Timeline

quarry rendered both as first-class surfaces (the docs tab, the file browser) and could find neither. Fossil searches documentation but not source; this does both, from one pass.

One tarball, not a read per file. cat per path costs a subprocess and a store open each: 2,199 files on sqlite is ~31s, against 2.2s for one archive — the tarball is the whole tree in one pass, which is the shape of the question.

Tip-only. Indexing every revision of every file would dwarf the rest of the index many times over to answer a question nobody asks; what a file used to say is a question its own history answers.

ref carries the version, not just the path: a link has to name the check-in the row was indexed at, and resolving a branch name at click time would be a second source of truth about which tree the row describes.

Binaries and files over 1MB are skipped and counted — a search that silently covers less than the tree beside it is the failure this change is about. closed check-in: 8df971b482 user: dmestas tags: tree-search

Reset drops every derived table, doc rows included, and the doc pass only fired on a tip change or once at boot — so a reindex left documentation silently unsearchable until somebody happened to commit, which on a mirror of a quiet project is never. Observed live: fossil's 173 documents went to zero after a rebuild and stayed there.

ReplaceDocs swaps the whole set, so a pass that found documents and could not read them would overwrite a good index with nothing. It now keeps the previous set and says so. A tree with genuinely no documents still writes an empty set, which is true. closed check-in: 24586a4771 user: dmestas tags: doc-reindex

Restores the backend the trunk merge dropped, and adds the two surfaces that get pressed.

The button sits in the stats strip beside the clone count, drawn only when there is an account behind it. Pressed state rides on aria-pressed as well as colour.

/praised is where praise is revisited and withdrawn. It exists apart from /people/{user} because that page lists CONTRIBUTORS derived from the DAG, and a praiser need never have authored anything. check-in: 0cba46d4ed user: dmestas tags: praise-ui

quarry rendered tree documents as a first-class surface — the docs tab, the reading furniture — and could not find them. A reader could open www/quickstart.wiki and get nothing searching for it. Fossil searches c d t w e f; quarry had everything but d.

Indexed at the default branch tip only, asynchronously, on the trigger warmReads already uses: a round that moved the tip, plus once at boot. Not through dagnats — dagnats is optional, a forge without one "simply has no CI rather than a broken one", and search must not inherit that. This is local derivation from the store, the same class of work as the fold, which Reindex already routes around the bus on purpose.

isDocPath is exported so the indexer and the docs listing share one definition of what a document is; two would drift and the drift would be invisible.

Adding the kind exposed how many places the list is written down: the index, two ops validators, the scope chips, hitOf. The validator was missed and refused every doc search before it reached the query. Two tests now hold the four in agreement. closed check-in: 57a2f8e3d2 user: dmestas tags: doc-search

Any signed-in account may praise any repository it can SEE — no capability letter, because the read gate has already decided whether the repository exists for this caller and past that gate praising is not a privilege anyone needs granting.

Per account, not per contributor: someone who has never authored an artifact can still say a repository is worth attention, which is why this cannot hang off the contributor pages.

praise.list filters by the VIEWER, never the owner. Praise is public; the existence of a private repository is not, and this is where the two meet.

UI deferred: the button belongs in the repo stats strip, which is on the unlanded clone-counts branch. check-in: f1a6be9a77 user: dmestas tags: praise

They fell through hitOf's switch with no URL, so the title rendered as — which a browser resolves to the current page. Clicking a forum result reloaded the search. The index found them, the facet chips offered them, and the one step that turns a hit into somewhere to go was missing.

Their refs are not artifact hashes: searchDoc files a forum post under its thread root and a technote under its event id, which is what a reader wants to land on.

The test walks index.SearchKinds rather than a hand-written list, so a kind added to the index without a link here fails instead of shipping a dead result. closed check-in: 69f5a86ccd user: dmestas tags: search-links

The capabilities table sized to an <input size=28> it could not shrink, so the whole page was wider than the viewport and every field ran off the right edge. It scrolls in its own box now.

Webhooks were a three-column table with an input and three buttons in one cell — the same bug, added by me. Rows stack instead.

Help text cut to one line; the noun chips and the field's datalist say the rest. closed check-in: 0a60aac30c user: dmestas tags: settings-mobile

A clone reads and writes nothing, so it is the one kind of forge activity that leaves no trace in the DAG at all.

