊 Changes **3 files changed** (+46 additions, -3 deletions) <details> <summary>View changed files</summary> 📝 `docs/model-artifacts.md` (+2 -1) 📝 `src/model/SafetensorsHeader.ts` (+13 -2) 📝 `tests/boundary/model/modelArtifact.test.ts` (+31 -0) </details> ### 📄 Description ## Summary `inspect_model_artifact` rejects a safetensors file whose header has `"__metadata__": null` with `file_format_rejected` ("Safetensors header requires unsigned integer declarations"). The reference reader declares that field as `Option<HashMap<String, String>>` (`HashMetadata` in `safetensors/src/tensor.rs`), so null is valid there and the same bytes load fine with safetensors itself. Fixes #6656 ## Problem and expected behavior Smallest trigger: the issue's 82-byte file, one U8 tensor `x` with shape `[1]` and `"__metadata__": null`. On main a281455f, `ModelArtifactService.inspect` with `detail: "complete"` fails as above; the same file with `__metadata__` left out or set to `{"a":"b"}` passes. The header JSON reader in `SafetensorsHeader.ts` has no case for `null`, so it falls into the integer branch. Expected: no metadata rows and the tensor reported as usual. ## Change and scope - `HeaderJson` now parses the `null` literal. - A root `__metadata__` of null is skipped like a missing one. - `objectOf` rejects null explicitly (`typeof null` is `"object"`), so null anywhere else still fails like it does in the reference reader: a tensor entry, a metadata value, a shape dimension, offsets or dtype. - `docs/model-artifacts.md` notes that null `__metadata__` reads as absent. - Not included: other JSON literals (`true`, `false`, floats). The reference reader rejects those in every header position too. Contribution scope: ordinary fix within existing boundaries; no scope decision needed. ## Contract and boundary impact - Semantic owner and earliest changed stage: `HeaderJson` and `readSafetensorsHeader` in `src/model/SafetensorsHeader.ts`. - Contribution scope fit and linked maintainer decision: not applicable (bug fix, #6656). - CLI and MCP/tool-catalog contract: none; a previously rejected valid file is now inspected. - Provider, bridge, target-format, or platform compatibility: safetensors headers with null `__metadata__` are now accepted. - Evidence, artifact, provenance, or reconstruction contract: none. - Target selection, authority, upstream behavior, and cleanup details: not applicable. - Generated metadata, package, or installation impact: none. Expected effects (check known effects; use documentation-only on its own). Unchecked items do not establish absence; leave uncertainties visible. - [x] Read existing files or captured data. - [ ] Launch or attach to a process. - [ ] Inject instrumentation or run scripts/code. - [ ] Change process memory, registers, files, or configuration. - [ ] Contact network services. - [ ] Recover protected code or work around protections. - [ ] Documentation or usability only; no target interaction. - [ ] Not sure yet; investigation or maintainer guidance needed. ## Evidence and regression coverage - Tests added or updated: `tests/boundary/model/modelArtifact.test.ts` adds `reads null safetensors __metadata__ as absent, as the reference reader does` (no metadata rows, tensor `x` U8 `[1]`) and three rows in the `file_format_rejected` table: null tensor entry, null metadata value, null shape dimension. - Base reproduction: the issue's file fails on main a281455f through `ModelArtifactService.inspect`. The new test fails on main's source with: ``` Error: Cannot open artifact: Safetensors header requires unsigned integer declarations Caused by: ModelFormatFailure: Safetensors header requires unsigned integer declarations ``` - User-visible CLI/MCP output: the issue's file now reports tensor `x` (`U8`, shape `["1"]`, byte length 1) and an empty metadata list. - Remaining proof gaps: I checked the reference behavior from the safetensors source (`Option` field) and didn't run the Python package here. For evidence-bearing changes: - [x] Observed, derived, and inferred claims remain distinguishable. - [x] Artifact identity, source provenance, and failed attempts remain preserved. - [x] Unsupported, incomplete, unavailable, or uncertain outcomes remain visible. ## Validation performed - `npm run test:focused -- tests/boundary/model/modelArtifact.test.ts`: 87 passed (the new positive test fails on main's source) - `npm run test:focused -- tests/acceptance/modelArtifactCli.test.ts tests/process-global/modelHostReadFailure.test.ts`: 12 passed - `npm run check:fast`: typecheck and lint passed ## Compatibility, safety, and release - Breaking changes or migration steps: none. - Real Hopper/Ghidra, browser, or OS coverage: Linux x64, Node 24.18.0. - Package or release metadata impact: none. - Security, privacy, process, or containment review: none; the parser still rejects null everywhere the reference does. ## Author readiness Human owner: @Sriram-PR - [x] I reviewed the final diff and relevant existing code, understand the change and its edge cases, and take responsibility for the complete PR, including AI-assisted work. - [x] I checked the evidence above, confirmed it supports the current change, and recorded meaningful validation gaps. - [x] I declare this PR ready for initial maintainer review. ## Before merge - [x] This PR addresses a concrete problem or an agreed enhancement. - [x] Material scope and authority questions are resolved, and any required maintainer scope decision covers the final diff. - [x] The PR has one focused outcome and the title follows `type(scope): outcome`. - [x] Related issue is linked, or the reason for not linking one is stated above. - [x] Tests cover changed observable behavior and meaningful failure paths. - [x] Owning docs, contracts, and generated metadata are updated where needed. - [x] User-visible CLI/MCP changes include representative output. - [x] I checked the final diff for secrets, unrelated cleanup, and unsupported claims. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-11-01 21:47:33 +00:00

closed this issue
added the
pull-request
label

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

morluto/rea#7395

No description provided.

Rows
Columns

[PR #6710] [CLOSED] fix(model): read null safetensors __metadata__ as absent #7395