ocrd 3.1.2__py3-none-any.whl → 3.3.0__py3-none-any.whl

ocrd/cli/workspace.py

@@ -88,8 +88,8 @@ def workspace_cli(ctx, directory, mets, mets_basename, mets_server_url, backup):
 @pass_workspace
 @click.option('-a', '--download', is_flag=True, help="Download all files")
 @click.option('-s', '--skip', help="Tests to skip", default=[], multiple=True, type=click.Choice(
-    ['imagefilename', 'dimension', 'pixel_density', 'page', 'url', 'page_xsd', 'mets_fileid_page_pcgtsid',
-     'mets_unique_identifier', 'mets_file_group_names', 'mets_files', 'mets_xsd']))
+    ['imagefilename', 'alternativeimage_filename', 'alternativeimage_comments', 'dimension', 'pixel_density', 'page', 'page_xsd',
+     'url', 'mets_fileid_page_pcgtsid', 'mets_unique_identifier', 'mets_files', 'mets_xsd']))
 @click.option('--page-textequiv-consistency', '--page-strictness', help="How strict to check PAGE multi-level textequiv consistency", type=click.Choice(['strict', 'lax', 'fix', 'off']), default='strict')
 @click.option('--page-coordinate-consistency', help="How fierce to check PAGE multi-level coordinate consistency", type=click.Choice(['poly', 'baseline', 'both', 'off']), default='poly')
 @click.argument('mets_url', default=None, required=False)

ocrd/processor/base.py

@@ -779,10 +779,14 @@ class Processor():
         to handle cases like multiple output fileGrps, non-PAGE input etc.)
         """
         input_pcgts : List[Optional[OcrdPage]] = [None] * len(input_files)
-        assert isinstance(input_files[0], get_args(OcrdFileType))
-        page_id = input_files[0].pageId
+        input_pos = next(i for i, input_file in enumerate(input_files) if input_file is not None)
+        page_id = input_files[input_pos].pageId
         self._base_logger.info("processing page %s", page_id)
         for i, input_file in enumerate(input_files):
+            if input_file is None:
+                grp = self.input_file_grp.split(',')[i]
+                self._base_logger.debug(f"ignoring missing file for input fileGrp {grp} for page {page_id}")
+                continue
             assert isinstance(input_file, get_args(OcrdFileType))
             self._base_logger.debug(f"parsing file {input_file.ID} for page {page_id}")
             try:
@@ -792,7 +796,10 @@ class Processor():
             except ValueError as err:
                 # not PAGE and not an image to generate PAGE for
                 self._base_logger.error(f"non-PAGE input for page {page_id}: {err}")
-        output_file_id = make_file_id(input_files[0], self.output_file_grp)
+        output_file_id = make_file_id(input_files[input_pos], self.output_file_grp)
+        if input_files[input_pos].fileGrp == self.output_file_grp:
+            # input=output fileGrp: re-use ID exactly
+            output_file_id = input_files[input_pos].ID
         output_file = next(self.workspace.mets.find_files(ID=output_file_id), None)
         if output_file and config.OCRD_EXISTING_OUTPUT != 'OVERWRITE':
             # short-cut avoiding useless computation:

{ocrd-3.1.2.dist-info → ocrd-3.3.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ocrd
-Version: 3.1.2
+Version: 3.3.0
 Summary: OCR-D framework
 Author-email: Konstantin Baierer <unixprog@gmail.com>
 License: Apache License 2.0