The sync edge classifies the protocol's cards: clone SEQNO 1 opens a new clone, later rounds do not, so a big store cloned in forty rounds still counts once. Two sinks: the vocabulary event for anything reacting, and a JetStream counter for how many, which converges across nodes instead of losing increments to a read-modify-write. check-in: dc058fa551 user: dmestas tags: clone-counts

Four commits: the fan-out onto the vocabulary, per-subscription event selection, the settings surface, and the RFC correction that preceded them. check-in: 69fc3f06d1 user: dmestas tags: trunk

Subscriptions were CLI-only. The settings page now lists them, takes an endpoint plus its event patterns, pauses, resumes and deletes — gated on 'a' like everything else there. Patterns are checked before saving.

The section is gated on the store being attached, not on the list being non-empty, or the first subscription could never be created. closed check-in: 1c5d758256 user: dmestas tags: hooks-vocabulary

A webhook took everything or nothing. It now names vocabulary patterns: '*', a bare noun for every action on it, or an exact kind. Checked at the moment it is set, because a webhook that silently delivers nothing reads as a broken forge. check-in: e9547dcc49 user: dmestas tags: hooks-vocabulary

A ticket closed by push notified nobody, because only ops queued webhooks. One consumer on quarry.evt.> turns every raised event into deliveries, so arrival path stops mattering.

Deletes kindOf: the delivery's kind was sliced off an idempotency key. check-in: f2ea5f5c1a user: dmestas tags: hooks-vocabulary

The index holds the state an op would have read, so a pushed artifact gets the same event a forge op would have raised. Ticket transitions and landing verdicts compare folded state before and after; forum, wiki and technote need only the artifact or one lookup.

Op sites stop raising derived events, so 2.1's split is now enforced rather than documented. check-in: 06942dcc5d user: dmestas tags: trunk

The forge half of zit#144. A stack is a chain of dependent landings that land bottom-up; quarry renders the chain, refuses accept above the bottom, and flags a segment whose base was rejected.

No new index and no new artifact: the edge is already implied by a landing's target, so a stack is a reading of landings the way a landing is a reading of forum posts. closed check-in: 041ca46258 user: dmestas tags: strata-rfc

checkin.created, branch.created and tag.* come off the artifact; branch.moved comes from the computed branch after RecomputeBranches, because only a branch's first check-in carries a sym- tag.

Raising is not idempotent while folding is, so the fold asks the index which hashes are new before writing them. A rebuild raises nothing. check-in: a1b816cd96 user: dmestas tags: trunk

RFC 0005 picked zit.evt., which the fold consumer already claims with a trailing wildcard — every vocabulary event would have been handed to the artifact parser. Own stream, own root, tests pinning both. check-in: 3bca39d8f5 user: dmestas tags: trunk

I probed HTTP and concluded dagnats had no scheduler. Wrong lens — triggers are NATS and KV native. It has cron, subject, webhook, durable HTTP, worker-contributed kinds and debounce. So quarry builds none of that: it owes well-named subjects, and the vocabulary IS the hierarchy. check-in: 3aef7f3211 user: dmestas tags: trunk

RFC 0005 §1 as internal/evt, named at the eight sites that raise events instead of sliced out of a TxnID — so a ticket change that closes one is ticket.closed, not ticket. Measured the window: a gap IS detectable (first_seq - ack_floor) but the next fetch succeeds silently, so every consumer must check. dagnats has no scheduler; quarry owns the clock. check-in: f08add75f5 user: dmestas tags: trunk

The stream is a one-hour window, so events split: DAG-derived ones are re-derivable after a gap, operational ones need OLTP backing. A cron run and an ad-hoc run are not events — three trigger sources, and schedules take fossil's 'a' so a repo admin owns their own. check-in: 60206e85b2 user: dmestas tags: trunk

Events are noun.action, subscribed by name. Fossil has no such vocabulary — its hooks are artifact-level and its alerts are eight letters — so the delivery semantics come from its warning that a distributed system cannot promise exactly-once. check-in: dc2bb819a3 user: dmestas tags: trunk

A diff is two zit subprocesses on purpose, run at once. Adding their durations printed more time than the whole request took — "rendered in 410.1ms (zit 706.8ms×2)", and "79.94s (zit 159.05s×2)" on a cold tcl page. A number larger than the total reads as a broken page rather than as two calls overlapping.

