sorokin.oleg.edu
2df7f580aa
Replace the oversized header and reference rail with stable editor slots and a retained supporting drawer. Bind diagnostic, result and menu actions to their original publication and visit; preserve effective selections on navigation and dismiss Actions, Find and the drawer one surface per Escape. Retain exact Windows source captures, eight final gates (871 desktop passes), independent reviews, native selection/rename/recovery trials and the original failures. All 526 tested inputs and staged evidence were byte-checked before commit. Combined preview integration and packaging remain with issue 27; Linux remains paused.
5.3 KiB
5.3 KiB
Initial Text native evidence: independent review
PASS — no factual mismatch found in the bounded report/archive review. This is acceptance of the evidence's stated scope, not acceptance of the failed native editing workflow or the current correction. No builds, original helpers, GUI actions or production edits were performed.
- Manifest SHA matches the supplied
991d54095f5466b717aefde99682e0c7d3136091bdf5525d8f21c0629d3cb0a4. All 106 payload paths, byte sizes and SHA-256 values match, with no unlisted files. All 106 copied payloads also match their original private trial/gate/helper files at review time. - Both source ZIPs pass CRC and exact manifest membership/byte hashes: attempt 3 has 523 inputs; attempt 4 has 524. Attempt 4's base CBC identity, source ZIP, Cargo development artifact, GUI size/hash and six successful unchanged-input gate receipts match the report. Raw stdout confirms 859 membership entries, 20 Guard tests, 26 focused passes, 856 full-suite passes/0 failures/3 ignored, plus the separately invoked ignored timing test. The five-sample Confirm median is 60.0406ms and explicitly excludes returned Task polling/rendering. GUI build ended 02:37:30.596393Z.
- Attempt 3's raw panic and totals match the retained failed-fixture account (853/1/3). The report correctly avoids treating it as a production defect fixed by a workaround.
- There are 28 screenshot metadata states and 30 JPEG payloads (the native dialogs supply multiple surfaces). The report says 28 states, not 28 JPEGs. Metadata 009 gives 1102x732 outer dimensions. Selected visual readbacks 009/013/016 support the stated 486/306/246 editor-slot heights and 180/240 drawer budgets. At 013 the expanded group shows three complete actionable rows and a partly clipped fourth; this is not a count of group headings.
- Visual readbacks 021 and 023 show the same scrolled source, with the initially selected alias no longer visibly highlighted after Find Close. Capture 024 shows the malformed insertion and Stale/disabled retained results. Capture 027 shows the correct direct-control replacement; 028 restores the alias and saved status while retaining Stale. These support the reported failure/control distinction. The source-mechanism diagnosis and predicted line-selection risk are correctly distinguished from what was exercised natively; this review did not independently rerun that mechanism.
- Launch/closure identify the same owned PID 2272, GUI, private profile/temp/cwd and limited PATH. They record 02:38:06.847206Z to 02:55:30.057373Z, exit 0, empty raw logs and unchanged executable/DLL. The read-only oracle receipt and stdout establish exact saved model/identity/30 reachable sources; its 37-file physical map also independently matches the complete original work-directory membership/bytes at review time. The report correctly limits this oracle to disk state.
The historical source/binary distinction is explicit: these are uncommitted captured attempt-4 inputs, not a source-commit build claim or a published release. Find correction and a fresh native trial remain required, and deferred workflows are not claimed complete. No native 840px alias-summary overflow is claimed; focused raw logs correctly show old flags fit 840px and clip at 720px.
Exact reviewed bindings
C:/ar-text-9e8r6t/harness/text-workspace-native-initial-20260913.md— 6269 bytes; SHA-256e2deccb8d2d51858250241671ab19bcf4f424d6a8e98bf666068214877231d77C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/manifest.json— 19208 bytes; SHA-256991d54095f5466b717aefde99682e0c7d3136091bdf5525d8f21c0629d3cb0a4C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/attempt-4/inputs.json— 97811 bytes; SHA-25683ec5a331d807767ee34441ad49e30cc5a5990c79432f664ce044eb869016875C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/attempt-4/inputs.zip— 2187961 bytes; SHA-256e1e566295bbbd6c1d571f7c8908b7344397293496481790f0f8dacab7eba5366C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/attempt-4/complete.json— 256 bytes; SHA-256268f46d4b072c039bafaed211eb311a36f55fdae3f26976b62e2827262417690C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/attempt-4/gui-result.json— 1904 bytes; SHA-2569f98aa0b0a363b38aad811f3d5068e1f74e7890a30ebaf4c45d7d2c50c8a8ef8C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/attempt-4/timing-stdout.log— 7226 bytes; SHA-2565ae357431fa8d99f76e193aa08f1192fb7ed99d8c3667966ce97790eca8d98feC:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/trial-text-1/launch.json— 1663 bytes; SHA-25691e989fd914f9a6de564752e62e67778bf9827039bf0895579b2cb753b4c4289C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/trial-text-1/closed.json— 398 bytes; SHA-256a1df27edfe0349bbb596e09fc2907e330eb3f1165646cfe4c5d1b2266b502df1C:/ar-text-9e8r6t/harness/experiments/text-workspace-initial-20260913/trial-text-1/verification-representative-selection-probes-closed/result.json— 7378 bytes; SHA-256257c67421d0efd459ec13aa9ee2ce017e904a1486f8db338ea9ac29cbc351907
All individual payload bindings, including the inspected captures, are transitively fixed by the verified manifest above. Later working-source changes are outside this historical review.