shifathosn
/
gridcraft
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
gridcraft/docs/gaps.md

9.8 KiB
Raw Blame History

Gaps

Last reviewed: 2026-10-11 · Last updated: 2026-10-11 · Change: minor (§15: the ODD* bond functions landed) · Target: Microsoft Excel (Microsoft 365)

Every known shortfall against Excel, one entry each, ranked by user impact (what blocks someone using GridCraft for real work first). This is the work list: pick from the top, check the open PRs first (50 open on 2026-10-10; several already address entries here), and update the entry when you land work. Estimates are Opus 5.5 agent-hours. Area detail lives in the linked parity docs; the summary is in target-app-parity.md.

Kinds: F feature · U UI/UX · FF file format · H hardware · L localization · P performance · S stability.

Beta blockers

No alpha blockers: every core workflow passes the alpha gate.

1. XLSX output not proven in Excel or on real files (FF)

  • Missing: a run over a real-world XLSX corpus, and confirmation that Excel opens what we save with no repair prompt (PivotTables, charts, CF, validation, dynamic arrays, tables).
  • Evidence: round-trip tests cover our own files only (crates/xlsx/src/tests/); PivotTable round-trip noted as "unverified in real Excel" since 10-07; #64 (dynamic-array flag) shows the class of bug.
  • Impact: the main format; a repair prompt or silent loss destroys trust.
  • Estimate: 15–25 h, plus owner time in Excel. file-format-parity.md

2. Large workbooks don't open; recalc is single-threaded (P, H)

  • Missing: streaming read for ~100 MB XLSX; multi-threaded recalc; 100k-dependent edit under 100 ms (was ~0.4 s); whole-sheet operations bounded.
  • Evidence: #175; no threading in crates/calc; PR #152 (whole-sheet Fill/Remove Duplicates out of memory), PR #151 (current region past 10,000 rows).
  • Impact: finance and data users routinely have big workbooks.
  • Estimate: 25–40 h. target-app-parity.md, hardware-parity.md

3. Unmodelled XLSX content is dropped on save (FF)

  • Missing: pass-through of parts and extensions we don't model (form controls, slicers, timelines, external links, OLE objects, custom XML, query tables, data model, chart details).
  • Evidence: reader warns and skips (crates/xlsx/src/read.rs); external links "not supported".
  • Impact: open-edit-save silently simplifies colleagues' workbooks.
  • Estimate: 10–15 h. file-format-parity.md

4. Charts lose type and options on round trip; shallow chart model (F, FF)

  • Missing: kinds that survive XLSX (stock, combo, histogram, box, waterfall, funnel, treemap, sunburst), axis scaling, label options, trendlines, error bars, per-point formatting.
  • Evidence: Chart holds only title, legend position, two flags and axis titles (crates/model/src/features.rs); PRs #158, #161 open; charts can't be resized (#190). Smooth lines (#62) and Switch Row/Column (#116) landed.
  • Impact: charts are in most business workbooks.
  • Estimate: 25–35 h for beta (40–60 h full). chart-parity.md

5. Text in non-Latin scripts doesn't render or export (L)

  • Missing: CJK font fallback, complex shaping (Arabic, Devanagari, Thai), bidi, non-Latin PDF.
  • Evidence: system-font fallback for CJK, Thai and Hebrew landed in #38 (closed #32); #142 (Chinese text) awaits confirmation; #54 (symbols as boxes on Fedora); no shaping for Arabic or Devanagari; PDF writer uses 14 standard fonts only.
  • Impact: blocks most of the world's users even with an English UI.
  • Estimate: 12–20 h. localization-parity.md

6. Calculation correctness bugs (F)

  • Missing: 15-significant-digit equality; spill updates after structural edits and cleared blockers; 3-D span deletion; qualified names after sheet rename; iterative calculation.
  • Evidence: #57, #133, #139, #140, #134; Enable Iterative Calculation stored but unused; PRs #148, #155, #156, #157.
  • Impact: wrong numbers are worse than missing features.
  • Estimate: 10–15 h. function-parity.md

7. No legacy .xls; no password-protected workbooks (FF)

  • Evidence: #121. Since #17 (closed #11) both are recognised through a CFB sniff and refused with a clear message, but neither opens.
  • Impact: common in finance, government and anything older than 2007; encrypted files are routine in business.
  • Estimate: 28–42 h (.xls read 20–30, encryption 8–12). file-format-parity.md

8. PivotTables: no slicers, calculated items, PivotCharts; unverified in Excel (F)

  • Evidence: parity-checklist.md missing insert.slicer, insert.timeline, insert.pivotChart, table.insertSlicer; calculated items and GETPIVOTDATA missing (calculated fields done).
  • Impact: pivots are the analyst's main tool.
  • Estimate: 20–30 h for beta (25–40 h full).

