fix(flows): move quote pricing and forecast accumulators to CEL value envelopes

[hotlong]

Fixes #1984

Description

Moves the six flow value expressions that objectstack validate --strict flags as {…} template dialect to CEL value envelopes: quote_generation → create_quote fields.discount_amount / fields.total_price, and the four forecast_snapshot bucket accumulators (add_pipeline / add_best_case / add_commit / add_won, all from sumBucket()). User-visible change: none.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• Code refactoring

Related Issues

Fixes #1984
Related to #1983 (left these six on purpose)

Changes Made

• Quote: round(double(oppRecord.amount) * (DISCOUNT / 100.0) * 100.0) / 100.0 and the 1.0 - … analogue. round() returns an int in CEL, so the divisor is the decimal 100.0 (/ 100 would integer-divide and drop the cents). DISCOUNT is one spliced fragment, (!has(vars.discount) || isBlank(vars.discount) ? 0.0 : double(vars.discount)): the template read a cleared discount (null / "") as 0, and a bare double(null) errors, which would have failed the quote. Authored with expression(src, 'cel') (precedent: event_attendee.object.ts) because the P tag JSON-quotes an interpolated string and cannot splice a fragment.
• Forecast: TOTAL + (!has(ITEM.amount) || isBlank(ITEM.amount) ? 0.0 : double(ITEM.amount)), again via expression() because the source splices the node's variable names. The guard keeps a null/absent amount at 0 instead of failing the owner's sweep; double() coerces a string amount.
• Comments beside both sites rewritten for the CEL form; the * 1 rationale above sumBucket and in the file header no longer describes the template path.
• Changeset: patch.

Testing

• New tests added: test/flow-quote.test.ts (new describe — 1,234.56 at 10% → 123.46 / 1,111.10; cleared discount null and "" → 0 / full price) and test/forecast-snapshot-amounts.test.ts (real AutomationEngine via flow-harness: a null amount sums as 0 and the stale row is still overwritten; decimal amounts keep their decimals).
• Existing pins untouched and green, including Generate Quote fails for most non-zero discounts: the flow writes raw IEEE-754 products into two 2-decimal money fields, and the rejection never reaches the user #1206's 30% / 70% of 180,000 and the string-amount forecast case.
• pnpm verify exit 0 at 151392a4 (174 files, 3728 passed, 1 skipped; hygiene:tokens clean on main's ceilings — src/revenue business semantics ~15,665 / 16,000, src/sales ~54,323 / 55,000).
• npx objectstack validate --strict: 14 warnings on base 39ba05e2 → 8 on this branch; the 6 removed are exactly the six sites, the 8 left are byte-identical to base (hierarchy-security capability, page:card description, six position-routed approval nodes).
• e2e not run (starts a server; flow behaviour is covered through the real engine above).

Ablations (each from the committed state, restored with git checkout HEAD -- PATH, restoration proved by HEAD blob hash equality and empty git diff HEAD):

mutation	on-disk proof	result
quote divisors / 100.0 → / 100	decimal count 2 → 0, int count 0 → 2	keeps the cents red (1 failed / 5 passed)
forecast guard → bare double(ITEM.amount)	guard count 1 → 0	sums a null amount as 0 red: expected 9999999 to be 30000
DISCOUNT → bare double(vars.discount)	guard count 1 → 0	both cleared-discount cases red (2 failed / 4 passed)

Behaviour parity probe (same harness, template on base vs CEL on branch, identical outputs): 1,234.56@10 → 123.46 / 1,111.1; discount null / "" → 0 / 180,000; discount "30" → 54,000 / 126,000; 99.99@33 → 33 / 66.99; forecast with null + 1,000.25 + 0.1 + 0.2 → 1,000.5500000000001 in both dialects, and a null-only owner → 0.

Checklist

• I have added a changeset (.changeset/flow-values-cel.md, patch)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• My changes generate no new warnings (strict: 14 → 8)
• I have added tests that prove my fix is effective
• New and existing unit tests pass locally with my changes

Additional Notes

Acceptance notes

• Beyond the issue table: the quote's discount guard. The issue named the null-amount hazard for the forecast only; measuring the template baseline showed a cleared quote discount (null / "") also priced as 0%, and an unguarded double(discount) errors on it. Guarded so Acceptance §2 (pricing unchanged) holds; pinned by a test.
• oppRecord.amount is not guarded: crm_opportunity.amount is required + notNull, so a null is unreachable, and a missing oppRecord already fails create_quote on the required crm_account (comment on edges e4a/e4b).
• An ABSENT amount key (sparse driver) is not reachable through flow-harness, which materialises declared columns as null. Measured directly against ExpressionEngine (17.6.0) with the celScope shape: the guard returns 0 for null, absent and "".
• The accumulator's initial value from reset_totals (0) adds to a CEL double without an overload error — measured through the real engine by the tests above.
• test/flow-scheduled.test.ts keeps a fixture comment saying the string-amount case is caught by "the * 1 coercion in the accumulator". Left untouched because Acceptance §3 asks for the existing forecast tests to pass untouched; its assertion still holds (CEL double() coerces). A one-line wording follow-up if wanted. Carrier: none.

---

Generated by Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated
hotcrm	Ignored		Oct 2, 2026 9:22pm UTC