Polish pass after the big refactor in 2d114fc.
== Header elevation slot propagated ==
shared/elevation.{js,css} surface a header checkbox for admins.
30-minute sudo-style cookie window (Max-Age=1800, SameSite=Lax).
Only renders when /.profile/access reports can_elevate=true; quiet
for non-admins. Slot added to all 7 tool templates and concat'd
into all 7 build.sh files; admin in any tool now sees the toggle.
Three text-rename ride-alongs in archive/classifier/transmittal
templates: "Add Local Directory" → "Use Local Directory" (the same
rename that landed in browse earlier in this branch).
== Docs ==
- CLAUDE.md gets an "Admin elevation is sudo-style" paragraph in
the "Things that bite if you forget" section.
- AGENTS.md gets a dedicated "Admin elevation (sudo-style)" section
alongside "Bearer tokens" — same depth as the existing auth docs.
== Helper file splits ==
The retired form editor's shared helpers got bundled into a single
zddc_admin.go in the cleanup; that name is now misleading. Split by
concern:
- admin_helpers.go: hasAnyAdminScope (the only admin-specific helper)
- paths.go: resolvePath, urlPathOf, chainDirs (URL ↔ filesystem path
math — used by several profile / zddc-file handlers)
- profile_assets.go (renamed from zddc_admin_assets.go): custom CSS
pipeline. URL renamed from /.profile/zddc/assets/ → /.profile/assets/
since /.profile/zddc/ no longer hosts an editor.
- treeEntry moves to profilehandler.go (alongside AccessView, its
only consumer).
- writeError moves to profileprojects.go (its only consumer).
== Smell cleanup ==
- zddc.HasAnyAdminGrant(fsRoot, email) — new elevation-independent
primitive that walks the cascade and reports whether email is named
in any admin: list anywhere. Replaces the synthetic-elevated probe
hack in enumerateAccess (`Principal{Email, Elevated: true}` was
"lying" to the elevation gate to ask what it would say). The handler's
hasAnyAdminScope collapses to a 4-line wrapper that gates on
p.Elevated and delegates.
- Access-log middleware records `elevated` per request, so forensics
can distinguish "admin acting as user" from "admin exercising power."
- browse/js/app.js's ?file= deep link walks multi-segment paths. Each
intermediate segment is matched + expanded; the leaf gets
selected/previewed. Auto-shows hidden when any segment starts with
. or _. Silently no-ops on unresolved segments.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
New shared/zip-source.js: a ZipDirectoryHandle / ZipFileHandle pair
that exposes a JSZip instance behind the File-System-Access surface
(values/entries/keys, getDirectoryHandle/getFileHandle, getFile) —
read-only, with a zip-slip guard. Mirrors shared/zddc-source.js's
HTTP polyfill. Wired into archive's and browse's build.sh (both
already bundle JSZip).
archive: a .zip whose name minus ".zip" parses as a transmittal-folder
name is now scanned as that transmittal folder. Offline, the zip is
opened in the browser (ZipDirectoryHandle) and its members enumerated
exactly like an uncompressed folder's files — table/export/hash paths
are unchanged (they go through file.handle.getFile()). Online, the
scanner recurses into the server's "<…>.zip/" virtual-directory
listing, so members come back as "<…>.zip/<member>" URLs the server
extracts on demand — no whole-zip download.
browse: the offline (file://) zip path is migrated onto the shared
adapter — expanding a .zip now opens it as a ZipDirectoryHandle and
its members become ordinary dir/file nodes handled by the normal
fetchFsChildren path (nested zips fall out by recursion). The bespoke
flat-entry walker (loadZipChildren / setZipDirChildren / zipEntries /
zipParentId / zipPath / _zipSyntheticDir) is gone — one zip
implementation repo-wide. Markdown members inside a zip are flagged
read-only (the ZipFileHandle refuses createWritable; server "<…>.zip/"
URLs 405 on PUT).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
User report: project root listings showed both "Archive" (PascalCase on
disk) and "archive (empty)" (lowercase virtual) — confusing duplicates.
This sweep:
1. Test fixture migrated to lowercase canonical folder names.
tests/data/test-archive.sh now creates archive/, received/, issued/
on disk. Three projects also get human-friendly .zddc titles
("Wabash Industrial Refit — Phase 1", etc.), and Project-3 carries
a display: override demonstrating the new map. Party names
(PartyA/B/C) stay unchanged — non-canonical.
