adobe-to-docusign-migrator

Commit Graph

Author	SHA1	Message	Date
Paul Huliganga	447a89923a	docs: comprehensive project documentation update architecture.md — full rewrite to reflect current v2 state: - Accurate component map and pipeline stages - Session lifecycle (server-side files, cookie signing, rotation) - Multi-account DocuSign support flow - Audit log record schema - Batch job in-memory state caveat documented - Security design table (log sanitizer, session signing, PDF checksums) - Known limitations table (retry gaps, shard config, CI fixtures) PRODUCT-SPEC.md — remove phantom migration_service.py and pdf_coords.py that were in the original spec but never implemented; document where pipeline orchestration actually lives README.md — add Production deployment section covering: - Reverse proxy / HTTPS requirement for OAuth callbacks - Required env vars table - SESSION_SECRET_KEY rotation procedure - Adobe shard configuration (EU2 / NA1 / others via ADOBE_SIGN_BASE_URL) - DocuSign sandbox-to-production switch - Session store maintenance (stale file cleanup) field-mapping.md — add Multi-Document Templates section explaining documentId assignment, page number behaviour, and the known limitation for multi-doc templates where page numbers are not rebased per document Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-23 09:51:38 -04:00
Paul Huliganga	342e8c3471	feat(phases-8-13): blueprint alignment — normalized schema, validation, migration options, retry, security, batch Phase 8: Normalized intermediate schema (src/models/normalized_template.py, src/services/mapping_service.py) - Platform-agnostic NormalizedTemplate as pipeline bridge - PDF SHA-256 checksums computed during normalization - 13 tests passing Phase 9: Validation service + report builder (src/services/validation_service.py, src/reports/report_builder.py) - Blockers (no recipients, no documents) halt migration - Warnings (no fields, unassigned roles, unsupported features) logged - Structured MigrationReport with per-template status - 20 tests passing Phase 10: Migration options API (web/routers/migrate.py) - source_template_ids, dry_run, overwrite_if_exists, include_documents options - Backward compatible with legacy adobe_template_ids field - 7 tests passing Phase 11: Retry with exponential backoff (src/utils/retry.py) - retry_with_backoff and async_retry_with_backoff decorators - 429/5xx detection via check_response_retryable - 14 tests passing Phase 12: Security hardening (src/utils/log_sanitizer.py) - SanitizingFilter redacts tokens, JWTs, base64 PDF content from logs - 15 tests passing Phase 13: Batch migration API - POST /api/migrate/batch — async background job with job_id - GET /api/migrate/batch/{job_id} — poll progress and results - 6 tests passing Full suite: 108/108 tests passing Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>	2026-04-21 02:19:38 -04:00
Paul Huliganga	b9a4e75da9	Initial project scaffold (Cleo)	2026-04-14 19:21:17 -04:00

Author

SHA1

Message

Date

Paul Huliganga

447a89923a

docs: comprehensive project documentation update

architecture.md — full rewrite to reflect current v2 state:
  - Accurate component map and pipeline stages
  - Session lifecycle (server-side files, cookie signing, rotation)
  - Multi-account DocuSign support flow
  - Audit log record schema
  - Batch job in-memory state caveat documented
  - Security design table (log sanitizer, session signing, PDF checksums)
  - Known limitations table (retry gaps, shard config, CI fixtures)

PRODUCT-SPEC.md — remove phantom migration_service.py and pdf_coords.py
  that were in the original spec but never implemented; document where
  pipeline orchestration actually lives

README.md — add Production deployment section covering:
  - Reverse proxy / HTTPS requirement for OAuth callbacks
  - Required env vars table
  - SESSION_SECRET_KEY rotation procedure
  - Adobe shard configuration (EU2 / NA1 / others via ADOBE_SIGN_BASE_URL)
  - DocuSign sandbox-to-production switch
  - Session store maintenance (stale file cleanup)

field-mapping.md — add Multi-Document Templates section explaining
  documentId assignment, page number behaviour, and the known limitation
  for multi-doc templates where page numbers are not rebased per document

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-23 09:51:38 -04:00

Paul Huliganga

342e8c3471

feat(phases-8-13): blueprint alignment — normalized schema, validation, migration options, retry, security, batch

Phase 8: Normalized intermediate schema (src/models/normalized_template.py, src/services/mapping_service.py)
  - Platform-agnostic NormalizedTemplate as pipeline bridge
  - PDF SHA-256 checksums computed during normalization
  - 13 tests passing

Phase 9: Validation service + report builder (src/services/validation_service.py, src/reports/report_builder.py)
  - Blockers (no recipients, no documents) halt migration
  - Warnings (no fields, unassigned roles, unsupported features) logged
  - Structured MigrationReport with per-template status
  - 20 tests passing

Phase 10: Migration options API (web/routers/migrate.py)
  - source_template_ids, dry_run, overwrite_if_exists, include_documents options
  - Backward compatible with legacy adobe_template_ids field
  - 7 tests passing

Phase 11: Retry with exponential backoff (src/utils/retry.py)
  - retry_with_backoff and async_retry_with_backoff decorators
  - 429/5xx detection via check_response_retryable
  - 14 tests passing

Phase 12: Security hardening (src/utils/log_sanitizer.py)
  - SanitizingFilter redacts tokens, JWTs, base64 PDF content from logs
  - 15 tests passing

Phase 13: Batch migration API
  - POST /api/migrate/batch — async background job with job_id
  - GET /api/migrate/batch/{job_id} — poll progress and results
  - 6 tests passing

Full suite: 108/108 tests passing

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-21 02:19:38 -04:00

Paul Huliganga

b9a4e75da9

Initial project scaffold (Cleo)

2026-04-14 19:21:17 -04:00

3 Commits