[marcin-kordas-hoc]

Context

The PR stops a function or operator from claiming an error it only read, and gives #SPILL! errors an address. It is part 5 of 7 of HF-131, stacked on #1764.

• Why: an error that exists before any function sees it, such as a broken reference, reached the enclosing call without an origin, so that call claimed it. =SUM(A2:A99999999999) reported SUM. #SPILL! errors had no address.
• What changed:
  • The parser, the interpreter, DependencyGraph.ts, the dependency transformers and NumericAggregationPlugin.ts set the origin where they build the error: reference, removed reference, literal or OFFSET.
  • ParsingErrorVertex.ts and CellContentParser.ts: parser and user input.
  • Interpreter.ts: an array literal whose rows differ in length reports array literal. An error inside an array result names the operation that built the array.
  • FormulaVertex.ts: setNoSpace() makes the formula vertex the root of its #SPILL! error.
  • JSDoc of originFunction and argumentIndex, and docs/guide/known-limitations.md: the full list of origin names.
• Result:

Formula	Before (#1764)	After
=SUM(A2:A99999999999)	SUM	reference
=DATE(2020,A99999999999,1)	DATE, argumentIndex: 1	reference, no index
=SUM(Sheet1!A2:Sheet2!B2)	SUM	reference
=SUM(A2) after row 2 is removed	SUM	removed reference
=SUM({1,2;3})	SUM	array literal
=INDEX(A1:A2+B1:B3,3,1), array arithmetic on	INDEX	addWithEpsilon
=1+	none	parser
#REF! typed into a cell	none	user input
=#REF!	none	literal
=SEQUENCE(2) blocked by a value below	address: none	address: Sheet1!A1

How did you test your changes?

• Paired tests: handsontable/hyperformula-tests#63 on fix/hf-131-rootless-origin, in unit/error-origin-function.spec.ts and unit/error-address-preservation.spec.ts.
• At the previous head, these two specs passed against this PR's code (45 tests); 15 of them failed against HF-131 (4/7): expose which function produced an error, and which argument #1764's head.
• The two new cases for a range that SUM or MAX reads (several sheets, a missing sheet) fail without the NumericAggregationPlugin.ts change, reporting SUM.
• CI is green on this head, except the repository-wide FOSSA license check.

Compatibility with Excel and Google Sheets

Not applicable: error types and displayed values do not change.

Public API changes

• New values of DetailedCellError.originFunction: reference, removed reference, parser, user input, literal, array literal.
• DetailedCellError.address of a #SPILL! error is now the cell whose array could not spill, instead of undefined.

Out of scope and follow-ups

• #SPILL! and #CYCLE! keep originFunction undefined, because no function produced them. This is stated in known-limitations.md.
• The name removed reference is open for review: it follows the other lowercase names (parser, user input, literal).

Types of changes

• Breaking change (existing functionality no longer works as before) – agreed with the maintainers
• New feature or improvement (a non-breaking change that adds functionality)
• Bug fix (a non-breaking change that fixes an issue)
• Additional language file, or a change to an existing language file (translations)
• Change to the documentation

Related issues

1. Part of Improve the formula error messages #1547 (HF-131).

Checklist

• My pull request meets the Definition of Done. External contributors: see also Contributing to HyperFormula.
• I have signed the Contributor License Agreement.
• I read my own diff end to end before requesting a review.
• I added a CHANGELOG.md entry, written by the changelog rules – or none is needed (docs, tests or tooling only).
• I updated the documentation and JSDoc to match the change.
• Breaking change only: I added a section to the migration guide.
• I updated the dev-docs/ pages this change affects – or none needed it.

[qunabu]

Task linked: HF-131 Verbose and actionable formula error messages

[cloudflare-workers-and-pages]

Deploying with    Cloudflare Workers

The latest updates on your project. Learn more about integrating Git with Workers.