2. New .zddc display: schema. Maps a child entry's on-disk name to a
human-friendly label. The on-disk name stays canonical (lowercase
for project-root folders); only the rendered label changes. Match
is case-insensitive. Example:
display:
archive: "Records"
working: "In-Progress"
No upward cascade — a parent .zddc doesn't relabel grand-children;
each directory sets display: on its own children.
3. listing.FileInfo gets a DisplayName field. fs.ListDirectory reads
the directory's .zddc display map and stamps DisplayName per entry.
The field is omitempty so listings without overrides stay
byte-identical to before.
4. Virtual canonical project-root folders (archive/working/staging/
reviewing) are now emitted by zddc-server (fs.ListDirectory) at any
project root where the on-disk variant is absent in any case. This
replaces the client-side injection in browse and lets the display:
map apply to virtual entries the same way it applies to real ones.
Browse drops its withVirtualCanonicals helper; the loader carries
display_name through from the server's listing.
5. Archive app project picker dropdown shows the .zddc title of each
project (sourced from ProjectInfo.Title in the server's project
list), falling back to the folder name when no title is set. When
they differ, the folder name is rendered in muted mono after the
title for traceability. data-name still carries the canonical
folder name so URL state stays stable.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
When zddc-server auto-serves the archive tool at a sub-folder URL (e.g.
/Project-1/Archive/PartyA/Issued, no trailing slash, no archive.html
filename), the tool's autoConnectHttpSource was stripping the last URL
segment under the assumption it was a filename. Result: archive scanned
the PARENT directory (/PartyA/) instead of /Issued/, showing
transmittals from sibling folders the user hadn't asked for.
Fix: detect the URL shape before deciding what the scan root is.
- Trailing slash → URL is already a directory; use as-is.
- Last segment has a dot → filename (archive.html); strip to parent.
- Last segment has no dot → bare directory name (auto-served URL);
treat the entire path as the scan root and append '/'.
Now /Project-1/Archive/PartyA/Issued shows only the transmittal folders
inside Issued/ — the locally-mounted-folder experience the user expected.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Removes every runtime CDN load. The "ship the record player with the
record" philosophy: a downloaded .html file works offline against any
file the user can open, with no network dependency at runtime.
Newly vendored under shared/vendor/:
- xlsx.full.min.js (SheetJS, 928 KB) — XLSX/XLS preview
- utif.min.js (UTIF, 57 KB) — TIFF preview
Already there but now used by mdedit too:
- jszip.min.js, docx-preview.min.js
Call sites updated to drop the `await loadLibrary(URL)` pattern —
since the vendor JS is concatenated into the inline <script> at build
time, window.XLSX / window.JSZip / window.UTIF / window.docx are
available synchronously from page load.
Per-tool changes:
- archive/build.sh: +xlsx, +utif
- classifier/build.sh: +xlsx, +utif
- transmittal/build.sh: +xlsx, +utif
- mdedit/build.sh: +jszip, +docx-preview, +xlsx, +utif
(mdedit was the only tool not yet
bundling any of the preview deps)
- browse/build.sh: +utif
- archive/js/table.js, classifier/js/preview.js,
transmittal/js/files-preview.js, mdedit/js/file-tree.js (×2):
drop the `await loadLibrary('…cdn…')` lines.
- shared/preview-lib.js:
drop the loadLibrary(UTIF) / loadLibrary(JSZip) wrappers; assume
window.UTIF and window.JSZip are present.
Net bundle-size delta after baking:
archive: +990 KB → ~1.47 MB
browse: +57 KB → ~292 KB
classifier: +990 KB → ~1.43 MB
mdedit: +1100 KB → ~2.09 MB
transmittal: +990 KB → ~1.63 MB
Docs (AGENTS.md, ARCHITECTURE.md) updated: removed the "runtime CDN
loading exception" paragraph and the table row that flagged xlsx as
CDN-loaded.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three tools (archive, browse, classifier) independently implemented
an empty-state pattern with three different CSS class naming
conventions and slightly different rules:
archive: .empty-state + .empty-state-content (BEM-less)
browse: .empty-state + .empty-state__inner (BEM)
classifier: .empty-state + .empty-state-content (BEM-less)
Same visual intent ("nothing's loaded yet — here's a welcome card
with instructions"), implemented three times with subtly different
spacing, no shared body styling for h2/p/ul/li, and incompatible
class names that prevented a future tool from copy-pasting the
pattern.
