Brings in the A4 (millimetre) report designer alongside the existing
matrix editor, migrates the frontend from react-scripts to Vite/Vitest,
and fixes four defects found while using the A4 band tools.
Layout — layout.js now implements the standard banded-report model:
everything prints where it was drawn, and a band growing per data row is
the only thing that moves anything. Previously `paginate` re-flowed every
band from the top margin, so a band placed halfway down the page
previewed at the top, and `header` was treated as a page header and
hoisted above the statics drawn before it. A header is a COLUMN header:
it prints in place and repeats only on a continuation page. Statics now
take part too — one drawn below a band keeps its drawn gap as that band
grows, and follows onto the next page rather than vanishing when pushed
past the bottom margin. The displacement is keyed on the band's design
bottom, not on an instance's laid-out y: every row of a band shares one
design position, so keying on the instance measured the gap from row 1
and let the static overlap rows 2..n.
Band children — a child's x is relative to its band, as its y already
was. layout() emitted it raw, so a field printed measured from the paper
edge instead of the band's left margin. Instances now carry bandX.
Band placement — a band's x comes from the page margins, so creation
takes only the click's Y (the factory's x was being overwritten by the
spread order) and dragging is locked to the vertical axis. Bands render
as a strip, with top/bottom rules only, matching the matrix editor.
fromApiReport reflows on load so reports saved with a stray x straighten
themselves. Matrix bands are constrained to vertical too.
Drag lifecycle — useElementDrag re-subscribed its document listeners on
every render during a drag, because the effect depended on an unmemoized
onDrag. Each re-subscribe removed the mouseup listener before adding the
next, measured as add/remove/add/remove/add across two mousemoves, and a
release landing in that gap was lost: the element kept following the
cursor until the next click. Snapping made it routine, since a held
guide stops x/y changing and leaves the guide overlay's render sitting
between the move and the release. onDrag and dragData now go through
refs so the effect depends on [isDragging] alone.
Note: the matrix editor has no test coverage, so its band constraint is
verified by reading only. The A4 suite is 201 passing tests.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A selector is a nomenclature whose value decides which other nomenclatures
appear beneath it: picking "Ship" reveals the ship fields, "Warehouse"
reveals the warehouse ones. Previously every vehicle nomenclature rendered
at once, so the operator had to know which were relevant and a vehicle
could silently carry both a ship and a warehouse.
Visibility is one rule - a nomenclature is hidden if some selector option
reveals it and that selector's value is not the revealing option - kept in
nomenclatures/visibility.py and mirrored in services/nomenclatureVisibility.js.
A nomenclature no selector references always shows, so installations without
selectors are unchanged.
The serializer, not the UI, enforces the invariant: a value for a hidden
nomenclature is rejected, as is a selector value that matches no option.
The frontend clearing values on switch and pruning hidden keys before save
is convenience on top of that.
Selector values join lookups in the quick-fill combination projection, so
applying a saved combination sets the source and its revealed fields in one
click rather than collapsing "Ship + cargo" and "Warehouse + cargo" into
one indistinguishable row.
Configuration rejects nesting (a selector revealing another selector) and
double-reveal (one nomenclature revealed by two selectors, leaving its
visibility ambiguous). The admin UI filters both out of the reveal list as
well, so an invalid configuration cannot be built in the first place.
Note: introducing a selector over existing data hides values already
entered - vehicles predating the selector have no selector key, so anything
it reveals becomes invisible until a source is chosen. The values remain
stored. Worth a data check before configuring the first selector.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Four related backend features. They are committed together because they
interleave in api/views.py and api/serializers.py - one hunk holds the
combinations ViewSet, the licensing helpers and LicenseView - and
splitting by hunk would produce commits that do not import.
ExtraCombination records previously-used combinations of lookup
nomenclature values so an operator can refill the Additional Data fields
in one click. Deduplicated by a canonical order-independent hash, since
JSONField equality compares serialized text on SQLite and so depends on
key order.
Weighing replaces set_tare/set_gross with a single measure action that
assigns by weight - lighter becomes tare, heavier gross - because trucks
arrive both loaded and empty. A second, lighter reading moves the first
to gross carrying its own timestamp, user and manual fields, so
provenance follows the value rather than the slot. Manual entries record
who, when and why, gated on the previously unenforced can_measure and
can_manually_measure flags.
ScaleProtocol stores how to read each indicator, so a new scale type is
a configuration change rather than a new binary on every station PC.
Licensing is offline and node-locked: the server verifies an Ed25519
signature against a fingerprint of four hardware components, tolerating
one change, and counts concurrently connected stations. Sites are
air-gapped, so there is no activation server. A 14-day grace period
keeps an unlicensed or newly-invalid install working, because a licence
check must never stop a weighbridge without warning.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
gitea