0.8.1

Changed

• Code lens titles are computed on demand, per visible lens, instead of for every
  record on every edit.

0.8.0

Added

• gedcom.virtualIndent.enabled indents each line by its level, as a decoration
  that never touches the file ([#2]). gedcom.virtualIndent.width sets the
  columns per level.
• Dates are validated: a month that is not a month, or a day the month cannot
  have. Every calendar, both spellings of the calendar escape.

0.7.1

Fixed

• Show Tree did nothing at all when no GEDCOM file was in front of you ([#5]).
  The view is contributed behind editorLangId == gedcom, so there was no view
  to focus and .focus resolved silently. It now says so, and where the view
  exists but has not come up yet — activation, the context key and the click all
  racing on a cold editor — the container is opened directly rather than the
  result being assumed. An integration test covers the cold path, which every
  other test warms past.

Changed

• The README lists every setting, with its default and what it does.
• glob is pinned through overrides to 13. Mocha pulled 10, which npm warns is
  unsupported; nothing else in the tree was on it, and no advisory was open.
• @playwright/browser-chromium is approved under npm's install-script allowlist,
  so npm install no longer stops to ask about it.

0.7.0

Added

• Notes holding markup can be rendered instead of read as characters, with a
  Text / HTML switch. gedcom.details.noteFormat sets the starting choice,
  default text; the switch lasts the session. Rendering copies the parsed tree
  through an allowlist of formatting tags — the only attribute kept anywhere is
  href, on http(s) only.
• fixtures/notes/ — HTML in notes, split mid-URL by CONC in 5.5.1 and
  declared with MIME in 7.0. Both carry well-formed, unclosed and crossed
  markup, since no rule in either specification inspects any of it.

Fixed

• The toolbar button did not open the tree panel (#5). The editor title bar
  passes the resource and its own {groupId, editorIndex} context; read as the
  code lens's (uri, line), that revealed the editor at a nonsense position and
  never opened the panel.
• An inline note was reported as a broken pointer (#2). 5.5.1 defines
  NOTE, SOUR, OBJE and REPO as either a pointer or the content in place,
  and the registry models only the pointer form — so most exporters' notes were
  called malformed.

Changed

• "Show Graph" is now Show Tree, the panel is Tree, and the code lens says
  see in the tree. Setting keys and command ids are unchanged.
• Records are counted in English: 3,010 individuals · 1,422 families, not
  3,010 INDI · 1,422 FAM. Used by the lens, the header hover and the details
  panel alike.
• Dates are said in words, in every calendar GEDCOM defines. ABT 3 NOV 1901
  reads "About 3 November 1901"; a French Republican or Hebrew date keeps its own
  month names and is labelled with its calendar. Gregorian carries no label.
• The lens above a record leads with the name, then the sex in English and the
  years: Victoria Hanover · Female · 1819–1901 · 2 parents · 1 spouse.
• Inlay hints read as captions: Died age 73, Male.

Testing

• The integration tests assert the panel opens and follows the cursor,
  not merely that a command did not throw. They also fail on any unhandled
  rejection in the extension host, and run against Insiders as an advisory CI
  job — where #5 came from, and the one combination nothing covered.

0.6.0

Two themes. A GEDCOM file is allowed to indent itself, and files in the wild do —
so those are now read, highlighted and folded like any other, whatever they indent
with. And the panels stop showing the reader the file's own shorthand: slashes
around a surname, F for a sex, a URL that cannot be clicked, a hover that
answers "Text."

Added

• Indented files are read as first-class. The specification puts the level
  number first on the line with nothing before it, but plenty of exporters
  indent anyway to show the hierarchy, and such a file is now lexed,
  highlighted, folded and validated exactly as a flat one is. detectIndentation
  reports the habit — spaces, tabs or both, the width per level, and any line
  that departs from it.

  A tab's width is the only thing a file cannot state. Where tabs and spaces are
  mixed it is solved by search over the widths anything has ever defaulted to,
  because the direct calculation needs a space-only line at the same level to
  measure against and a file whose first levels are tabs never provides one.

• Seven fixtures/style/ files carrying the same family tree at one, two and
  four spaces per level, at tabs, and at tabs-then-spaces with the tab standing
  for two, four and eight columns. A test asserts all seven parse to the same
  tree, which is the load-bearing claim: the level number states the hierarchy
  and the whitespace states nothing.