Promote a single consolidated rule set to shared/base.css using
BEM naming throughout:
.empty-state — base (flex centered, padding)
.empty-state--overlay — modifier: position absolute,
top 50px to clear app-header,
z-index 10. Used by archive
and classifier (their empty
states sit OVER the main
layout).
.empty-state__inner — content card (left-aligned,
text-muted, max-width 640)
.empty-state__inner--centered — modifier: tighter max-width
500, centered text, 2rem
padding. Used by tools whose
welcome screen reads as a
centered card.
.empty-state__inner h2/p/ul/ol/li — typography defaults
.empty-state__inner .note — italic small-print
.welcome-list — bullet list with left-aligned
text + auto margins; safe to
nest inside a centered card.
Per-tool changes:
- archive/template.html, archive/js/app.js: rename
.empty-state-content → .empty-state__inner empty-state__inner--centered;
add .empty-state--overlay to the outer .empty-state container.
Also the runtime-injected unsupported-browser markup in
showUnsupportedBrowserMessage() and the showHttpErrorState
selector.
- classifier/template.html: same renames.
- archive/css/layout.css + components.css: delete .empty-state*
and .welcome-list rules (now in shared).
- classifier/css/layout.css: same. Keep .empty-state.drag-over
locally — classifier is the only tool whose empty state acts
as a drop target.
- browse/css/base.css: delete .empty-state* (shared covers it).
browse's template was already using .empty-state__inner so no
template change needed.
LOC: shared/base.css gains ~70 lines; per-tool overrides lose ~85
combined. Net -15. More importantly, future tools can reuse the
pattern by adding two divs and (optionally) the --centered or
--overlay modifiers; no copy-paste required.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
archive/js/events.js defined a 10-line debounce function inside the
events IIFE that was never called anywhere in archive/. Dead code,
confirmed by grepping the whole archive/ tree for debounce(.
The plan was to extract debounce to shared/util.js so this file and
mdedit/js/utils.js could share one implementation, but mdedit's debounce
has only one caller (editor.js) so a shared abstraction would be
premature. Just delete the dead copy.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Same pattern as the browse fix. archive, transmittal, classifier
previously CDN-loaded jszip + docx-preview on first preview of a
.zip / .docx file via shared/preview-lib.js's loadLibrary helper.
That meant each first-preview blocked on a CDN round-trip + parse,
and broke entirely under restrictive networks or CSPs.
Vendor both libs under shared/vendor/ and concat them at the top of
each tool's build, ahead of init.js. window.JSZip + window.docx are
now defined immediately on page load. Drop the redundant loadLibrary
calls (and classifier's stray <script src="cdn..."> tag in the
template, plus archive's bespoke loadJSZip helper in export.js).
xlsx (SheetJS) intentionally stays CDN-loaded — at ~900 KB it's too
large to inline, and only fires on .xlsx preview which is a rarer
path.
Bundle size impact (uncompressed):
archive: 304 KB → 476 KB (+172 KB)
transmittal: 449 KB → 621 KB (+172 KB)
classifier: 252 KB → 424 KB (+172 KB)
With the gzip middleware (~75% reduction on HTML) and ETag-cached
revalidation now in place, the wire-size delta is ~40 KB per tool
on the first load and 0 on every subsequent load until redeploy.
drag-drop.js and the unsupported-browser handler in app.js both
referenced getElementById('app'), but the template's root has
id="appContainer". The mismatch was masked in production because
sourceMode='http' skips dragDrop.init() — only file:// (sourceMode=
'local') tripped over it, throwing "Cannot read properties of null
(reading 'addEventListener')" at app load.
Surfaced while header-standardizing the other tools; fixed by
pointing both callers at #appContainer.
User report: opening an .html file with a '../.archive/' hyperlink in
a new tab works (zddc-server intercepts and serves the right file),
but clicking the same link inside the file previewer does nothing.
Two combined causes:
1. The previewer's iframe was loaded from a blob: URL (built from
the file's bytes). Relative URLs in the iframe resolve relative
to the blob URL — '../.archive/X.html' becomes 'blob:.../.archive/
X.html', which is gibberish. The browser never sends a request to
the server, so the .archive interception never fires.
2. sandbox="" disables every iframe capability including popups,
so even <a target=_blank> is silently swallowed.
