Adds the second unit's measurement
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
@@ -5,7 +5,7 @@
|
||||
- **Feature Slug**: prerendering
|
||||
- **Project Type**: Brownfield
|
||||
- **Start Date**: 2026-10-31T14:15:01Z
|
||||
- **Current Stage**: CONSTRUCTION - Code Generation (unit 2: display-fallbacks): B0 to B6 done; B7 measurement running
|
||||
- **Current Stage**: CONSTRUCTION - Code Generation (unit 2: display-fallbacks): measured; logo fix and summary running
|
||||
- **Branch**: feature/prerendering (from master c4c2f83); commits and pushes allowed for this feature (owner, 2026-10-31)
|
||||
- **Worktree**: K:\Development\SlpSoftware\Projects\AndreaSlopFotografie-prerendering
|
||||
|
||||
|
||||
@@ -323,3 +323,15 @@
|
||||
**Context**: Code Generation Part 2, unit 2.
|
||||
|
||||
---
|
||||
|
||||
## Code Generation - Unit 2 Measurement
|
||||
|
||||
**Timestamp**: 2026-11-01T02:40:00Z
|
||||
|
||||
**User Input**: "ga verder waar je gebleven was"
|
||||
|
||||
**AI Response**: "Resumed the measurement sub-agent after a usage limit. Results in display-fallbacks/code/measurement-display.md. BR-DMS-05 not triggered: E 0.93, D 0.91, E2 0.93 on mobile. Finding V-D7: the lazy light logo gives an empty logo for about a frame on client navigation to / on desktop. Decided by the main session, as it follows from decision O4: load that logo after the page load event. V-D10 blocked: the Lighthouse CLI is not installed."
|
||||
|
||||
**Context**: Code Generation unit 2.
|
||||
|
||||
---
|
||||
|
||||
+112
@@ -0,0 +1,112 @@
|
||||
# Measurement - Unit 2 (display fallbacks)
|
||||
|
||||
Requirements: NFR-PR-P-01, P-03; rules BR-DMS-01 to 05. Neutral reading only; the owner decides on targets.
|
||||
|
||||
**Method**: `lhci collect`, 3 runs per URL, median, same PC as the unit-1 gate measurement (Node 24.16.0), `config/previewServer.ts` on port 4173, configs not edited, not-found URL `/404.html`. Commit under test: d11fcb1 (clean tree). Measured 2026-11-01 (01:19 to 01:44). Script: `summarize2.mjs` in the scratch folder (medians; LCP element and images from the median-LCP run). Master and unit 1 "A" figures are copied from `start-migration/code/measurement-gate.md`. CLS is shown to 3 decimals.
|
||||
|
||||
## V-D5 decision rule (BR-DMS-05): not triggered
|
||||
|
||||
Mobile `/portfolio`: E 0.93, D 0.91, E2 0.93. E is not below 0.9, not below D, and E and E2 differ by 0.00 (limit 0.02). Desktop is 0.99 for E, D and E2. Nothing goes to the owner on this rule.
|
||||
|
||||
| /portfolio | Score (runs) | FCP ms | LCP ms | TBT ms | CLS | LCP element | Image KB before real LCP |
|
||||
|---|---|---|---|---|---|---|---|
|
||||
| Mobile E | 0.93 (0.93, 0.93, 0.92) | 1358 | 3232 | 63 | 0.001 | tile 1 `img` (f41a1916) | 78 |
|
||||
| Mobile D | 0.91 (0.91, 0.89, 0.91) | 1357 | 3459 | 61 | 0.002 | tile 1 `img`, lazy | 222 |
|
||||
| Mobile E2 | 0.93 (0.93, 0.92, 0.93) | 1359 | 3233 | 62 | 0.001 | tile 1 `img` (f41a1916) | 78 |
|
||||
| Desktop E | 0.99 | 367 | 931 | 0 | 0.004 | tile 3 `img` in `picture` | 332 |
|
||||
| Desktop D | 0.99 | 369 | 870 | 0 | 0.004 | lazy tile `img` (Aanstaande vader...) | 551 |
|
||||
| Desktop E2 | 0.99 | 368 | 931 | 0 | 0.003 | tile 3 `img` in `picture` | 332 |
|
||||
|
||||
Earlier figures: 2026-10-27 eager 0.80, deferred 0.91 (mobile); unit-1 run "B" 0.90 (grid visible, deferred), A 0.94. D in this run (0.91) matches the deferred figures; E (0.93) is 0.02 higher than D on mobile with 227 ms better LCP; on desktop E LCP is 61 ms later than D, scores equal.
