chore: unified file domain model

[Israeltheminer]

Summary

• Add lib/shared/file-types.ts as single source of truth for all file type definitions across the platform
• Migrate 13 files to import MIME types, extensions, accept strings, size limits, and classification helpers from the shared registry
• Remove ~120 lines of duplicated file type logic scattered across chat uploads, document uploads, import forms, attachment processing, and parse routing

What's in the registry

• MIME constants: MIME_TYPES.PDF, MIME_TYPES.DOCX, etc.
• Grouped sets: IMAGE_MIME_TYPES, DOCUMENT_MIME_TYPES, SPREADSHEET_MIME_TYPES
• Classification: isImage(), isTextFile(), isSpreadsheet(), isParseable()
• Extensions: extractExtension(), getDisplayExtension()
• Accept strings: CHAT_UPLOAD_ACCEPT, DOCUMENT_UPLOAD_ACCEPT, SPREADSHEET_IMPORT_ACCEPT
• Size limits: CHAT_MAX_FILE_SIZE, DOCUMENT_MAX_FILE_SIZE
• Parse routing: getParseEndpoint()
• Display labels: getFileTypeLabelKey()

Files migrated

File	What was replaced
use-convex-file-upload.ts	10-line MIME array + hardcoded size limit
chat-input.tsx	Hardcoded accept string + MIME-to-label chain
document-upload-dialog.tsx	Long accept string + MAX_FILE_SIZE_BYTES
use-document-upload.ts	MAX_FILE_SIZE_BYTES constant
vendor-import-form.tsx	.endsWith() checks + hardcoded accept
product-import-form.tsx	Same pattern
customer-import-form.tsx	Same pattern
parse_file.ts	14-line getParseEndpoint()
extract_extension.ts	37-line function → re-export
process_attachments.ts	Inline isTextFile + image classification
document-helpers.ts	25-line getFileExtension() → re-export
document-icon.tsx	Local getExtension()
file-parsing.ts	isExcelFile() → uses shared isSpreadsheet()

Test plan

• Verify chat file upload accepts images, PDFs, Word, PowerPoint, and text files
• Verify document upload dialog accepts all supported document types
• Verify vendor/product/customer import forms accept .xlsx, .xls, .csv only
• Verify document icons render correctly for all file types
• Verify file parsing routes to correct endpoints (PDF, DOCX, PPTX)
• Verify attachment processing correctly classifies images, text files, and documents

Summary by CodeRabbit

• Refactor
  • Centralized file-handling configuration across the platform, including file type validation, size limits, and display labels.
  • Unified file upload logic for documents, chat attachments, spreadsheet imports, and other file-related features.
  • Standardized file type detection and validation to ensure consistent behavior throughout the application.

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR centralizes file type handling by introducing a new shared file-types.ts module containing MIME type definitions, file classification helpers, upload accept strings, size limits, and parse endpoint routing. It then refactors multiple components and utilities across the codebase to use these centralized constants and helpers instead of hardcoded values or duplicated logic, moving file type validation, extension extraction, and MIME type checking from scattered locations into a single source of truth.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Possibly related PRs

• refactor(platform): update file upload components #344: Modifies file upload flows in the same files (use-convex-file-upload, chat-input, import forms) to centralize file-type handling and upload accept constants.
• feat: add file attachment support for chat agent with document parsing #12: Touches the file parsing helper (parse_file) and chat attachment parsing paths that are also refactored in this PR.
• fix: ui improvements #377: Updates chat file-upload/display logic and file-type utilities in overlapping files like chat-input.tsx and file-type extraction helpers.

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 35.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'chore: unified file domain model' accurately describes the main change: consolidating file-type logic into a centralized shared module.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch chore/file-domain-model

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 3

🤖 Fix all issues with AI agents

In `@services/platform/lib/shared/file-types.ts`:
- Around line 91-94: The isParseable and getParseEndpoint functions disagree on
allowed extensions which can cause silent fallback to the PDF parser; update
isParseable and the parsing-extension checks in getParseEndpoint to use the
exact same normalized extension set (include or exclude .doc/.ppt consistently)
and add a defensive processLogger/console.warn inside getParseEndpoint (or its
caller) to log when an unexpected extension is received and the code is falling
back to the PDF route; reference the isParseable function and getParseEndpoint
function names to locate where to align the extension list and insert the
warning.
- Around line 43-69: The MIME sets (IMAGE_MIME_TYPES, DOCUMENT_MIME_TYPES,
PRESENTATION_MIME_TYPES, SPREADSHEET_MIME_TYPES, TEXT_MIME_TYPES) are strict
allow-lists while the helper functions (isImage, isTextFile) use broader
heuristics (e.g., startsWith('image/') and extension checks); add brief doc
comments above each exported ReadonlySet (at least above IMAGE_MIME_TYPES and
TEXT_MIME_TYPES) explaining that the sets are for strict
validation/allow-listing and that isImage() and isTextFile() intentionally
perform looser classification for routing/UX, so they are not interchangeable —
mention examples like image/svg+xml and .txt/.log to clarify intent.
- Around line 218-225: Update getFileTypeLabelKey to detect spreadsheet MIME
types so they don't fall through to 'file': use the existing
SPREADSHEET_MIME_TYPES (and/or MIME_TYPES.XLS, MIME_TYPES.XLSX, MIME_TYPES.CSV)
to check if mimeType matches any spreadsheet entry (e.g.,
SPREADSHEET_MIME_TYPES.includes(mimeType) or compare against MIME_TYPES
constants) and return an appropriate label such as 'spreadsheet' (or specific
keys like 'xls'/'xlsx'/'csv' if you prefer); add this check before the final
return so spreadsheet MIME types are handled by getFileTypeLabelKey.