Skip to content

Changelog

All notable changes to this project will be documented in this file.

The format loosely follows Keep a Changelog.

2026-09-07

Added

CSV Bulk Import for Students

  • Added student_import.py service with CSV template generation, preview, and single-transaction import
  • Added a downloadable CSV template via the students page (served in memory, students_template.csv)
  • Added a + BULK IMPORT button next to + Add Student that opens a native file picker (.csv)
  • Added an import preview dialog that buckets rows into to add / to skip / errors before committing
  • Students are de-duplicated on the (first_name, last_name) business key (case-insensitive); duplicates in the file and existing database rows are skipped, never overwritten
  • Rows missing a first or last name are reported as row-level errors rather than silently skipped
  • Added tests/test_student_import.py covering add, in-file and database de-duplication, invalid-row errors, and template output (4 tests)

GLOW (ACB Large Print Toolkit) Integration

  • Registered GLOW as a CLI QA tool (glow = acb-large-print) in qa_service.py and the Execution Service allow-list
  • Added a Large Print (ACB/GLOW) Pipeline in pipeline_service.py
  • Added run_glow_audit and GLOW binary discovery to toolchain_binaries.py
  • Added a GLOW card to the binary integrations dashboard

FIDO (DAISY Labs AI) Integration

  • Registered FIDO as a manual-review QA tool (manual_review=True) since it has no stable CLI
  • Documented FIDO's desktop-only workflow in tools.ini / tools.ini.example

File Picker Dialogs

  • Added a reusable file_picker component that stages selected files into FILES_DIR (components.py)
  • Replaced manual file-path typing in job attach dialogs and the print job model-file dialog

Open Project Folder

  • Added an Open Project Folder affordance to job Files cards via the open_folder helper

Changed

  • Configured tools.ini and tools.ini.example with glow = acb-large-print and a commented # fido = entry
  • Updated README and MkDocs documentation to cover GLOW/FIDO integration
  • Documented the CSV bulk-import workflow in the Students user guide

2026-08-20

Added

Interactive Setup Assistant

  • Added setup.py interactive terminal setup assistant (OS detection, prerequisite checks, .secrets creation with auto-generated secrets, tools.ini setup from example, validation)
  • Added setup.sh launcher for macOS/Linux
  • Added setup.bat launcher for Windows

Auto-Setup Integration

  • App now auto-runs setup.py when .secrets is missing on first launch

Changed

  • load_secrets() in app.py no longer crashes on missing .secrets; instead launches the setup assistant automatically
  • Updated README.md quick-start to reflect new auto-setup flow
  • Updated MkDocs installation, configuration, troubleshooting, and deployment docs

2026-07-15

Added

Test Suite

  • Created tests/conftest.py with shared database fixtures
  • Created tests/test_services.py with 60 service-layer tests (RBAC, authentication, SLA monitoring, workflow DAG, persistent queue/analytics/provenance, artifact retention, audit log, event stream, distributed workers, multi-tenant, metadata validation, compliance reporting, execution service, worker runtime, base analytics, base provenance, workflow queue)
  • Created tests/test_api.py with 16 API endpoint tests (healthcheck, workflow listing/enqueue, analytics summary, provenance events, API key authentication enforcement)

Build Infrastructure

  • Added sys._MEIPASS detection in app.py for PyInstaller frozen-mode favicon resolution
  • Added version.py for dynamic date-based versioning (YYYY.M.D format)
  • Added _base_path() and _get_version() helpers in app.py

API Security

  • Added RBAC permission-gated FastAPI dependencies (_dep_workflow_read, _dep_workflow_manage, _dep_analytics_view, _dep_governance_manage)
  • Added governance.manage permission to the operator role

Workflow Queue Service

  • Fleshed out workflow_queue.py with full WorkflowQueueService implementation (in-memory priority queue with enqueue, next_job, complete_job, fail_job, list_jobs)

Changed

Build Scripts

  • Fixed --add-data paths in build_linux.sh, build_macos.sh, build_windows.bat (resources directory is at project root, not under accessibility_mgr/)
  • Fixed --icon path in build_windows.bat to resources\icons\icon.ico
  • Removed --icon=...favicon.icns from build_macos.sh (no .icns file exists)
  • Fixed build_windows.bat entry point path to use backslash

Containerfile

  • Fixed build order: copy full source before uv pip install (hatchling needs source to build wheel)
  • Replaced curl-based healthcheck with Python urllib.request (curl not in python:3.12-slim)
  • Removed premature COPY tools.ini.example tools.ini

podman-compose.yml

  • Removed deprecated version: '3.8' key

Dynamic Versioning

  • Replaced hardcoded version = "2026.6.9" with dynamic = ["version"] using hatchling code source
  • FastAPI app version now reads from importlib.metadata at runtime
  • Health endpoint now includes version field