• A sentence about what each tag is for, in hovers. The registry we vendor is
  a machine's view of the format — tags, payload types, cardinalities, labels —
  and carries no prose, so a hover over OCCU answered with "Text.", which the
  reader could see for themselves.
  Roughly ninety structures now carry a written gloss, and the payload line is
  dropped where the gloss has already said what belongs there. Across the fixture
  corpus this takes the hovers that said nothing from several hundred to none.

• Links and image previews in the details panel. A URL anywhere in a field —
  a media object's FILE, a WWW, an address written into a note — is now a
  link that opens in the browser, and media the file identifies as a picture is
  shown as a thumbnail beneath its row.

  http and https only. A FILE payload is free text from a document the
  reader may merely have been sent, and javascript:, file: and vscode: URIs
  would otherwise be handed to the machine by a click; the scheme is checked in
  the extension host rather than trusted from the panel.

  Previews are on by default and gedcom.details.imagePreviews turns them off.
  Off rewrites the panel's content security policy rather than merely suppressing
  the <img>, so the panel is then unable to make a request rather than
  disinclined to — the point of the setting is that a photograph hosted by a
  genealogy site does not get to learn that the file is being read, and from
  where. Even on, only https is permitted.

Fixed

• A media object written inline was reported as a broken pointer. GEDCOM
  5.5.1 lets OBJE, among others, be written either as a pointer or with the
  record's substructures in place of the payload, and the second form is common
  in real files. Only the payload-less form is exempt: 1 ASSO @I1@ df is still
  wrong however many substructures follow it.

• A date's time of day was dropped from the details panel. TIME hangs under
  DATE rather than beside it, in 5.5.1 and 7.0 alike, and the panel read only
  the date — so a change record, whose entire purpose is to say when, showed
  14 FEB 1998 and threw away 09:22:41. The parse tree was right throughout;
  only the reading of it was wrong.

Changed

• A name is read rather than copied. The slashes in /Family/ Personal are
  how GEDCOM marks the surname, not punctuation, and the details panel was
  printing them. It now shows the name without them, in the order the file wrote
  — that order is itself information, since the format exists to carry names from
  cultures that write the surname first — and lists the parts beneath, from the
  GIVN and SURN substructures where the file states them and from its own
  reading of the string where it does not.

  Where a person holds several names, the TYPE beneath each becomes its label,
  so two rows are no longer both called "Name".

• A structure carrying nothing says so. 1 _MAYBE with no payload and nothing
  beneath it used to render as "recorded", a word the file never said. It is now
  marked as having no value, in italic and dimmed, so that a tag someone wrote and
  left blank is visible as exactly that.

• Coded values are shown in English. The details panel expands an enumerated
  payload to its meaning — Sex: Female, not Sex: F — for every enumeration
  the extension models, not only SEX. Inlay hints capitalise the same labels,
  since a hint is a caption beside the line rather than a clause continuing it.

• The details panel names a media object by its own title rather than by its tag,
  and says what kind of file it is — read from the FORM where the file writes
  one and from the path's extension where it does not. A residence with an
  address but no place now shows the address instead of the word "recorded".

0.5.2

Tooling and documentation only — nothing in the extension itself changed.

0.5.1 was tagged and never shipped: the manifest was left at 0.5.0, so the
release workflow's version guard refused it, which is precisely what that guard
is for.

Fixed

• The release workflow could not be re-run. It skips the Marketplace when
  VSCE_PAT is absent and tells you to add the secret and run it again — but
  gh release create fails outright once the release exists, so the advice was
  impossible to follow. An existing release is now updated in place.

• dist/preview/ no longer ships in the VSIX. It only exists once somebody has
  run vp run preview, so the extension gained 40 kB of developer artifact on
  some machines and not others.

• The README badges were dead. shields.io has retired its entire
  visual-studio-marketplace family — version, installs, downloads and rating
  all answer "retired badge". They are replaced by a static Marketplace link
  plus release, build and licence, none of which can rot.

  The obvious substitute, vsmarketplacebadges.dev, works and was rejected: the
  Marketplace renders images only from an allowlist of hosts, so those badges
  would have looked right on GitHub and broken on the listing itself. Losing
  version, installs and rating costs nothing there, since the Marketplace page
  already shows all three.

Changed

• Developer guidance moved from README.md to CONTRIBUTING.md. The README is
  the Marketplace listing, so build instructions and package layout were being
  shown to every prospective user.