|
||||
|
||||
## Mobile, all URLs
|
||||
|
||||
| URL | Score master / A / U2 | LCP master / A / U2 | FCP U2 | TBT U2 | CLS U2 | LCP element (U2) |
|
||||
|---|---|---|---|---|---|---|
|
||||
| `/` | 0.75 / 0.85 / 0.85 | 4293 / 4280 / 4356 | 1359 | 71 | 0.001 | heading `span` |
|
||||
| `/over-mij` | 0.86 / 0.90 / 0.92 | 3535 / 3608 / 3233 | 1359 | 65 | 0.002 | `img` portrait |
|
||||
| `/portfolio` | 0.90 / 0.94 / 0.93 | 2926 / 3007 / 3232 | 1357 | 59 | 0.001 | `img` tile 1 |
|
||||
| `/soorten-fotoshoots` | 0.85 / 0.89 / 0.92 | 3750 / 3683 / 3306 | 1357 | 62 | 0.009 | `img` newborn |
|
||||
| `.../newborn` | 0.85 / 0.90 / 0.91 | 3525 / 3610 / 3456 | 1357 | 62 | 0.021 | `img` first story photo |
|
||||
| `.../zwangerschap` | 0.84 / 0.88 / 0.90 | 3676 / 3834 / 3607 | 1358 | 68 | 0.006 | `img` first story photo |
|
||||
| `.../gezin` | 0.84 / 0.88 / 0.89 | 3678 / 3908 / 3681 | 1357 | 64 | 0.002 | `img` first story photo |
|
||||
| `.../loveshoot` | 0.82 / 0.86 / 0.88 | 3768 / 4133 / 3831 | 1357 | 63 | 0.002 | `img` first story photo |
|
||||
| `.../portret` | 0.84 / 0.89 / 0.90 | 3673 / 3759 / 3607 | 1358 | 65 | 0.002 | `img` first story photo |
|
||||
| `/contact` | 0.87 / 0.94 / 0.94 | 3345 / 3007 / 3007 | 1357 | 66 | 0.001 | paragraph `p` |
|
||||
| `/privacyverklaring` | 0.90 / 0.96 / 0.96 | 2844 / 2630 / 2780 | 1357 | 60 | 0.001 | logo `img` |
|
||||
| Not found | 0.91 / 0.97 / 0.97 | 2603 / 2556 / 2629 | 1356 | 63 | 0.001 | logo `img` |
|
||||
|
||||
Times in ms. Scores across the 3 runs per URL are identical or within 0.01.
|
||||
|
||||
## Desktop, all URLs
|
||||
|
||||
| URL | Score master / A / U2 | LCP master / A / U2 | FCP U2 |
|
||||
|---|---|---|---|
|
||||
| `/` | 0.99 / 0.99 / 0.99 | 915 / 866 / 866 | 370 |
|
||||
| `/over-mij` | 0.99 / 1.00 / 1.00 | 770 / 752 / 708 | 369 |
|
||||
| `/portfolio` | 0.99 / 0.99 / 0.99 | 868 / 887 / 931 | 368 |
|
||||
| `/soorten-fotoshoots` | 0.99 / 1.00 / 1.00 | 770 / 769 / 706 | 368 |
|
||||
| `.../newborn` | 0.99 / 1.00 / 1.00 | 748 / 809 / 747 | 368 |
|
||||
| `.../zwangerschap` | 0.99 / 1.00 / 1.00 | 769 / 809 / 767 | 368 |
|
||||
| `.../gezin` | 0.99 / 1.00 / 1.00 | 768 / 808 / 767 | 368 |
|
||||
| `.../loveshoot` | 0.99 / 0.99 / 1.00 | 767 / 850 / 786 | 368 |
|
||||
| `.../portret` | 0.99 / 1.00 / 1.00 | 767 / 808 / 766 | 367 |
|
||||
| `/contact` | 0.99 / 1.00 / 1.00 | 783 / 728 / 708 | 367 |
|
||||
| `/privacyverklaring` | 1.00 / 1.00 / 1.00 | 641 / 586 / 625 | 367 |
|
||||
| Not found | 1.00 / 1.00 / 1.00 | 577 / 589 / 626 | 367 |
|
||||
|
||||
Desktop is 0.99 or better on all 12 URLs. Desktop TBT is 0 everywhere; CLS at most 0.014 (`/soorten-fotoshoots`).