Record now takes the start instant instead of a duration, because a duration cannot say whether two calls overlapped, and the union of the spans is what a reader actually waited. closed check-in: 4aaf089616 user: dmestas tags: timing-wall

A contributor is a name an author's client wrote on an artifact; an account is a login on this forge. On a mirror the two barely intersect — fossil has 151 contributors and this forge has two accounts, sharing none — so the page says so, and badges the few that are both.

Read-only on purpose. A grant control here would offer to give write access to a string, and on a mirror those strings are chosen by whoever can commit upstream. The door goes to settings, where every row is a real account.

account.list is forge-wide and site-admin only, so a repo admin gets the door without the badges rather than an error. closed check-in: 8fb0a669aa user: dmestas tags: people-admin

"151 contributors" in fossil's About panel went to the fleet page: 481 names across 25 repositories, repo nav gone, search rescoped to the whole forge. people.list has taken a repo parameter all along and nothing in the UI reached it.

The per-person line differs from the fleet's on purpose. Counting repositories inside one repository always says 1, so the span of somebody's work here is the fact worth carrying instead.

Contributors(0) meant a hundred, silently, which is why the listing said "of 100" under an About panel saying 151. It means all of them now. closed check-in: 278e021086 user: dmestas tags: repo-people

zit#136 shipped --mimetype, so a remark written in the composer is stored as the Markdown it previewed as. zit#138: the title card holds the whole -m message, so the heading takes its first line. check-in: 73367a3820 user: dmestas tags: trunk

Every other row in the About panel is icon + label, and .about-sub indents 1.5rem to clear that icon. These two had no icon, so their excerpt and date hung indented under nothing. A wrapping title also needs baseline alignment, or its later lines slide back under the icon.

The note now says who wrote it, as the feed does. check-in: d6a6146502 user: dmestas tags: trunk

Notes sat third in the sidebar, under Tags. What a project has been doing is a better first answer to "what is this" than which versions it has cut — and with no tab, this is a main door to them, which reads poorly third. closed check-in: d1dde173e0 user: dmestas tags: notes-first

zit#134: landing show has to walk the whole store to FIND the record before it can answer about it, and quarry already folded that record out of its own index. zit merge --probe names the two check-ins, so the walk disappears -- 0.005s against 0.18s on this forge's own store, and the gap widens with store size, not with the number of landings.

Probing the same tips the cache is keyed on also closes a race: the key was built from resolved tips while the probe re-resolved a branch name, so a push landing between the two cached an answer under a key it did not describe.

An unresolvable tip now skips the probe instead of keying on ':' -- every landing whose branch could not be resolved shared that one cache entry and could be served another landing's verdict. check-in: bb05986ad8 user: dmestas tags: trunk

Without a tab, /timeline?y=e is where somebody browsing notes stands, and it offered no way to the page that reads them as a feed. The other two doors — the About panel and the docs page — are not where a reader who just filtered a timeline is looking. check-in: 16e8738cb6 user: dmestas tags: trunk

A landing is a forum thread — that is what a stock fossil clone reads — but it is a decision record with its own surface. Off the listing and out of the count, which must agree. An existing thread URL redirects to the record rather than 404ing a link that used to work. check-in: 302ca7fae6 user: dmestas tags: trunk

Fossil registers one timeline event per technote; quarry registered one per artifact. libfossil has 50 technote artifacts under 26 event ids, so its timeline showed a superseded title five rows above its own correction. The collapse is in the event trigger — insert, then evict every loser for the event id — which is order-independent, as the fold requires: a clone delivers artifacts in wire order.

Notes are no longer a tab. Fossil's built-in menu has none, and libfossil, which uses them as a milestone feed, overrode mainmenu to add TODOs and FIXMEs without adding one. They are reached from the About panel, the docs page and the timeline filter instead, and the listing is a feed with excerpts rather than a list of links.

Every edit stays readable, as a wiki page's revisions are.

