v3.12.1

Fixed:

• only use multiprocessing if max_workers > 1, not just when using METS Server, #1352
• ensure that file paths are relative to workspace directory, #1213, #1353

Removed:

• DOCKER_RABBIT_MQ_FEATURES env var not needed anymore, #1354

v3.11.0

Fixed:

• 🔥 CUDA base images working with newer Python, Ubuntu, Tensorflow, Torch and Numpy, #1350

v3.10.1

Fixed:

• Missing imports in shell_processor.py

v3.10.0

Removed:

• 🔥 Drop support for Python 3.8, #1345

Changed:

• 🔥 Upgrade from Ubuntu 20.04 to 22.04 for the docker base image, #1345
• 🔥 Restrict supported tensorflow version to < 2.16 (for support for v1 compat), #1345
• 🔥 Upgrade PAGE XML API to include PRImA-Research-Lab/PAGE-XML#24 for properly recursive RegionType, #1341

Fixed:

• Support timeouts (OCRD_PROCESSING_PAGE_TIMEOUT) for processor calls in more cases, #1345
  • In multi-processing setup, via pebble (which replaces loky) ProcessFuture and ProcessPool
  • With some C python extensions (such as tesserocr)
  • In non-networked local processor calls via cysignals

Merged PRs

• replace page.xsd with current PRImA master; regenerate PAGE API by @bertsky in #1341
• Processor: replace loky with pebble to enforce worker timeouts by @bertsky in #1345

Full Changelog: v3.9.2...v3.10.0

v3.9.2

Fixed:

• Require beanie version compatible with pydantic >=2, #1342, #1349

v3.9.1

Added:

• ocrd network client check-status has a --verbose flag for more detailed job status, #1348

v3.9.0

Changed:

• Support multiple output file groups for processors, #1344
  • OcrdPageResult: replace by proxy class OcrdPageResultVariadicListWrapper with list semantics and variadic constructor (with the original class now under SingleOcrdPageResult)
  • Processor.process_page_file: handle results from process_page_pcgts
    as lists:
    • split output_file_grp with commas (just as input_file_grp)
    • iterate over output file groups and OcrdPageResult
    • log error if there are more results than output file groups (that will get lost)
    • raise FileExistsError (in order to skip actual computation) iff output file exists for all output file groups
    • make output files (and file IDs), and save images etc for each output independently
• PAGE API: get_AllRegions available for all region types, not just PAGE root, #1344
• ocrd_network: Update RabbitMQ from 3.12 to (latest) 4.2, #1348
• ocrd_network: Fix and improve logging for network integration tests, #1348

Fixed:

• 🔥 do not log RabbitMQ credentials, #1346, #1348

Added:

• test combinations of OCRD_* config variables and multi-output, #1344

v3.8.1

Fixed:

• Include ocrd-command and ocrd-merge in the ocrd-all-tool.json, #1347

Merged PR:

• add ocrd-command and ocrd-merge to distributed ocrd-all-tool.json by @bertsky in #1347

v3.8.0

Added:

• ocrd-command processor to run arbitrary PAGE transformation CLIs, #1343
• various parameter presets for ocrd-command, #1343
• ocrd-merge processor to join multiple PAGE inputs by concatenation, #1343
• test coverage for ocrd-filter, ocrd-command, and ocrd-merge, #1343
• Resource Manager Server as ocrd_network analogon of ocrd.cli.resmgr, #1309
  • ocrd network resmgr-server for triggering Resource Manager Server (RMS) in the background
  • Processing Server also deploys RMS on each processing host

Fixed:

• Page.get_ReadingOrderGroups: sort by index, use OrderedDict as result
• OcrdAgent.notes: convert to dict to accommodate pydantic 2 with older lxml
• ocrd.resource_manager: ensure necessary + reduce unnecessary updates of user database
• ocrd.resource_manager: deduplicate entries (newer wins) before updating user database
• ocrd resmgr download: extract archives independent of whether they are URLs or local paths
• ocrd resmgr download: if --overwrite, ensure the old res gets removed
• ocrd resmgr download: default to data location instead of first in list of allowed
• ocrd_utils.list_all_resources: filter module non-resource files w/ more anti-patterns
• ocrd_utils.list_all_resources: no subpaths except for cwd location, OCR-D/spec#263, #1315
• ocrd_utils.list_all_resources: filter resources via media (MIME) type, if specified, #1315

Merged PR

• drop ocrd-distributed resource_list.yml for good by @bertsky in #1322
• resmgr download: implement git clone by @bertsky in #1340
• Continuation of #1309: Implementation of the resource manager server (issue #1294) by @MehmedGIT in #1319
• add builtin processors ocrd-command and ocrd-merge by @bertsky in #1343

Full Changelog: v3.7.0...v3.8.0

v3.7.0

Changed:

• 🔥 upgrade and adapt ocrd_network to pydantic v2, #1342

Removed:

• 🔥 drop bashlib processors, retain only the Python API, #1339