|
||||
|
||||
## JavaScript (gzip)
|
||||
|
||||
| Measure | Master | A | U2 |
|
||||
|---|---|---|---|
|
||||
| Entry JavaScript on `/` | 181,161 | 206,550 | 208,100 |
|
||||
| All JavaScript | 284,705 | 305,234 | 306,472 |
|
||||
|
||||
Script transfer on mobile (median, bytes): `/` 222,660 (A 221,110), `/over-mij` 211,506 (A 209,938), `/contact` 225,724 (A 224,665).
|
||||
|
||||
## BR-DMS-03: expected against measured (A against U2, mobile)
|
||||
|
||||
- Better score and LCP as expected: `/over-mij` (+0.02, -375 ms), `/soorten-fotoshoots` (+0.03, -377), newborn (+0.01, -154), zwangerschap (+0.02, -227), gezin (+0.01, -227), loveshoot (+0.02, -302), portret (+0.01, -152). The expected 77 to 226 ms was exceeded on 4 of the 7 pages.
|
||||
- `/contact`: equal to A (0.94, 3007 ms); better than master.
|
||||
- `/portfolio`: 0.93 against A 0.94, LCP 3232 against 3007 (+225 ms), because a photo is now the largest paint. Better than B (0.90, 3607) and than D (0.91, 3459).
|
||||
- `/`: 0.85 and 4356 ms, equal score and +76 ms against A (expected "unchanged").
|
||||
- `/privacyverklaring` (+150 ms) and not found (+73 ms) LCP are later than A with equal scores; the LCP element is the dark logo, now lazy. Not an expected change; reported.
|
||||
|
||||
## BR-DMS-04: still open
|
||||
|
||||
- `/` at 0.85.
|
||||
- Below 0.9 on mobile: gezin 0.89 and loveshoot 0.88. Zwangerschap and portret are at 0.90.
|
||||
- LCP against master: loveshoot +63 ms, gezin +3 ms; the other shoot pages are below master.
|
||||
|
||||
## V-D14: largest paint per page and width (unthrottled Chrome probe, last LCP entry after 2.5 s)
|
||||
|
||||
- `/portfolio`: 412 px tile 1 (`f41a1916`), 768 px tile 2 (`f41a1922`), 1350 px tile 3 (`f41a7883`). All first-row tiles, the same photos as the earlier probe without the eager row.
|
||||
- `/`: heading `span` at 412, 768 and 1350 (the hero photo is not the candidate).
|
||||
- `/over-mij`: portrait (`about-intro`) at all widths. `/soorten-fotoshoots`: `shoot-newborn` photo. Shoot pages: first story photo (`story-verwachten`) at all widths.
|
||||
- `/contact`: paragraph `p` at 412 and 768; aside photo (`contact-aside-photo`) at 1350. `/privacyverklaring`: logo `img` at 412, `h1` at 768 and 1350.
|
||||
- Raw output: `probe25.out` in the scratch folder.
|
||||
|
||||
## V-D7: lazy logos and contact portrait
|
||||
|
||||
- Image KB requested before the real LCP on non-home pages: the light logo (24.8 KB) is no longer in the list; only the dark logo (23.5 KB) is (`/over-mij` 107 KB, soorten 97 KB). `/` still requests both logos (total 272 KB with the 224 KB hero). The contact portrait is 5.9 KB on mobile contact, where it is displayed.
|
||||
- Score and LCP A to U2: see the BR-DMS-03 list. The gains on photo pages combine this with the visible `Reveal`; this measurement does not separate the two.
