paged.data — database publishing

The database-publishing plugin — a binding DSL, a sync engine, record-flow pagination, and batch automation over DuckDB-WASM, publishing datasets other plugins can consume.

In short paged.data turns data into pages — a binding-expression DSL, a resolution/sync engine, record-flow lowering with grouping and per-group footers, and end-to-end print automation, all over a vendored DuckDB-WASM. It also publishes governed datasets that other plugins (e.g. paged.sheet) consume.

What it does

A publishing-grammar binding DSL (53 registry-driven functions) and a binding/sync engine (Linked / Pinned / Overridden / Stale states).
Record flow / pagination: nested grouping, multi-level per-group footers, parent-path on spill.
Print automation: per-record / per-group / one-catalog batch plans, plus a CLI and a sandboxed Boa scripting path.
A data-provider contract: register as a provider and publish a record set without knowing its consumers; governed extracts with column metadata.
In development: InDesign Data Merge (planning, merging and export as a Data Merge template), a SQL query panel over DuckDB, local JSON, Parquet and XLSX sources, row diffs and sync review, and binding any object property to data (the universal binding round).

How it's built

A Cargo workspace (data-core, data-expr, data-sources, data-query, data-bind, data-lower, automation crates, data-js) plus a TS bundle, with DuckDB-WASM vendored behind an Arrow seam.

Capability state

● shipped & verified · ○ shipped, re-test pending · ◍ shipped, untested · ◐ partial · ✗ failing · · planned

Plugin · data 39 features · 100% green

Feature	Shell	Data
Print automation / batch generation (per-record / per-group / one catalog) `data.automation.batch`	·	●
Barcode / QR symbologies — clean-room EAN-13/UPC-A, Code-128, QR encoders + binding + VECTOR lowering (§9.7) `data.barcode.symbology`		●
Bindings authoring — first-class add-binding flow in the bindings panel `data.bind.authoring`		●
Change-highlighting on refresh — per-binding "what changed since last sync" report (§8) `data.bind.change-report`		●
Binding + synchronization engine — resolution graph, sync states, identity diff `data.bind.engine`		●
Field-mapping wizard — map source columns → variable bindings one-click (§9) `data.bind.field-mapping`		●
Record-preview stepper — resolve a binding against a chosen record index N (§9) `data.bind.preview-step`		●
Universal property binding — any host.objects property bound to data, one undo step per apply (ADR 558) `data.bind.property`		●
Row diff since the document was last written — added, removed and changed rows, and the bindings they reach `data.bind.row-diff`		●
Sync states per binding, with pin, unpin and accept-source `data.bind.sync-review`		●
Variable bindings are custom InDesign text variables; one Set re-bakes every instance (ADR 559, protocol 71) `data.bind.text-variables`		●
Illustrator-compatible variable-library XML — import + export (§16.4) `data.dataset.library-xml`		●
Data sets — capture / switch / delete, applied as ONE undo step (§16.4) `data.dataset.palette`		●
Export as InDesign Data Merge template (native placeholders + UTF-16 data source) `data.export.datamerge`		●
Binding-expression DSL — lexer, parser, evaluator and function library, registry-driven dispatch `data.expr.engine`		●
Governed-extract source (governed tables + optional metadata sidecar -> catalog) `data.governed.extract`		◐
Locale-aware display formatting (en/de v1) `data.i18n.display`		●
Locales as a data table (CLDR values), per-field locale and display patterns `data.i18n.locale-table`		●
Placeholder lowering — variable replacement + single-region dynamic table `data.lower.content`		◐
Data Merge planner — Single and Multiple Records laid out by InDesign's rule `data.lower.merge-plan`		●
Data Merge writer — merge the records into the document, matching InDesign `data.lower.merge-writer`		◐
Record flow lowers into text frames on pages added as the flow needs them `data.lower.recordflow-writer`		◐
Re-lowering a table, barcode, record flow or merge updates its content in place `data.lower.relower-in-place`		◐
v43 lowering consumers — in-text variables (D-01), image place (D-14), live record-flow (D-12), rules (D-13) `data.lower.v43-consumers`		◐
paged.data's own objects in host.objects — sources, queries, bindings, data sets, variables and typed commands `data.object-model`		●
Performance gates (§12.5) — engine-tier perf assertions `data.perf.gates`		◐
Bindings live in the document's labels; the session part is a cache; restore from an InDesign re-save (ADR 559) `data.persist.labels`		●
paged.data bundle (media.paged.data) — manifest + activate + 3 panels + DuckDB query glue `data.plugin.bundle`		◐
The data session is saved with the document and restored on open `data.plugin.persistence`		●
Register as an SDK data provider (publish RecordSet to other consumers, e.g. sheets) `data.provider.contract`		●
Data query panel — SQL with DuckDB diagnostics, filter/sort/group builders, preview grid `data.query.panel`		◐
Query engine (DuckDB-WASM) + RecordSet shape/order/hash — the Arrow seam `data.query.seam`		◐
Rule editor — checked condition, firing preview, document style, per-record paragraphs `data.rule.authoring`		●
Capability-gated data access + threat model (the suite's largest surface) `data.security.gates`		◐
Source adapters (inline + file at M0) + capability/consent gate + visible manifest `data.source.adapters`		◐
D-11 DB attach (SQLite/Postgres/MySQL) + DuckDB-WASM first-class engine + refresh-policy + remote `data.source.db-attach`		◐
Local import of JSON, NDJSON, Parquet and XLSX into DuckDB (data-xlsx for workbooks) `data.source.local-formats`		◐
Refresh policies per source — manual, on open, interval (remote, consented), never `data.source.refresh-policy`		◐
Illustrator "Variables" — the four traits, mapped onto the shipped binding model (§16.4) `data.variable.traits`		◐

paged.data — database publishing

What it does

How it's built

Capability state

Plugin · data 39 features · 100% green

Recent activity

On this page