Completes the AP source layer: GP was covered by the pm* modules, but the
CMS side had only cms.vend. These three are what
SQL/fiscal/analysis/open_ap_aging.db2.sql actually joins.
avtx 966,553 A/P voucher transaction events incremental on AVTSTM
vchr 792,348 vouchers entered, period to date full
open 538,544 open payables; FHCBAL is live full
AVTX is the artrn analog -- the per-event apply detail. It matters
because the AR aging in /opt/clean reconstructs balances from
application events rather than a single settle_date, so AP needs an
applications fact too, not just documents.
Grains verified exactly unique before drafting: AVTX
(avtco#,avtvh#,avttx#) 966,553/966,553; VCHR
(idcom#,idvch#,idbnk#,idfisy,idfisp) 792,348/792,348; OPEN
(fhcom#,fhvch#,fhpyy,fhpper) 538,544/538,544. All three reconcile
IN SYNC.
vchr and open stay full: no change signal, only business dates, and
FHCBAL is a live mutable balance.
Two things worth knowing for the next DB2 module:
- pipekit maps DB2 TIMESTMP to Postgres text, so a watermark resolver
doing COALESCE(MAX(ts), TIMESTAMP '...') fails on mismatched types.
Cast explicitly. cms.gtran was already the precedent.
- Watermarks resolve at engine step 2 but --create-dest builds the
table at step 3b, so a watermark reading its own dest table fails on
the first run. Pre-create the table or seed with a constant.
Note LGDAT.AP* is a dead end -- APVH has 14 rows, APVD has 22. CMS AP
does not live in AP-prefixed files.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Only pm00200 (vendor master) covered the AP side; every transactional
PM table was missing even though the parallel receivables (rm*) set was
fully built out. Add ten modules mirroring that AR structure:
pm00100 vendor class setup pm20100 open apply detail
pm00300 vendor address master pm30200 paid transaction history
pm00400 key master (doc index) pm30300 apply history
pm10100 open GL distributions pm30600 GL distribution history
pm20000 open transactions pm30700 tax history
All full reloads. The whole set is <=230k rows and loads in seconds over
the Postgres COPY path, so a watermark buys nothing — and a date-keyed
incremental carries the row-level purge drift documented for icstt.
pm20000 and pm30200 do have DEX_ROW_TS if that changes.
Column maps come from pipekit's own /api/introspect/columns over the
GPSERVER linked server. Descriptions are null throughout: GP populates
no MS_Description extended properties (CHG has zero), and the metadata
lives in the Dexterity dictionary rather than SQL Server.
The group is scheduled 0 1 * * * but left disabled, matching the
Accounts Receivable group.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
config/ had drifted from pipekit.db: group memberships and three
schedules existed only in the DB, cjobdr and cjobh had no sidecars at
all, and ab.{json,sql} was left over from a module no longer in the DB
(an apply would have recreated it). pipekit apply --dry-run now reports
that the database matches config.
Two real inconsistencies fixed in ffpdglr1.json: it was missing the
ffpdglr1_hot watermark added with the rollover hot set, so an apply from
files would have deleted the watermark while the committed SQL still
referenced {ffpdglr1_hot}; and its period resolver had been edited in the
UI without an export, so the file carried a stale format('%L', ...)
variant.
Purchasing members were all run_order 0, which happened to sort
poh, poh_deleted, poi alphabetically and satisfy the real constraints by
accident. Now explicit: poh 1, poi 2, poh_deleted 3 -- poi resolves its
watermark off cms.poh, and poh_deleted's first hook reads cms.poh.
Also describes eight GP modules that had no description (dest_description
is the half the sidecar format carries; the module-level description
lives only in the DB).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Add `pipekit export` / `pipekit apply` (pipekit/config_io.py) to serialise the
DB's config — drivers, connections, modules (+ columns, watermarks, hooks),
groups, schedules — to a git-trackable text tree under config/, and rehydrate
it. SQLite stays runtime state; definitions become diffable/reviewable/
revertible.
- config only: run_log/group_run/settings and per-run state columns excluded
- name-keyed refs (portable across databases); source_query in .sql sidecars;
columns as real JSON arrays for line-by-line diffs
- newline-normalised so CRLF-vs-LF is never a spurious change
- apply is create/update by name; child collections fully synced; top-level
deletes gated behind --prune; --dry-run prints the plan
- round-trip is identity (export -> apply --dry-run == nothing to do)
Commits the current config/ as the first baseline, capturing the freshly
populated columns_json for rm00101/rm00301/iv00101. Documented in SPEC.md.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