{ocrd-3.1.2.dist-info → ocrd-3.3.0.dist-info}/RECORD

@@ -18,7 +18,7 @@ ocrd/cli/ocrd_tool.py,sha256=EyD5VdLm2WTzQnR-hZKpn-D4-dsWr2PIE5IoY1O3mfE,7357
 ocrd/cli/process.py,sha256=8KD0i7LT01H9u5CC1vktYMEVpS67da_rp_09_EOECmw,1233
 ocrd/cli/resmgr.py,sha256=mk8KZweC_7ENAFnC6FvFf7Zv_W1wqJTmk0EMd9XSvf4,10132
 ocrd/cli/validate.py,sha256=nvageDaHCETcE71X5lu7i_4JKpgo9MrvJKinVPLYUTI,5727
-ocrd/cli/workspace.py,sha256=KTbSzIUrba5WoYETvM9ElRZVsDUHCGVvjoFgBGZS2nU,40468
+ocrd/cli/workspace.py,sha256=bsp6YXEgwABIUFbSENmxV1c4oxRwc2L-BpeDPlYfhHE,40501
 ocrd/cli/zip.py,sha256=MMJLw3OXWiJVfVtrdJcBkbB8vA1IzSautluazZRuCQ0,5910
 ocrd/decorators/__init__.py,sha256=PyXX7vxdWkRHixas9dWUtyO3YLczcly8ZEpfZDSMVp8,7639
 ocrd/decorators/loglevel_option.py,sha256=tgipROEu3t4hkwWvFssd80k2SbTBwBIC4WNE6Gc-XAg,798
@@ -26,7 +26,7 @@ ocrd/decorators/mets_find_options.py,sha256=d4oATKMP6bFQHNqOK6nLqgUiWF2FYdkPvzkT
 ocrd/decorators/ocrd_cli_options.py,sha256=lIvtE8re1VmpHm45u71ltE0QJS8nyd28HhLC7zGSvlo,2691
 ocrd/decorators/parameter_option.py,sha256=n8hYw7XVTd3i3tvpK8F1Jx_CqRp6EGF9qJVH95yj92Q,1076
 ocrd/processor/__init__.py,sha256=39ymNwYRdc-b_OJzzKmWCvo2ga3KdsGSYDHE1Hzkn_w,274
-ocrd/processor/base.py,sha256=GcfVrgCvfHbrxngwl1VzcaZ5z7QV2e1Cn7CIjBYdcHc,59480
+ocrd/processor/base.py,sha256=yN_sMfwm2B89wtr2ShNkEtcTjXNqnvAtjM4TbWTUNCk,59929
 ocrd/processor/helpers.py,sha256=gIc6PdvOS1sR0UkYlrdZopImAXxXglDBNpgNZGWHO7Y,10987
 ocrd/processor/ocrd_page_result.py,sha256=eDkpyVHcpaBzTHXiGrcNk9PP9Xr-XZru2w_uoX_ZeNA,510
 ocrd/processor/builtin/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -36,7 +36,7 @@ ocrd/processor/builtin/dummy/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NM
 ocrd/processor/builtin/dummy/ocrd-tool.json,sha256=VoI37paWiUyMkTN5Qqau8R1Clmw24-HcZu4wjy1Br9Y,2311
 ocrd_modelfactory/__init__.py,sha256=NyJT1uSvmeEwibRFOkh0AEoVnYfP0mzxU--pP23B-TQ,4404
 ocrd_models/__init__.py,sha256=A0aj0mOraNb-xfiUueACdoaqISnp0qH-F49nTJg2vCs,380
-ocrd_models/constants.py,sha256=fI6Qz4OPOm6UBLQ_P2dlpjcwB0XFJZ7AgxxKqgc75X0,2724
+ocrd_models/constants.py,sha256=kvvAAro_1YOTRWwFgbrGEFeDZ8_u0S624Y3icNNk4Oo,2987
 ocrd_models/mets-empty.xml,sha256=dFixfbxSXrgjZx9BfdIKWHX-khNmp7dNYaFe2qQSwCY,1203
 ocrd_models/ocrd_agent.py,sha256=E9OtDhz9UfKb6ou2qvsuCL9NlO1V6zMb0s8nVq8dVos,5609
 ocrd_models/ocrd_exif.py,sha256=wRSprHxCy9LCXw41Fi9kp-CbFc5NFX9ZFIFNszB41qk,4585
@@ -115,14 +115,14 @@ ocrd_validators/parameter_validator.py,sha256=_5Y3IS24Sf_xHBkB3TE3jB9VTCbbjWO8bS
 ocrd_validators/processing_server_config.schema.yml,sha256=8NQbhSshm1exTvbdYiu694rZZ-Xe70_vQtsJ0nd7ZCM,5432
 ocrd_validators/processing_server_config_validator.py,sha256=lQ2-ZxsvbFki_SvE_N4_1ptBnBHcwOTJ5grtL2G9F8A,810
 ocrd_validators/resource_list_validator.py,sha256=cFMj0n_x-tjhuNUpjgEvPP8iPVm7lme9TWAaqATasV0,776
-ocrd_validators/workspace_validator.py,sha256=rEXIwjtNpt8HcTv94fKed3vVlA3U4z7Xmm1ZL1VHC84,17892
+ocrd_validators/workspace_validator.py,sha256=JNPsRVPgQI0vsaxcs_c3qj22GagdZcgO3v9u3sbBbBI,20340
 ocrd_validators/xlink.xsd,sha256=8fW7YAMWXN2PbB_MMvj9H5ZeFoEBDzuYBtlGC8_6ijw,3180
 ocrd_validators/xsd_mets_validator.py,sha256=4GWfLyqkmca0x7osDuXuExYuM0HWVrKoqn0S35sFhHU,467
 ocrd_validators/xsd_page_validator.py,sha256=BNz_9u-Ek4UCeyZu3KxSQoolfW9lvuaSR9nIu1XXxeE,467
 ocrd_validators/xsd_validator.py,sha256=6HrVAf6SzCvfUIuQdIzz9bOq4V-zhyii9yrUPoK2Uvo,2094
-ocrd-3.1.2.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-ocrd-3.1.2.dist-info/METADATA,sha256=gtLDkn539WBI_GpwiwsMe5iIv_bmePJCFqgRfkf9Gnc,10442
-ocrd-3.1.2.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
-ocrd-3.1.2.dist-info/entry_points.txt,sha256=4hcJ2LkK_OlIabHnKgFit35Ap7b5Lz1Gb4hzkxV0Kiw,152
-ocrd-3.1.2.dist-info/top_level.txt,sha256=pUgiN42t4KXC5rvpi6V8atza31XP4SCznXpXlVlvomM,75
-ocrd-3.1.2.dist-info/RECORD,,
+ocrd-3.3.0.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+ocrd-3.3.0.dist-info/METADATA,sha256=K8u-P2RKlBlGPPq8h8sv5hLhWi0XgRmhF-Bf-F3Qgpc,10442
+ocrd-3.3.0.dist-info/WHEEL,sha256=iAkIy5fosb7FzIOwONchHf19Qu7_1wCWyFNR5gu9nU0,91
+ocrd-3.3.0.dist-info/entry_points.txt,sha256=4hcJ2LkK_OlIabHnKgFit35Ap7b5Lz1Gb4hzkxV0Kiw,152
+ocrd-3.3.0.dist-info/top_level.txt,sha256=pUgiN42t4KXC5rvpi6V8atza31XP4SCznXpXlVlvomM,75
+ocrd-3.3.0.dist-info/RECORD,,

