Files

6.1 KiB

R4 J1: results (to fill during the device acceptance)

Procedure: R4_J1.md. One column per device actually run; PASS, FAIL or NON TESTÉ per cell, with a note under the table for every FAIL, incident, timing and memory event. A device or browser not run stays NON TESTÉ.

Status 2026-10-18: the owner reports the functional checks they executed as passed, with no failure, on a computer (browser and version to be given). Status 2026-10-20 (R4.9): the recette is re-issued on the R4.8 build (the export on the core, no TypeScript engine left); J1.13 added; every cell stays as the owner left it, nothing below was run by Claude. iPhone / Safari iOS: not run, deferred by the owner to a later date; it stays NON TESTÉ and blocks the final acceptance of R4.3 (plan §A.4, OD-6: a device not run is never reported validated). The J1.8 / J1.12 timings and the J1.9 / J1.10 memory observations on the computer are pending too; every cell below stays NON TESTÉ until the owner confirms it, nothing is inferred. Status 2026-10-21 (R4.10): the owner ran the recette on the computer with Chrome, Firefox and Brave, and on Android with its browser; Safari iOS not run (it stays NON TESTÉ until the owner's results on the real iPhone; a simulator never fills its column and is never a memory proof). The result per check was declared without the split per browser: it sits in the last column. A device cell is filled only where the owner named the browser (J1.7). Brave is outside the OD-6 list, reported as run. Status 2026-10-21 (R4 close): the owner ran J1 on a real iPhone with Safari iOS and reports it as passed as a whole, with no blocking anomaly. The declaration covers the recette, not each check: the iPhone column carries it as PASS déclaré per row, and the details the grid asks for (iPhone model, iOS and Safari versions, surface, the J1.8 and J1.12 timings, the J1.9 and J1.10 memory observations, the J1.7 picker delay on the phone) were not given and are not inferred (note below).

Device, browser, version Surface (1 dev / 2 prod CSP) Date
(to be confirmed by the owner) 2026-10-18
iPhone, Safari iOS (model and versions not given) not given 2026-10-21
Check Computer, Chrome Computer, Firefox Computer, Brave (outside OD-6) iPhone, Safari iOS (run 2026-10-21, declared as a whole) Android, its browser Declared 2026-10-21, no browser split
J1.1 reference PDFs open the editor NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.2 hostile folder: message per file NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.3 rescue copies NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.4 engine unavailable (--csp without) NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.5 engine recovered NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.6 interrupted import NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.7 offline import NON TESTÉ PASS, picker delay (see note) PASS, picker delay (see note) PASS déclaré, no delay reported NON TESTÉ PASS, performance reserve
J1.8 L export (≈ 42 MB) opens, time to the editor NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré, no time given NON TESTÉ PASS déclaré
J1.9 six L imports in a row, tab alive NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré, no memory observation given NON TESTÉ PASS déclaré
J1.10 twenty picks of values-1m.pdf NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré, no memory observation given NON TESTÉ PASS déclaré
J1.11 rescue copy under and over the cap NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré
J1.12 timings (demo, L) NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré, no measure NON TESTÉ PASS déclaré, « quelques secondes », no measure
J1.13 bilan in progress kept after a refused import NON TESTÉ NON TESTÉ NON TESTÉ PASS déclaré NON TESTÉ PASS déclaré

Notes (FAIL, incidents, timings, memory warnings, reloads):

  • J1.7 (owner, 2026-10-21, Brave and Firefox, surface 2: production build, port 4173, production CSP). First declared FAIL: offline, « Importer » did not open the file picker and the browser seemed unusable. Rectified the same day after the mini-recette (R4_MINI_RECETTE.md, M1 and M2): with the system's network cut, the file picker takes about one minute to open, from Ctrl+O in an empty tab as from the Faunosteo button alike; once open, the offline import works. Functional PASS with a performance reserve on the picker's opening. The behaviour is seen outside Faunosteo, so no proof of an application defect remains; the cause of the delay is not established (not measured, not traced). Diagnosis: RUST.md §10.12; the recette server's cache headers fix stays (a demonstrated, separate defect).
  • J1.12: « quelques secondes » from the pick to the editor, no measure; the 20 s timer is neither confirmed nor moved by it (the §B.3 rule needs the timings).
  • iPhone, Safari iOS (owner, 2026-10-21): J1 reported as passed as a whole, no blocking anomaly. Missing for the per-check justification the grid asks for, not invented: the iPhone model, the iOS and Safari versions, the surface (dev or production CSP), the J1.8 and J1.12 timings, the J1.9 and J1.10 memory observations (tab alive, reloads, warnings), whether the J1.7 picker delay seen on the computer appears on the phone. The items of PDF_IMPORT_HARDENING.md §6 that close on these observations (5, 6, 9, 20) stay open (§7 there).

Recommendation (owner, 2026-10-21): accept with follow-ups. R4.3 is accepted on the computer (Chrome, Firefox, Brave), Android and the real iPhone; the follow-ups are the J1.7 picker-delay reserve (cause not established), the device timings and memory observations not reported, and M3 of R4_MINI_RECETTE.md not run (complementary, not blocking).