|
||||
- Logo flash, probe on client navigation `/` to `/over-mij` and back (logos already cached from `/`): no frame with a displayed, incomplete logo.
|
||||
- **Finding**: fresh load of `/over-mij`, then client navigation to `/` at 1350 px: the light logo is requested only then, and 1 sampled frame shows the light logo displayed (46 px) while incomplete with the dark logo already hidden (an empty logo area for that frame on a local server; longer on a slow network). At 412 px no such frame (light logo not displayed). Not fixed; for the owner.
|
||||
|
||||
## V-D10: hero photo and LCP in a trace
|
||||
|
||||
Trace not obtained: `pnpm exec lighthouse` does not exist (`Command "lighthouse" not found`; only `@lhci/cli` is installed). V-D10 stays open. Known: the LHR of `/` names the heading `span` as the LCP element on mobile and desktop, and the probe agrees.
|
||||
|
||||
## Moved scripts against FCP and LCP (unthrottled, 1350 x 940)
|
||||
|
||||
`/`: three `inline-*.js` files, request start 8 ms, finished at 17 to 20 ms; FCP 76, LCP 76. `/over-mij`: finished at 17 ms; FCP 56, LCP 56. All before first paint.
|
||||
|
||||
## Real against simulated LCP (mobile, median-LCP run)
|
||||
|
||||
Observed LCP is 46 to 100 ms on localhost (unthrottled) against 2.6 to 4.4 s simulated. Image bytes requested before the observed LCP (U2): `/` 272 KB, `/over-mij` 107, `/portfolio` 78, soorten 97, newborn 97, zwangerschap 127, gezin 141, loveshoot 176, portret 120, contact 29, privacy 23, not found 23 (image requests with `networkRequestTime` before the observed LCP).
|
||||
|
||||
## State after the measurement
|
||||
|
||||
Build D was temporary. `PortfolioGallery.tsx` (8f7d3b24...) and `routes/portfolio.tsx` (aae65baf...) were restored from scratch copies; SHA-256 equal, `git status --short` equal to before (clean), `dist` rebuilt from the real code (48em and 64em preload links present), then E2 was measured. No git write command was used.
|
||||
+4
-4
@@ -8,16 +8,16 @@ Status values: `open` (not yet settled), `OK`, `BLOCKED`, `DEVIATION`.
|
||||
| V-D2 | With JavaScript on, the twin image is never requested (request log on `/soorten-fotoshoots` and `/portfolio` before any scroll), and the preload scanner does not read inside `noscript`. Also that Emotion writes the shared class once (one `<style>` for the two images). | 3 | Twin never requested with JavaScript on; one Emotion class. | Built dist, Chrome, 1350x940, JavaScript on. /soorten-fotoshoots: 12 image requests in total, 12 distinct URLs, none twice; 4 before load + 500 ms (both logos, newborn, newborn-detail: photos in or near view); set of requested URLs equals the set of currentSrc of the displayed img elements. /portfolio: 18 requests, 18 distinct, none twice, 8 before load + 500 ms (logos and 6 photos near the viewport), requested set equals the displayed set. No console message. Raw HTML: no preload link for a twin. Emotion: one style block data-emotion="css qvpj2z" (soorten) and one for css-1qa5q9y (portfolio tile class), shared by the deferred image and its twin. | OK |
|
||||
| V-D3 | `<noscript>` inside `<head>` and inside the header and the panel hydrate without a mismatch in `hydrateRoot(document)` (the shell hydrates `<html>`), and the head hoisting of `HeadContent` keeps the CSS in the head. | 3 (head, twin), 7 (header and panel) | Settled OK: head, twin, header and panel hydrate cleanly with both noscript menu links in a fresh build. | Fresh pnpm build (CP2), pnpm e2e: hydration.e2e.ts, smoke and interaction pass on all 11 pages, /404, /404.html, /onbekend and /soorten-fotoshoots/onbekend at 390 and 1280 px: no console error or warning, no CSP violation, no React #418. Raw HTML: one noscript with one style in head on every page (displayHtml.e2e.ts); header and mobile panel carry nav-noscript-open and nav-noscript-close; document noscript count equals the HTML count. | OK |
|
||||