Fixes an authoring bug found on the way: zitcli never passed --time, so an edit re-filed the note at the moment it was edited and it moved in the timeline. The filing time is now a field on both forms. closed check-in: ac80214995 user: dmestas tags: notes

Two names whose hues land 7 degrees apart painted one block. Near collisions are a birthday bound, not bad luck, so the bar resolves what is in the thread — stepping a colliding voice to the identicon's other lightness. The hue never moves: it is the identity. check-in: e1e9619d0d user: dmestas tags: trunk

An HTML response carried no cache directive and no validator, which does not mean do not cache — it lets the browser choose a freshness lifetime. Safari chose one, and served a timeline rendered before a deploy while the server answered with another; the reader sees stale state with no way to tell.

no-cache rather than no-store: the copy is still kept and reused after revalidation, so the back button stays instant. Only reuse without asking is wrong, and every one of these pages is a view of moving state. check-in: ece3b20f3d user: dmestas tags: trunk

A URL with credentials (http://user:pw@host/) leaves them on document.baseURI even though location is cleaned, and fetch() refuses any URL carrying credentials — so every enhancement failed at once: diff toggle, unread poll, diff sentinel. A server-stated <base> fixes all of them without script, since the server knows the URL from its own routing and never from userinfo.

The toggle also swallowed its own fallback: replaceState to a credential-free URL throws on such a page, and it ran before @get while __prevent had already cancelled the link — neither swap nor navigation. Cosmetics now cannot fail the click. check-in: c35d8c4b5b user: dmestas tags: trunk

Fossil routes every surface through one helper (hyperlink_to_user); quarry had thirty-two places rendering a name and none of them clickable. One 'who' helper now, applied to the twenty that name an author — the account button and participant stacks stay decorative.

timeline.list gains user=, paired with y=all: what someone did includes their wiki edits and ticket changes, not just check-ins. check-in: a18be6d78d user: dmestas tags: trunk

The ring is stroked 2.2, so it paints to r+1.1; a 4.8 disc beside it read a fifth smaller. Internals are the ring's outer edge, filled. check-in: e1f3ed4bac user: dmestas tags: trunk

Mark then name, in the page's colour, orange on hover. The sigil is syntax and the mark already says this is a person; a name that is orange everywhere stops meaning clickable anywhere. check-in: dda6810776 user: dmestas tags: trunk

An @mention now renders the same glyph as the person's byline, through the resolver seam so markup still knows nothing about accounts. The identicon moved to internal/idn — web and api each hashed the name separately and agreed by luck. Blockquotes had no rule at all. check-in: 0ba22dea9e user: dmestas tags: trunk

f-post-h and f-post-b are where the padding lives; naming the wrapper alone gets a border round unpadded text. Check-ins use the same bordered list every other listing in the forge uses. check-in: 20373a1806 user: dmestas tags: trunk

The composer's textarea has no border of its own — it is built for the f-reply wrapper the ticket page and forum use, and without it the tabs and text floated unframed. The diff is a details now: this page is where a decision is made, and 600 lines between the proposal and the buttons means scrolling past the change to reach them. check-in: 59d7d8fa4c user: dmestas tags: trunk

Textually clean, but branchTipOf called Timeline with the old positional signature — ported to TimelineFilter. check-in: dae2180a89 user: dmestas tags: trunk

Fossil lets an author correct a check-in comment after the fact; quarry showed the superseded C card as if it were current, on 697 fossil check-ins. The correction is now applied at read, batched in AttachCI, with the original kept beside it — the stored card is an artifact fact and stays untouched. hidden and closed surface too.

Branch diff is the branch against its fork point, which is not the landing page's base (the target's current tip). Offered only where a fork point exists.

The manifest view renders the eight cards that are not the file list and counts the other 980, linking the tree.