Fix per tool:
- archive (table.js): for HTML preview, use file.url (the real
server URL) directly when available; fall back to blob only for
File-System-Access-API mode where there's no server to intercept
anyway. Now relative links in archived HTMLs resolve against the
actual server origin and the .archive interception fires as
designed. Sandbox loosens to allow-same-origin + allow-popups +
allow-popups-to-escape-sandbox so resources within the iframe
load and link clicks (default target / target=_blank / middle-
click) work normally. allow-scripts is intentionally NOT set —
archived HTML still cannot run JS in the popup's origin.
- transmittal (files-preview.js) + classifier (preview.js): same
sandbox loosening for consistency. These tools' files are
typically local (FileSystemAccessAPI), so the file.url branch
doesn't apply — relative URLs that depend on a server still
won't resolve in local mode (intrinsic limitation, no server).
Tested behavior preserved:
- PDFs: unchanged (no sandbox, browser's PDF viewer handles).
- Images / docx / xlsx / tiff / zip / text: unchanged.
- HTML in zddc-server-backed archive: relative '../.archive/' links
now navigate the iframe to the correct target file.
HTML files in the file previewer (archive, transmittal, classifier
popups) were dispatched to the text renderer because 'html'/'htm'
are in shared/preview-lib.js's TEXT_EXTENSIONS (which is shared with
the syntax-highlighting code path). Result: opening an .html file in
preview showed its source as a <pre> block, not the rendered page.
Fix in each tool's popup builder + dispatcher:
- Add 'html' / 'htm' to the iframe branch (alongside pdf), so the
popup ships an <iframe src=blob:...> instead of an empty
#previewContent div. The blob's MIME type from getMimeType()
is already 'text/html', so the browser renders natively.
- Skip the text-render dispatch for html/htm (the iframe is enough).
- Add to the HTML iframe so an arbitrary archived
HTML file cannot run scripts, navigate top, submit forms, or
open popups in the popup-window's origin. PDFs don't need this
since the browser's PDF viewer is sandboxed natively.
classifier/js/preview.js uses a getPreviewType() switch instead of
chained ifs; adds 'html' as its own preview type (checked BEFORE
'text' since html is in TEXT_EXTENSIONS).
mdedit already handled HTML specially (file-tree.js has an isHtml
check); no change there.
TIFF was already rendered via the shared zddc.preview.renderTiff
canvas viewer in all four tools — no change needed for that path.
If TIFF preview appears broken on the live prod server, that's the
v0.0.9-alpha-baked-in image; the fresh stable redeploy fixes it.
Four entangled change-sets from one session, committed together because
their file-level overlap (build.sh, docs, embedded/, watcher.go, …) makes
post-hoc separation noisy:
* fix(archive): nested-party + folder-type cascade
transmittalIsUnderVisibleParty short-circuited on the first matched
party segment, only checking the immediately-next segment for a
folder-type marker. Paths like BM/sub/Issued/<txn> bypassed the Issued
toggle entirely. Replaced with isUnderHiddenFolderType (full-path) +
any-segment party match. Eight new Playwright cases pin the contract
in tests/archive-cascade.spec.js.
* refactor(zddc-server): scope .archive index by project
archive.Index now buckets by top-level segment
(.ByProject[<project>].ByTracking[<tracking>]). Resolve and AllEntries
take a project parameter; handler extracts it from contextPath's first
segment. /.archive/ at root returns 404 — stable refs must be
project-rooted. Within-project (tracking, rev) collisions emit a WARN
with both paths. Cross-project tracking-number duplicates no longer
collide.
* perf(zddc-server): lazy-load expensive bits of the profile page
serveProfilePage now ships a minimal shell: Email, EmailHeader,
IsSuperAdmin (root .zddc only). Visible projects + admin subtrees +
editable scaffolds populate client-side via /.profile/access. Subtree-
admin scaffolds live in <template id="tmpl-subtree-admin">; pure
non-admins receive no live admin form. ScanZddcFiles now memoized,
invalidated on .zddc events by the watcher and writer helpers.
* feat: lockstep release + redesigned releases page
sh build.sh --release [version|alpha|beta] is the canonical lockstep
cut: every tool (5 HTML + zddc-server) bumps to the same coordinated
version. zddc-server binaries now committed under website/releases/
with the same cascade chain as HTML tools (no more Codeberg release-
asset publication). zddc/release.sh deprecated (kept as a guard);
shared/publish-codeberg-release.sh removed.