9. Keyboard-first users: key tips, shortcuts, context menus (U)

  • Evidence: #33, #49 (key tips: initial Home-tab support landed in #43 on 10-10, the rest of the ribbon not yet), #110, #21 (Insert Copied Cells), #28 (selection), 59 command shortcuts vs Excel's 200+.
  • Estimate: 12–20 h. ui-parity.md

10. Crashes and freezes reported from the field (S)

  • Evidence: #120 (Intel UHD crash on Windows), #117 (error on close, macOS), #166 (freeze opening from Finder; PRs #164, #167), #76 (flashing on Windows 11), #130 (caught panic in table styles).
  • Estimate: 8–12 h plus an autosave/recovery soak.

After beta

11. No localization at all (L)

  • Ribbon-only catalog in five languages (#31, #104, #82; ~12% of strings each); dialogs, menus and messages English; en-US function names and separators (#45); German in PR #163. Estimate: 100–165 h including #5. localization-parity.md

12. Get & Transform, external data, data model (F)

  • No Power Query editor, connections, refresh, From Web/database/folder sources, relationships or data model (data.fromWeb, data.relationships missing). Estimate: 60–100 h.

13. Analysis tools: Solver, Analysis ToolPak, Forecast Sheet (F)

  • data.solver, data.analysisToolPak, data.forecastSheet missing (the FORECAST.ETS family the Forecast Sheet builds on is in). Estimate: 18–30 h.

14. Page Layout view and print polish (U, F)

  • No paper view with rulers and in-place header/footer editing; web print exports only (#141, PR #143). Estimate: 12–20 h.

15. Remaining functions and an oracle suite (F)

  • 18 missing, 9 stubs; no Excel-computed oracle corpus; #137, #138. Estimate: 16–26 h.
  • 20 missing, 9 stubs; no Excel-computed oracle corpus; #137, #138. Estimate: 17–29 h.
  • 21 missing, 9 stubs; no Excel-computed oracle corpus; #137, #138. Estimate: 18–30 h.
  • 21 missing, 9 stubs; trim-reference operators (A1:.E10) not parsed; no Excel-computed oracle corpus; #137, #138. Estimate: 18–30 h.
  • 18 missing, 9 stubs; no Excel-computed oracle corpus; #137, #138. Estimate: 18–30 h. function-parity.md

16. Collaboration and review (F)

  • No co-authoring, Show Changes, comment navigation, notes ↔ comments conversion; cloud storage requested (#86). Estimate: 35–55 h (co-authoring scope needs the owner).

17. Insert objects (F)

  • SmartArt, WordArt, equations, signature line, objects, 3-D models, screenshot; 8 shape kinds vs ~170; pictures in cells (#61, PR #105). Estimate: 25–40 h.

18. Accessibility (U)

  • AccessKit on, but no accessible grid/ribbon tree, no high contrast, no checker. Estimate: 12–18 h. ui-parity.md

19. XLSB/ODS beyond values; minor formats (FF)

  • XLSB and ODS import values only, no write; no .prn, .dif, .slk, XML Spreadsheet 2003; CSV writes raw instead of displayed values; templates saved with the workbook content type. Estimate: 15–23 h.

20. Multiple windows; pen pressure (H)

  • One window (#162, PR #171); no pressure or tilt in ink. Estimate: 10–14 h. hardware-parity.md

21. Automation and add-ins (F)

  • VBA out of scope (dropped on open, #6); no Office Scripts compatibility; no plugin or custom function API. Estimate: 25–40 h plus an owner decision.

22. AI features (F)

  • No in-app assistant, natural-language formulas or Analyze Data; agent control via MCP is already strong. Estimate: 30–50 h; model choice needs the owner.

23. Mobile (platform)

  • No iOS, iPadOS or Android layout. Estimate: 30–60 h.

24. Fonts and look (U)

  • Font weights beyond regular/bold (#168), system font list (#4), Calibri metrics (#29), macOS chrome (#169, #170), Linux title bar (#54), icon consistency across apps (#173). Estimate: 7–12 h.

Revision history

Date Change Summary
2026-10-11 minor §15: ODDFPRICE, ODDFYIELD, ODDLPRICE, ODDLYIELD added (18 functions missing)
2026-10-11 minor §15: PERCENTOF and EUROCONVERT added (20 functions missing)
2026-10-11 minor §15: BAHTTEXT added (21 functions missing)
2026-10-11 minor §15: TRIMRANGE added (21 functions missing; trim-reference operators still missing)
2026-10-11 minor §16: Allow Edit Ranges landed (#307)
2026-10-11 minor §8: calculated fields added
2026-10-11 minor §8: PivotCharts added
2026-10-11 minor §13, §15: FORECAST.ETS family landed (#314)
2026-10-10 minor Updated for the 10-10 merges (#17, #31, #38, #62, #82, #104, #116) and the new chart resize bug (#190)
2026-10-10 minor Noted that no gap blocks the alpha gate
2026-10-10 major Created: 24 ranked gaps with evidence, from the full re-measure and the 60 open issues