* fix(period): let klarmarkera close a migrated year whose native vouchers are balance-sheet only
markPeriodClosedExternally decided "migrated" by asking whether the period
had any non-imported verifikat at all. That was a proxy for the thing the
guard protects (a bokslutsverifikat transferring 3xxx-8xxx, BFL 5-6 kap),
and it shut out the migrated first year whose only native voucher re-keys
the opening balance (1930 D / 2081 K aktiekapital) after a failed SIE
import. With the next year's IB already imported, the normal year-end
refused too (NEXT_PERIOD_HAS_IB), so the year had no closing path at all.
The guard now asks the ledger the real question: are there lines on result
accounts (BAS class 3-8) in the period? Id-only entry fetch plus per-chunk
head counts with early exit, no line fetch, no journal_entries!inner embed.
The refusal message names what was found and why the normal year-end is
the remedy.
The loaded årsredovisning view also gets the FyPicker in its header. The
no-period branch auto-jumps to the remembered scope (or the newest year)
before its own picker is ever seen, so a user whose scope pointed at the
historical year had no way to reach the current year's årsredovisning
except by changing the scope on some other page and coming back.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RZjtjw5xnC2SVaEr5HqG8Q
* fix(period): PR #2393 round 1: require next-year IB for balance-sheet-only klarmarkera, bound result range to class 3-8
Triage of the swedish-compliance review:
- Balance-sheet-only native years bypassing the bokslut: partly real. The
result transfer is not the only thing the normal year-end does; it also
posts the next year's IB. A natively bookkept year with only balance-
sheet vouchers (dormant AB, aktiekapital deposit only) could have been
klarmarkerad and its balances would never have reached the next year.
The leg now additionally requires the next period to carry IB already,
which is exactly the shape where the normal year-end refuses
(NEXT_PERIOD_HAS_IB) and klarmarkera is the only path. Without that IB
the normal year-end works and stays the remedy, and the message says so.
The suggested IB-correction tag on entries was not adopted: the ledger
already answers the question, a tag would be one more concept to know.
- Unbounded account range: real, cosmetic. Bounded with lt '9' so class 9
interna poster no longer count as result accounts. Storage format is the
4-digit BAS string every classifier in the codebase already relies on
(isBalanceSheetAccount reads charAt(0)); the text compare is as robust as
those.
Tests: next year with IB passes, next year without IB refused, no next
year refused, lt('account_number', '9') asserted.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RZjtjw5xnC2SVaEr5HqG8Q
---------
Co-authored-by: Jakob Wennberg <311770904+jakobwennberg-oss@users.noreply.github.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Jakob Wennberg
5d735097a1