You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 900995e
Browse filesBrowse the repository at this point in the historyBrowse files
The CHANGELOG's 1.5.0 entry gains its HPO bullet: pineforge-hpo 0.13.2
pins pineforge-codegen <1.4, and HPO support for codegen 1.4 and 1.5
(enum options as Enum.member tags) comes in pineforge-hpo 0.14.0.
Five files that still called 1.5.0, or a 1.5.0 feature, unreleased or
targeted for it now read as released once the tag lands:
docs/pages/abi-stability.md (the RunFailureCode exception, the
symbol-inventory counts line, the recorded-outputs line),
docs/pages/index.md (its counts line, matching the README wording),
the run-failure page (the outputs and selected-window rows' status, the
two 1.5.0 outputs reasons, and the 24 catalog rows that said none was
released), docker/README.md (the recorded-outputs section) and
docs/outputs.md. Only the status words change; "not part of 1.4.0"
stays where it is true history, and no number, claim or other sentence
moves.
No file outside these six changes.
Co-Authored-By: Claude Code <[email protected]>
Copy file name to clipboardExpand all lines: CHANGELOG.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -73,6 +73,7 @@ does not update a statically linked strategy library.
73
73
`PINEFORGE_PRIVATE_FIXTURES` — the verification requires every pinned row to
74
74
run: a row that skips there fails the run. Without the overlay, those rows
75
75
skip in public runs and the row floors account for them.
76
+
-**HPO.** pineforge-hpo 0.13.2 pins pineforge-codegen <1.4; HPO support for codegen 1.4 and 1.5 (enum options as Enum.member tags) comes in pineforge-hpo 0.14.0.
Copy file name to clipboardExpand all lines: docs/pages/run-failure-codes.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -93,7 +93,7 @@ codes it tested the text alone, so a script stopped by an empty
93
93
94
94
The tables below list the catalog at this tree.
95
95
96
-
The `outputs_limit` and `outputs_rejected` rows below are unreleased, targeted for 1.5.0, and not part of 1.4.0; so are the 24 selected-window and report-contract rows described [after the tables](#run_failure_codes_selected_window). The other 38 codes retain their released 1.4.0 status. The `strategy_library_incompatible` row counts the two unreleased 1.5.0 outputs reasons, `outputs_api_mismatch` and `outputs_manifest_invalid`, in addition to its six released 1.4.0 reasons.
96
+
The `outputs_limit` and `outputs_rejected` rows below are released in 1.5.0, and not part of 1.4.0; so are the 24 selected-window and report-contract rows described [after the tables](#run_failure_codes_selected_window). The other 38 codes retain their released 1.4.0 status. The `strategy_library_incompatible` row counts the two released 1.5.0 outputs reasons, `outputs_api_mismatch` and `outputs_manifest_invalid`, in addition to its six released 1.4.0 reasons.
97
97
98
98
<!-- BEGIN generated from docker/run_failure_codes.json: python3 scripts/test_gen_run_failure_catalog_diff.py --write-docs-tables -->
99
99
@@ -224,7 +224,7 @@ three: `report_post_execution_failed` (the phases after the strategy has
224
224
finished), `engine_command_value_invalid` (an invalid command value the engine
225
225
derived itself) and `window_pruner_unsupported` (the optimizer's trial pruner).
226
226
Their classes, argument kinds and English are the contract's; all are
227
-
`retryable: false` and `since: 1.5.0`, and none is released. The registry
227
+
`retryable: false` and `since: 1.5.0`, and all are released. The registry
228
228
declares them and validates their arguments, which is all this page states
229
229
about them: whether a build raises one is the build's own, and a row in the
230
230
catalog is not proof that the source paths it names enforce anything.
0 commit comments