Status	Name	Latest Commit	Preview URL	Updated (UTC)
✅ Deployment successful! View logs	hyperformula-docs	ae97e21	Commit Preview URL Branch Preview URL	Oct 23 2026, 07:18 AM

[github-actions]

Performance comparison of head (ae97e21) vs base (e745be2)

                                     testName |    base |    head | change
--------------------------------------------------------------------------
                                      Sheet A |  486.32 |  491.26 | +1.02%
                                      Sheet B |  155.64 |  153.02 | -1.68%
                                      Sheet T |  140.13 |  138.64 | -1.06%
                                Column ranges |  464.97 |  466.52 | +0.33%
                                Sorted lookup | 13714.2 | 14803.9 | +7.95%
Sheet A:  change value, add/remove row/column |   14.63 |   15.33 | +4.78%
 Sheet B: change value, add/remove row/column |  124.81 |  125.58 | +0.62%
                   Column ranges - add column |  148.27 |   145.5 | -1.87%
                Column ranges - without batch |   449.4 |  456.22 | +1.52%
                        Column ranges - batch |  114.06 |  115.17 | +0.97%

[marcin-kordas-hoc]

Ran a closed-loop comparison of this stack's origin/address behavior against real Excel via
MS Graph (13 scenarios, full write-up + raw evidence in a private handoff repo — happy to share the
methodology if useful). Everything on the CRUD-address side of this PR checked out — a REF error
from a deleted column reports the same type as Excel and the address stays consistent before/after
the delete.

One thing surfaced that isn't a bug, but is worth a product call: the CRUD-produced #REF! from
Transformer.ts's column/row-removal branches never calls .withOrigin(...), so originFunction
comes back undefined for it — confirmed live, the field is genuinely absent, not empty. That's
consistent with how the mechanism works (an AstNodeType.ERROR node has no entry in
originNameForAstNode, so the stamp is a no-op), but neither this PR's own "deliberate limits"
nor known-limitations.md's existing bullet (which only covers the 6 parser reference-checking
branches) actually says whether this specific case is intentional.

Should a CRUD-removed reference get an originFunction too (e.g. something like 'row/column removed'), or is leaving it unattributed here the right call, same as the parser branches? Either
answer is fine by me — just flagging that it isn't currently written down either way.

[cursor]

Stale Bugbot comment from a previous run.

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 2a81d2f. Configure here.}

[Tobiadefami]

The implementation looks good.

[codecov]

Codecov Report

❌ Patch coverage is 97.87234% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 97.38%. Comparing base (e745be2) to head (ae97e21).

Files with missing lines	Patch %	Lines
hyperformula/src/interpreter/Interpreter.ts	96.00%	1 Missing ⚠️

Additional details and impacted files

@@                       Coverage Diff                       @@
##           feat/hf-131-origin-function    #1765      +/-   ##
===============================================================
+ Coverage                        97.36%   97.38%   +0.02%     
===============================================================
  Files                              207      207              
  Lines                            16327    16343      +16     
  Branches                          3624     3630       +6     
===============================================================
+ Hits                             15896    15916      +20     
+ Misses                             423      419       -4     
  Partials                             8        8              

Files with missing lines	Coverage Δ
hyperformula/src/CellContentParser.ts	100.00% <100.00%> (ø)
hyperformula/src/CellValue.ts	83.33% <ø> (ø)
...yperformula/src/DependencyGraph/DependencyGraph.ts	98.82% <100.00%> (ø)
hyperformula/src/DependencyGraph/FormulaVertex.ts	84.25% <100.00%> (ø)
...rformula/src/DependencyGraph/ParsingErrorVertex.ts	100.00% <100.00%> (ø)
...rformula/src/dependencyTransformers/Transformer.ts	98.70% <100.00%> (ø)
...src/interpreter/plugin/NumericAggregationPlugin.ts	97.23% <100.00%> (ø)
hyperformula/src/parser/FormulaParser.ts	98.09% <100.00%> (+0.47%)	⬆️
hyperformula/src/interpreter/Interpreter.ts	96.39% <96.00%> (+0.45%)	⬆️

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.