Also: unified/side-by-side now swaps in place via datastar, keeping the window and correcting the URL; the wiki note panel lets blocks use the full width while prose keeps its measure. check-in: ff7f5d3d69 user: dmestas tags: trunk

docs/landings.md covers the whole loop and says plainly that a landing is a forum thread plus branch tags. Writing it found the gap: there was no create control at all. A mirror offers its upstream instead. check-in: c6aa693938 user: dmestas tags: trunk

open, comment, reject, accept through the forge against a real store — accept merges to trunk via zit#131's bare-store path. Also checks the record still reads as an ordinary forum thread. check-in: 9a84a5d962 user: dmestas tags: trunk

The probe is a full-store scan by design and measures 2.0s here, paid on every view. Keyed on the two check-ins, so a push invalidates it by making a different key rather than by anyone clearing a cache. 1.9s cold, 50ms warm. check-in: 21d510a5bd user: dmestas tags: trunk

A proposal is changes with reasons attached; a diff keeps the changes and throws the reasons away. zit#133's display half, from our own index. check-in: 120277efc7 user: dmestas tags: trunk

zit#131 landed the probe and exit 5. A conflicted landing renders as blocked with its paths named and accept disabled; a refused accept is a state to show, not a fault to report. Unknown is never treated as clean. check-in: e0f59b6abf user: dmestas tags: trunk

Stats are tallied at parse time so a windowed diff reports the whole file, and sit on the file header as well as the summary line. Tags come from checkin_tag, newest word per name winning. Copy falls back through execCommand to selection: navigator.clipboard is undefined on the plain-HTTP origins quarry is usually served from. check-in: ac056aaa1d user: dmestas tags: trunk

The timeline already had them; the ticket list did not — its row uses .Created and my earlier pass patched a different template that uses .Opened. Also added to ticket timeline entries and the forum queue.

The note is no longer folded when long. It was folded past ~1200 bytes because fossil's auto-toc note runs three screens and pushed a timeline below the fold, but a note written about this artifact is the context you came for. The summary stays, so it can still be collapsed; the threshold and the NoteLong field it fed are gone rather than left dead. closed check-in: eb1f126e04 user: dmestas tags: idn-open

Fossil's Context panel: the check-in among its neighbours in time, which is the question somebody bisecting a regression is asking. Two bounded index walks, no store read.

Needed a new index — every index on checkin led with branch, so ordering by date alone was a full scan of 20,928 rows plus a sort, twice a page. checkin_by_date puts the panel back at ~10ms.

Nodes reuse the timeline's palette so a branch is one colour across the forge. One lane deliberately: three neighbours are a slice of time, not a graph. On a phone the row stacks and the drawn rail becomes a border, since an SVG node cannot stretch without distorting its circle.

Identicons now appear wherever a person is named — check-in metadata, context rows, timeline, technotes, chat, attachments, wiki history. closed check-in: f73b5b9f05 user: dmestas tags: ci-context

A list of what awaits a decision and a page per record: proposal, diff against target computed at view time, discussion, and the controls. accept rides `i` because it merges; everything else is a forum cap. Transitions render as events — zit stores them as replies, which is encoding the interface exists to hide. check-in: 7d2a820f65 user: dmestas tags: trunk

Comment links resolved only hashes and http URLs, so fossil's own Markdown and limitations rendered as raw markup. They now go through the same route table the wiki renderer uses, and an unmapped one degrades to its label instead of showing brackets.

About moves below the comment and metadata — fossil's order. Above the title it was an empty bar floating over the page name.

checkin.diff returned the raw patch on every call and nothing read it: ~78KB of duplicate serialised and parsed back per view. Now opt-in. Diffs are also memoised per immutable pair, like listings already are. closed check-in: 7a2510835d user: dmestas tags: checkin-page

They were reachable only from a panel at the foot of the page — 36 rows down on a small wiki, a hundred on fossil. That is where you put something a reader already knows about, not where they discover it. Counts now sit in the filter row: wanted 1, unlinked 7. closed check-in: 6554752b15 user: dmestas tags: struct-links

A landing (zit#127) is a reading of forum_post, not a new artifact kind, so the table is derived and recomputed whenever any post in its thread folds -- which is what makes a reply indexed before its root converge on the same row. Backfilled once at open for indexes whose posts predate the table. check-in: 3c6bee847a user: dmestas tags: trunk

A merge whose two parents are both on trunk puts two chains in front of one column. Claiming rail 0 anyway stopped drawing the line that was already travelling there — the gap — and overwrote the commit it was heading to. Test asserts every head resumes its predecessor's tail. check-in: 6f34d948a0 user: dmestas tags: trunk