| V-D4 | Fragment navigation to a fixed element does not scroll in Firefox and Safari, and Chrome on Android (only desktop Chrome was probed). If one scrolls, the fixed `header` target moves to a zero-size fixed element. | 17 | Chrome part OK; Firefox, Safari and Chrome on Android need a person. | e2e/noJsMenu.e2e.ts: opening via #mobile-nav and closing via #site-header leave scrollY at 300 on /over-mij and / at 390 x 800. | open (Chrome OK, other browsers need a person) |
|
||||
| V-D5 | The comparison of BR-DMS-05 (E against D on mobile, three runs each, same machine) and its report to the owner when E is below 0.9 or below D. | 26 | | | open |
|
||||
| V-D5 | The comparison of BR-DMS-05 (E against D on mobile, three runs each, same machine) and its report to the owner when E is below 0.9 or below D. | 26 | E mobile 0.93, D 0.91, E2 0.93; desktop 0.99 for all; rule not triggered | measurement-display.md | OK |
|
||||
| V-D6 | Height of every `Reveal` at 320, 360, 412, 768, 1350 px wide against the limit of BR-RVL-13. | 18 | Every hidden Reveal stays far below the limit (320: 4233 px, 360: 4833, 412: 6358, 768: 8033, 1350: 7333). | e2e/reveal.e2e.ts, hidden Reveals after hydration plus 2 s, tallest per page (px) at 320/360/412/768/1350: / 649/649/649/998/585; /over-mij 465/443/512/230/408; /portfolio 481/541/619/511/223; /soorten-fotoshoots 551/479/391/470/470; newborn 784/712/712/600/624; zwangerschap 784/712/712/600/600; gezin 784/712/712/600/600; loveshoot 736/688/688/568/336; portret 736/688/688/568/336; /contact 670/0/0/0/0; /privacyverklaring 0. | OK |
|
||||
| V-D7 | After the gate measurement: whether `loading="lazy"` on the light logo and the contact portrait changes score and LCP as estimated; that the logo does not flash on `/` or on a client navigation from `/` to another page and back. | 24, 25 | | | open |
|
||||
| V-D7 | After the gate measurement: whether `loading="lazy"` on the light logo and the contact portrait changes score and LCP as estimated; that the logo does not flash on `/` or on a client navigation from `/` to another page and back. | 24, 25 | Light logo gone from pre-LCP bytes off the home page; no flash with cached logos; one empty-logo frame on fresh /over-mij then client nav to / at 1350 px | measurement-display.md | DEVIATION |
|
||||
| V-D8 | The raw-HTML e2e tests that count `img` or links (`rawHtml.e2e.ts`) with `noscript` content present. | 20 | The unit-1 raw-HTML tests pass unchanged with noscript content present. | rawHtml.e2e.ts uses toContain on nav targets and removes noscript before reading text; it counts no img and no links, so no change was needed. Only the comment in the interaction.e2e.ts V15 block was updated. | OK |
|
||||
| V-D9 | The visible focus ring of the two menu links in Chakra's reset (a bare `a` outside Chakra components). | 17 | The focused open and close links have a visible focus ring. | e2e/noJsMenu.e2e.ts: computed outline width of nav-noscript-open and nav-noscript-close, reached with Tab, is at least 2 px. | OK |
|
||||
| V-D10 | Whether Chrome really excludes the viewport-filling hero photo from LCP (so text is the candidate on `/`), by a trace. Only explains `/`; changes no rule. | 25 (only if a trace exists) | | | open |
|
||||
| V-D10 | Whether Chrome really excludes the viewport-filling hero photo from LCP (so text is the candidate on `/`), by a trace. Only explains `/`; changes no rule. | 25 (only if a trace exists) | Trace not obtained: pnpm exec lighthouse not found | measurement-display.md | BLOCKED |
|
||||
| V-D11 | The first React 19 upgrade: the hoisting facts of BR-PRE-01 (tested by BR-PRE-04). | 15 | React hoisting facts hold in the current build. | e2e/displayHtml.e2e.ts, 70 tests over 11 pages and /404.html: every preload as=image href once per file; every priority photo has its preload with fetchpriority high from the img src; /contact has one media link (min-width: 64em) for the aside photo and no other photo link; /portfolio has React link for tile 1 plus exactly two media links (48em, 64em) for two other photos; light logo preload only on /; no preload for a lazy photo unless a non-lazy img uses the same file (the dark logo in header and footer) and none for a twin except the two media links of /portfolio; data-reveal="hidden" nowhere. | OK |