License

  • Changed classifier in pyproject.toml from MIT License to Apache Software License
  • Updated docs/index.md and docs/project/contributors.md to reference Apache 2.0
  • Replaced MIT license text in docs/license.md with Apache 2.0 summary

Documentation Accessibility

  • Removed all unicode emoji from 35 markdown files in docs/
  • Replaced checkmark symbols (✅, ❌) with text (Yes, No) in table cells
  • All section headers are now plain ASCII text

Service Fixes

  • Added conn.row_factory = sqlite3.Row to _connect() in 7 services: sla_monitoring.py, workflow_dag.py, artifact_retention.py, audit_log.py, event_stream.py, distributed_workers.py, multi_tenant.py

Technical Debt

  • Wired ACCESSMAN_LOG_LEVEL environment variable into app.py logging setup
  • Fixed CHANGELOG.md references to CHANGES.md in docs/project/release-process.md
  • Synced CHANGES.md content into docs/changelog.md

Removed

  • Removed unused json import from sla_monitoring.py
  • Removed unused dataclasses.asdict import from sla_monitoring.py and workflow_dag.py
  • Removed unused dataclasses.field import from workflow_queue.py

2026-05-17

Added

Enterprise Workflow Orchestration

  • Persistent SQLite-backed workflow queue
  • Workflow DAG orchestration engine
  • Workflow dependency resolution
  • Workflow replay and recovery engine
  • Distributed reconciliation service
  • Worker heartbeat monitoring
  • Worker failover detection
  • SLA monitoring infrastructure
  • Distributed worker registry

Accessibility Automation

  • Production DAISY Ace binary integration
  • Production EPUBCheck integration
  • Secure subprocess sandboxing layer
  • Accessibility toolchain telemetry
  • Artifact capture orchestration
  • Timeout-controlled binary execution
  • Binary integration monitoring dashboard

CI/CD Accessibility Governance

  • GitHub Actions accessibility validation workflow
  • CI/CD accessibility validation orchestration
  • Accessibility severity-threshold policy engine
  • Release-blocking governance enforcement
  • Automated EPUB accessibility validation
  • Pipeline execution telemetry
  • CI/CD governance dashboard

Security and Identity

  • Tenant-aware RBAC enforcement
  • Authentication middleware
  • API token validation
  • Credential vault abstraction
  • Permission evaluation engine
  • Multi-tenant organization infrastructure
  • Tenant membership management

Reliability Infrastructure

  • Webhook retry engine
  • Dead-letter queue support
  • Persistent event store
  • Signed event persistence
  • Artifact retention lifecycle management
  • Automated artifact cleanup
  • Distributed orchestration telemetry

Governance and Compliance

  • Operational provenance registry
  • Audit-grade event tracking
  • Governance workflow telemetry
  • Compliance-oriented workflow orchestration
  • Operational analytics persistence

Changed

  • Expanded repository architecture from NiceGUI workflow application into enterprise accessibility governance platform
  • Transitioned orchestration layer from in-memory execution to durable distributed workflow infrastructure
  • Transitioned accessibility tooling from execution stubs to production binary integration model
  • Expanded CI/CD support into deployable accessibility governance automation
  • Expanded RBAC implementation into tenant-aware authorization model
  • Expanded operational telemetry into audit-grade event persistence architecture
  • Expanded reliability infrastructure with replay/recovery semantics and distributed reconciliation

Planned

  • Real encrypted KMS/HSM-backed secret storage
  • Queue partitioning and distributed workload balancing
  • Tenant-aware API authorization middleware
  • Signed governance export generation
  • Immutable compliance snapshot exports
  • Workflow auto-remediation routing
  • Adaptive orchestration retry policies
  • Distributed auto-scaling workers

2026-05-09

Added

  • Initial NiceGUI application shell
  • Dynamic page loading for UI modules
  • Sidebar-based navigation
  • Responsive content layout
  • Dashboard integration
  • Graceful handling for incomplete UI modules
  • Application-wide color/theme configuration
  • Header/status display components
  • Architecture and workflow audit documentation
  • LLM remediation prompt specification (prompt.json)
  • Asset Registry NiceGUI page
  • Metadata workflow visibility dashboard
  • METS-inspired asset management concepts
  • Metadata lineage planning documentation

Changed

  • Replaced placeholder app.py implementation with a functional NiceGUI frontend
  • Updated project documentation to reflect current architecture
  • Updated README to document NiceGUI migration strategy
  • Updated repository structure documentation
  • Clarified technology stack and workflow functionality
  • Expanded application shell to include metadata-oriented workflows
  • Expanded project scope toward preservation-oriented accessibility production

Planned

  • Full migration from Textual TUI to NiceGUI
  • CRUD interfaces for all production workflows
  • Reporting and analytics
  • Authentication support
  • Accessibility auditing improvements
  • Multi-user operation support
  • Full metadata lineage tracking
  • Asset version graph visualization
  • Preservation export tooling
  • Workflow automation pipelines