Testing

• GitHub's rendering is now checked against GitHub's rendering. The Primer
  panel in vp run preview was a palette written from memory, which made the
  central claim of the colour design — that the semantic classes stay distinct
  where the palette is narrowest — one nobody could verify. It now runs the
  committed grammar through starry-night,
  the open reimplementation of GitHub's highlighter, coloured from the tokens
  @primer/primitives publishes.

  It found two things the approximation was hiding. markup.quote and
  entity.name.tag both resolve to pl-ent, so a citation and a name are the
  same colour on github.com; and variable.other resolves to pl-smi, whose
  colour is Primer's own foreground, so linkage tags are indistinguishable
  from ordinary text there. Six semantic classes come out as four colours in
  light and five in dark, not six. All of it is recorded in
  packages/grammar/test/prettylights.test.ts rather than assumed away.

  It also confirms the new payload-shape rules arrive as pl-ii, the class
  GitHub paints as invalid, rather than as ordinary text.

0.5.0

The theme of this release is that a GEDCOM file is mostly opaque identifiers and
undocumented codes, and every feature below removes a reason to leave the line
you are reading.

The graph panel now draws the family, not the file

GEDCOM stores a marriage as a FAM record that both spouses and every child
point at. Drawn literally that put a nameless join record between every pair of
relatives, made a grandparent four hops from a grandchild, and filled the third
column with boxes that answered no question anyone had asked.

• Families are collapsed. Nodes are people; edges are the relationships
  between them, derived by joining through the FAM records and then discarding
  them. A spouse and a child are now one hop away, not two.

• Columns are generations, not hop counts. Laid out by hops, a sibling and a
  grandparent shared a column because both are two steps from the focus — two
  generations side by side, saying something false about the family. Ancestors
  now run left of the focus and descendants right, the direction a family tree
  is read in.

• Couples sit together, joined by a marriage bar, and every child descends
  from one point on it. Two parents fanning independently to four children
  makes crossings unavoidable; one line per child from the marriage makes them
  impossible.

• Sibling groups hang beneath their parents, and siblings run oldest first
  within the group. Nothing in the ordering consults the current view, which
  is what stops a column rearranging itself every time the selection moves —
  a reader clicking along a row of relatives was watching the row reshuffle
  under them.

  This replaced a barycentre pass that ordered each column against its
  neighbours. That reads well on any one drawing and produces a different
  drawing every time, because the arrangement is computed from whoever happens
  to be on screen. Ordering by birth alone held still but tangled. Neither trade
  was necessary: hanging each family beneath its parents makes the two columns
  agree by construction rather than by search.

  Measured over 300 neighbourhoods of Linguist's Royal92.ged: 68% are drawn
  with no crossings at all, 87% with two or fewer, and the ninetieth percentile
  is three — against 54%, and a ninetieth percentile of fourteen, before any
  of this. Tests hold both the crossing figures and the stability ones.

• Marriages are shown, labelled with their year, alongside parent and child
  links. Siblings are drawn only where a family records no parents, since
  otherwise they are already two hops apart through one.

• Boxes carry dates — 1901–1975, or b. 1930 where only one is known.
  A tree full of people sharing a name is unreadable without them. Where the
  file records none the box shows the name alone: "Individual" under every name
  is a label with no information in it, and a row of them reads as though
  something failed to load.

• A family is still shown as itself when it is the record under the cursor,
  with its members around it and their roles named. Collapsing is right for
  families travelled through and wrong for the one being looked at.

• Ancestors / Descendants / Both buttons, because tracing a line back is a
  different task from following it forward, and each is half the graph.

• Sources, notes and media are left out unless gedcom.graph.includeReferences
  asks for them: a well-sourced person cites dozens, and they crowd out the
  family the panel exists to show.

• A family is never drawn as a box, not even when it is the record under the
  cursor. It is a join, not a person: no name, no dates, nothing to say that its
  members do not say better. Putting the cursor in one shows that family — the
  couple and their children — and highlights all of them.

• Edges run left to right whichever way the pointer is written, and labels are
  placed with a backing plate, nudged apart where two would collide. The gutter
  between columns was widened so a label such as Married 1874 fits in it, and
  labels are drawn after the boxes: painted before them, a marriage year
  ended up half hidden behind the spouse below it.

• Relationships that cross no generation — siblings, citations, and a couple —
  are routed down the side of their column. Drawn as a left-to-right curve such
  an edge doubled back on itself and dropped its label behind a box.

The details panel

A second view in the same GEDCOM panel, for everything the graph has to
discard in order to stay readable.