|
||||
| V-D12 | Visual check without JavaScript on every page at 390 and 1280 px (screenshots through `E2E_SCREENSHOT_DIR`, looked at, not compared): header, hero, nav, footer, empty areas left by hidden controls (`mb` of the filter bar, the gate's margin). | 16 | Looked at 24 screenshots (11 pages and the 404 page at 390 and 1280 px, JavaScript off). No broken-image box, no empty frame, no hidden control left behind; /over-mij has no empty area under the h1 (V15 resolved). Remaining empty-looking areas are design-accepted (see Findings). | e2e/noJs.e2e.ts passes 24 of 24: no visible button, every data-reveal opacity 1, every displayed img complete with naturalWidth > 0, every img[data-deferred] display none, displayed named imgs equal getByRole("img"), no email-pending or requires-js visible, no mailto link, footer has no Contact heading and no empty column; / has no hero controls and shows slide 1, one review; /portfolio has no filter bar and every tile href ends in an image file. Findings below. | OK |
|
||||
| V-D13 | `<picture>` with `display: contents` around the grid tile `DeferredImage`: the tile keeps its size, the hover zoom, the focus ring and the reading order; the `<source>` photo is requested once when `src` is set later; hydration is clean for tiles 2 and 3 (`<picture>` and `noscript` children). | 4 (proof), 14 (test), 19 (e2e) | Tile keeps size, zoom and focus; one request per photo; clean hydration in the browser (Step 4); e2e confirms request counts per width. | Step 4 evidence unchanged. e2e/images.e2e.ts at 390/768/1350: distinct x of the first row 1/2/3 equals the number of grid photos requested before load; every grid photo requested exactly once after a full scroll; no console message. Tile sizes against Step 4 were not compared again (Step 4 recorded identical boxes, not their values). | OK |
|
||||
| V-D14 | Which tiles Chrome reports as the largest paint at 412, 768 and 1350 px after the change (the probe saw `f41a1916` at 412, `f41a1922` at 768 and `f41a7883` at 1350 without the eager row), and that these are first-row tiles. | 25 | | | open |
|
||||
| V-D14 | Which tiles Chrome reports as the largest paint at 412, 768 and 1350 px after the change (the probe saw `f41a1916` at 412, `f41a1922` at 768 and `f41a7883` at 1350 without the eager row), and that these are first-row tiles. | 25 | Tile 1 at 412, tile 2 at 768, tile 3 at 1350: all first-row tiles | measurement-display.md, probe output | OK |
|
||||
| V-D15 | `/portfolio?shoot=gezin`: the preloads of unfiltered tiles 2 and 3 and the remounted first row of the filtered list do not fetch the same photo twice. | 19 | No image URL is requested twice on /portfolio?shoot=gezin. | e2e/images.e2e.ts: at 1350 x 940, after the filter applied (fewer tiles than before hydration) and a scroll through the page, every image URL was requested exactly once. | OK |
|
||||
|
||||
## Findings (Step 16 screenshots, Step 17 run)
|
||||
|
||||
+4
-4
@@ -433,7 +433,7 @@ All files in `web/e2e/`, Dutch titles and comments like the neighbours, using th
|
||||
|
||||
Preconditions for all of Part H: Step 23 green; the main session has confirmed that it holds port 4173 for this work; `git status --short` recorded to `<scratch>/status-before.txt`; `pnpm build` just run. The method is the one of unit 1: `lhci collect`, 3 runs per URL, median, same machine. Lighthouse runs may be backgrounded; wait for them (rule 13).
|
||||
|
||||
### Step 24: Measurement U2 and the moved-scripts check (BR-DMS-02, 03) [ ]
|
||||
### Step 24: Measurement U2 and the moved-scripts check (BR-DMS-02, 03) [x]
|
||||
|
||||
- Executor: Haiku for the runs, Sonnet for the script and the report.