Releases page redesigned as an action-first install guide: hero +
version dropdown that rewires every download link, channel chips for
always-visible alpha/beta access (state-aware labels: "tracks stable"
vs "active dev"), Path A (zddc-server with platform auto-detect from
UA), Path B (5 standalone tool HTMLs), version-pinning empowerment
narrative (drop-a-copy vs .zddc apps: cascade), channels explainer.
Channel-link verifier asserts every <tool>_{stable,beta,alpha}.html
resolves at the end of every build. Bootstrap-friendly: zddc-server
artifact checks skip until the first lockstep cut anchors the chain.
Tests: 167 Playwright + all Go packages green.
Docs: CLAUDE.md, AGENTS.md, ARCHITECTURE.md, zddc/README.md updated.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds shared/preview-lib.js with two cross-tool renderers:
- renderTiff (UTIF.js, lazy-loaded from CDN; PDF-style toolbar with
page nav, zoom, fit-width/fit-page; multi-page TIFFs decode lazily)
- renderZipListing (JSZip; sortable name/size/modified table, sticky
header, host-grouped paths)
Wired into the four tools that have a preview surface (archive, classifier,
mdedit, transmittal). Cross-document compatible so the same renderer works
for popup-window tools (archive/classifier/transmittal) and inline tools
(mdedit). Archive previously had no image branch at all — now previews
JPG/PNG/GIF/WebP/BMP/SVG natively, plus TIFF via UTIF, plus the ZIP listing.
Adds the dark-blue rounded-square favicon to each app's header (left of
the title) and to the website navigation. Single inline SVG, sized via
.app-header__logo (in shared/base.css) for tools and .brand-logo (in
website/css/style.css) for the website. Self-contained — the SVG carries
its own background, no wrapper styling needed.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Bundles a stretch of in-progress work across the SPA tools so the
tree returns to a coherent shippable state ahead of cutting a new
zddc-server stable image:
- landing: substantial rework of the project picker (sortable/filterable
table, presets refactor, ?projects= filter, ?v= channel propagation,
loading/error states)
- archive: presets cleanup, source.js refactor, filtering/url-state
alignment with the landing page
- mdedit: file-system module split, resizer, file-tree improvements,
base/toc styling tweaks
- transmittal/classifier: small template touch-ups for shared chrome
- shared: build-lib.sh helpers, new favicon.svg
- bootstrap, build.sh: pick up the channel-aware install/track zip
generation
- tests: new landing.spec.js, expanded archive/mdedit/build-label specs
- docs: CLAUDE.md picks up the zddc-server section and freshens the
alpha-build exception note
- regenerated artifacts: install.zip, track-{alpha,beta,stable}.zip,
*_alpha.html — these are produced by `sh build.sh` and per project
convention are committed alongside the source changes
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two small additions to the project-filter / channel-selector flow that
already worked end-to-end for HTTP-mode but were missing in the local
File-System-Access path and across landing→archive navigation:
* archive: scanLocalRecursive now applies window.app.projectFilter at
depth 0, mirroring the HTTP source's existing filter at source.js:316.
Loading archive.html?projects=A,B in local mode (file://) now virtually
merges A and B into one combined view, same as HTTP mode does today.
* landing: openArchive() reads ?v= from its own URL and passes it through
to the archive.html link it generates. This keeps the user on the same
channel (alpha/beta/stable/<version>) when they cross from the project
picker to the archive — without it, alpha-channel users would silently
drop back to whatever the deployment-default channel is at the
archive.html boundary.
Test exercises the local-mode filter via the existing mock-fs-api
fixture: three top-level projects, projectFilter set to {A, B}, scan
produces only A's and B's files. (The url-state.restore() URL parsing
path is well-trodden in the HTTP case — the test sets projectFilter
directly to isolate the new source.js change from a pre-existing init()
fragility in the mock environment.)
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ZDDC — Zero Day Document Control. A file-naming convention plus five
single-file HTML tools (archive, transmittal, classifier, mdedit,
landing) and an optional Go HTTP server (zddc-server) with ACL and a
virtual archive index. Self-contained, offline-capable, dependency-free.
See README.md for an overview, AGENTS.md and ARCHITECTURE.md for the
build/release/architecture detail, bootstrap/README.md for the
two-level deployment install pattern, and zddc/README.md for the
HTTP server.