• Selecting a person shows what they actually contain: occupation,
  residence, the place they were born, notes, citations, identifiers — composed
  generically from the registry, so a tag nobody anticipated appears rather than
  being dropped silently. Every field jumps to its line when clicked.
• With nothing selected it describes the file: the program that wrote it,
  the character set, the counts, and the submitter with their address and notes.
  None of that is a person or a family, and drawn into the graph the submitter
  became a box with no generation and no relationships. An unreferenced SUBM
  is found too — PAF-era files carry one that the header never points at, and
  Linguist's own Royal92.ged is one of them.
• Clicking a box in the graph now selects rather than navigates. Reading down
  a line of descent means looking at a dozen people in turn, and jumping the
  editor to each one loses the reader's place in the file for no benefit.
  Navigation is its own gesture, on a button that appears on the box.
• Text written across CONT lines is shown whole — monospace, wrapped, in a
  block of its own — whether the record is reached as the file's submitter or
  selected in its own right. Royal92.ged records the file's provenance as a
  twenty-eight line posting from 1992, and only its first line was reaching the
  panel; its three-line address fared no better. The parser's reassembly was
  never at fault, and a test now proves it: this layer was discarding the rest.
• A short list of vendor tags the registry has never covered are named in
  English — COMM is PAF's comment field, and it is what carries that
  provenance note. The list stays short deliberately: a confidently wrong label
  is worse than a bare tag.
• A record that is neither a person nor a family no longer appears in the graph
  as a lone box with no generation and no relationships. The panel says where to
  find its contents instead.
• The graph scrolls the selection into view when the neighbourhood is larger
  than the panel, on whichever axis actually overflows.
• Children descend from the parent whose line is being traced rather than from
  the midpoint between the couple. The midpoint is nobody — it floats in the gap
  between two boxes, and a reader following a descent cannot tell which of the
  two it belongs to.

Added

• Inlay hints, off the back of that: what each pointer names
  (1 FAMS @F1@ → John Smith + Jane Doe), what each coded value means
  (1 SEX M → male, 2 QUAY 3 → primary), a language tag's language, and how
  old the subject was at an event, computed against their own BIRT date. Each
  kind has its own setting, because the resolved names are indispensable in an
  unfamiliar file and noise in your own. An age carries the verb of its event —
  died age 73, married age 24 — and every hint is set apart from the payload
  so it does not read as though the file itself said 1 SEX M male.
• Code lens above each record: its shape in the tree, a clickable reference
  count that peeks every pointer to it, and a link into the graph panel. Above
  HEAD, the dataset summary the header does not carry — record counts and the
  span of years the file covers.
• Document links on WWW, EMAIL and URL-valued FILE payloads.
• Per-verb hovers. The rule they are written against is that a hover must
  answer the question the line provokes rather than restate the tag:
  • DATE — the weekday for an exact date, what a qualified date is claiming,
    and how old the subject was; under CHAN or CREA, how long ago instead,
    because whether a record is maintained is the actual question there.
  • AGE — the notation in words, cross-checked against the recorded dates.
    An age that disagrees with the file's own birth and event dates is flagged.
    Nothing else in the format checks the two against each other.
  • PLAC — the jurisdictions labelled from HEAD.PLAC.FORM, which almost no
    tool surfaces, plus coordinates from a MAP and a link to a map service.
  • NAME — split into given name, surname and suffix on the slashes, with a
    note when no surname is marked.
  • NCHI — how many children are claimed against how many are recorded, which
    is the research question rather than a restatement.
  • Enumerated payloads across QUAY, PEDI, RESN, SEX, FAMC.STAT,
    NAME.TYPE, MEDI, ROLE, FAMC.ADOP and the LDS ordinance statuses. When
    a value is outside its set, the alternatives are listed.
  • LANG resolved through BCP 47, FORM through media types in both the modern
    and 5.5.1 spellings, and the identifier families (REFN, UID, EXID,
    RIN, AFN, RFN) distinguished from each other.
  • A migration note on any tag the file's version removed, carrying the
    replacement — ROMN says to use TRAN.
• Cardinality stated in words: "Required, exactly one" rather than {1:1}.
• The detected version, in the status bar. It governs how every line in the
  file is read and is frequently guessed rather than declared, so a reader who
  disagrees with the guess needs to see it before anything else makes sense. An
  inferred version is flagged with the warning background; the tooltip explains
  how it was arrived at and what the file contains.
• Payload shapes checked in the grammar, so a wrong value is coloured as
  wrong while it is being typed rather ...