|
||||
- Implements: BR-DMS-02, 03; NFR-PR-P-01, P-03; V-D7 (first part).
|
||||
@@ -444,14 +444,14 @@ Preconditions for all of Part H: Step 23 green; the main session has confirmed t
|
||||
4. Trace (BR-DMS-02): for each of the 12 URLs keep the trace of one mobile run through Lighthouse gather mode (`pnpm exec lighthouse <url> -G --output-path=<scratch>/trace/<slug>` with the mobile defaults of `lighthouserc.cjs` and the Chrome flags it uses, against `pnpm preview --port 4173 --strictPort` started and stopped by you). If `pnpm exec lighthouse` does not exist or gather mode fails, write "trace not obtained" with the error in the report; this is not a STOP.
|
||||
- Done when: the folders and the script output exist; the comparison table is ready for Step 27.
|
||||
|
||||
### Step 25: Largest-paint probe, logo effect and moved scripts (V-D7, V-D10, V-D14) [ ]
|
||||
### Step 25: Largest-paint probe, logo effect and moved scripts (V-D7, V-D10, V-D14) [x]
|
||||
|
||||
- Executor: Haiku runs, Sonnet reads.
|
||||
- Implements: BR-PRE-05, 06; V-D7, V-D10, V-D14.
|
||||
- Do: (a) probe (scratch, unthrottled Chrome) of the largest-contentful-paint element per page at 412 x 823, 768 x 1024 and 1350 x 940 (a `PerformanceObserver` for `largest-contentful-paint` registered by `addInitScript`; last entry after 2.5 s; record tag, `url`, `size`, alt text) for the eleven pages; check against business-rules.md BR-PRE-06 and the domain-entities.md § 6 table, and for `/portfolio` state which tile it is at each width (V-D14, expected: a first-row tile). (b) Effect of the lazy logo and contact portrait (V-D7): from `<scratch>/u2` and the unit-1 numbers, `imgKB` before the real LCP on non-home pages should be about 25 KB lower; report the score and LCP change per page, and confirm by a probe on `/` after a client navigation to `/over-mij` and back that the logo does not flash (take frames). (c) If a trace was obtained (Step 24.4), state whether the hero photo is the LCP candidate on `/` (V-D10). (d) Time of the moved `inline-*.js` files against FCP and LCP for `/` and `/over-mij`, as unit 1 did.
|
||||
- Done when: results written for Step 27. A first-row tile that is not the largest paint, or a logo flash, is a finding for the report, not a STOP.
|
||||
|
||||
### Step 26: Portfolio comparison, build E against build D (BR-DMS-05, V-D5) [ ]
|
||||
### Step 26: Portfolio comparison, build E against build D (BR-DMS-05, V-D5) [x]
|
||||
|
||||
- Executor: Haiku for the runs, Sonnet for the temporary change and the report. Exclusive.
|
||||
- Implements: BR-DMS-05; NFR-PR-P-01; V-D5.
|
||||
@@ -464,7 +464,7 @@ Preconditions for all of Part H: Step 23 green; the main session has confirmed t
|
||||
6. **Decision rule (BR-DMS-05):** if the median mobile score of E is below 0.9, or below that of D, or if E and E2 differ by more than 0.02 (noise; then repeat with 5 runs and say so), the result goes to the owner with both tables before anything else is built on it. Do not switch to D, do not edit `lighthouserc*.cjs`, do not "fix" the score. Desktop must stay at 0.9 or better; below that is also reported.
|
||||
- Done when: the three result sets exist, the files are restored (checksum equal, `git status --short` equal to the recorded one), `dist` was rebuilt from the real code, and the tables and the decision-rule outcome are ready for Step 27.
|
||||
|
||||
### Step 27: Measurement report [ ]
|
||||
### Step 27: Measurement report [x]
|
||||
|
||||
- Executor: Sonnet.
|
||||
- Implements: BR-DMS-01 to 05; NFR-PR-P-01, P-03; CP5.
|
||||
|
||||
Reference in New Issue
Block a user