The edge resolved a CapSet and threw it away; it now rides the context, so an op can ask which letters the caller holds. Private tickets list, count and resolve for e, and for nobody else. One privacyWhere() so a fourth read path cannot silently ship without the clause. check-in: 9b2b188ece user: dmestas tags: trunk

The capability exists now (RFC 0004), so private_contact and contact are withheld by default and rendered to a holder of e. Default deny with no ungated-forge exemption: that rule would publish them. check-in: 86fadebbab user: dmestas tags: trunk

The new-ticket form rendered type/severity/priority/subsystem and the handler forwarded none of them. foundin was readable and writable nowhere. The sidebar editor listed six of nine fields. check-in: a83e03f2f3 user: dmestas tags: trunk

Three things the panel could not do. The lists were capped at twelve with "and 66 more" in plain text, which is a dead end wearing a summary's clothes — both now have a page, paged like everything else, reachable from the heading and from the count.

A wanted page named nobody. It links to the create form, which is the right action, but said nothing about WHO reached for the name — a to-do item with no brief. The index now records the asking pages and the page lists them.

Links counted link occurrences while its own doc and the tooltip both said pages; a page linking one absent name twice counted twice. Counts pages now.

Layout: the count sat 496px from its label in a 595px row, because space-between flung it to the far edge. closed check-in: af8678e35b user: dmestas tags: wiki-struct

One ticket changed nine times is nine rows here. "tickets 93,649" against 17,000 tickets had the right number and the wrong noun. check-in: bcdf35bcef user: dmestas tags: trunk

A chip counted its kind across the repository but paged a window of the whole timeline, so 234 tickets among thousands of check-ins gave ten pages with rows on almost none. Timeline takes a kind; event_by_kind covers it. check-in: 0238002099 user: dmestas tags: trunk

Every .idn rule was scoped to a parent — .avatar, .ft-meta, .f-post-h — so an identicon used anywhere new inherited no size, and an SVG with no size fills its container. The wiki history page rendered one person's mark at the width of the screen. A helper meant to be usable wherever a person is named has to be safe by default; the scoped rules are more specific and still win.

Also drops the centring from the doc grid. It indented the block away from the page title, which sits outside the grid: on a doc page the heading started at 392px and its first paragraph at 495px. closed check-in: 717dbf0f20 user: dmestas tags: idn-base

Sizing the grid to what each page happened to hold made the prose jump sideways every time a reader clicked from a page with headings to one without. The text should sit where it sat: the contents track is now always reserved, and an empty one costs nothing but space it was going to occupy anyway.

The left rail still follows :has — a doc page has no siblings to list, and reserving a gutter for nothing is a different mistake. closed check-in: 6d2a12980e user: dmestas tags: doc-reserve

Infinite scroll out, a real pager in: page numbers with both ends pinned, position, and every page a bookmarkable URL. The feed pages by offset now (a page is a position, not a place you arrived from) and says how far it reaches — 264K events is not a pager, it is an archive, and that lives on each repository's timeline. check-in: 98f77c8273 user: dmestas tags: trunk

fossil content links its documentation absolutely — the Cookbook has importing from Git — and the route mapping sent those to the file view. That was right when the file view was the only thing there was; since doc.get exists it has been putting readers in a source listing of a page quarry can render.

Leaves /uv/ dropped: unversioned content is a fossil feature neither quarry nor zit has, and that is why Download renders as plain text rather than as a link to a 404. closed check-in: d53f14d5b5 user: dmestas tags: doc-links-abs

Fossil's own docs link directories with a trailing slash — index.wiki carries a variety of techniques — and those all 404'd: the doc route found no renderable extension, redirected to the file view, and the file view cannot serve a directory. A real page was sitting behind the error the whole time (www/caps/index.md).

A directory now serves its index document, preferring the formats that render into the page, found with one listing rather than a probe per candidate. A directory with no index goes to the TREE view; only an actual file falls back to source. closed check-in: ff17aa2517 user: dmestas tags: doc-dirs

A fixed window is a dead end for anyone whose repositories are busy. Cursor is date~hash, so the merged stream cannot repeat or skip where two events share a timestamp. Sentinel

• loads the next page on scroll; noscript holds the same page as a link.