{ocrd-3.1.2.dist-info → ocrd-3.3.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.3.0)
+Generator: setuptools (75.3.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

ocrd_models/constants.py

@@ -28,6 +28,8 @@ __all__ = [
     'TAG_PAGE_TEXTEQUIV',
     'TAG_PAGE_TEXTREGION',
     'METS_PAGE_DIV_ATTRIBUTE',
+    'PAGE_REGION_TYPES',
+    'PAGE_ALTIMG_FEATURES',
 ]
 
 
@@ -72,6 +74,20 @@ PAGE_REGION_TYPES = [
     'Separator', 'Table', 'Text', 'Unknown'
 ]
 
+PAGE_ALTIMG_FEATURES = [
+    'binarized',
+    'grayscale_normalized',
+    'despeckled',
+    'cropped',
+    'deskewed',
+    'rotated-90',
+    'rotated-180',
+    'rotated-270',
+    'dewarped',
+    'clipped',
+]
+
+
 class METS_PAGE_DIV_ATTRIBUTE(Enum):
     ID = auto()
     ORDER = auto()

ocrd_validators/workspace_validator.py

@@ -7,6 +7,7 @@ from pathlib import Path
 
 from ocrd_utils import getLogger, MIMETYPE_PAGE, pushd_popd, is_local_filename, DEFAULT_METS_BASENAME
 from ocrd_models import ValidationReport
+from ocrd_models.constants import PAGE_ALTIMG_FEATURES
 from ocrd_modelfactory import page_from_file
 
 from .constants import FILE_GROUP_CATEGORIES, FILE_GROUP_PREFIX
@@ -98,6 +99,9 @@ class WorkspaceValidator():
         self.page_coordinate_consistency = page_coordinate_consistency
         # there will be more options to come
         self.page_checks = [check for check in ['mets_fileid_page_pcgtsid',
+                                                'imagefilename',
+                                                'alternativeimage_filename',
+                                                'alternativeimage_comments',
                                                 'dimension',
                                                 'page',
                                                 'page_xsd']
@@ -118,7 +122,7 @@ class WorkspaceValidator():
             mets_url (string): URL of the METS file
             src_dir (string, None): Directory containing mets file
             skip (list): Validation checks to omit. One or more of 
-                'mets_unique_identifier', 'mets_file_group_names', 
+                'mets_unique_identifier',
                 'mets_files', 'pixel_density', 'dimension', 'url',
                 'multipage', 'page', 'page_xsd', 'mets_xsd', 
                 'mets_fileid_page_pcgtsid'
@@ -145,8 +149,6 @@ class WorkspaceValidator():
             try:
                 if 'mets_unique_identifier' not in self.skip:
                     self._validate_mets_unique_identifier()
-                if 'mets_file_group_names' not in self.skip:
-                    self._validate_mets_file_group_names()
                 if 'mets_files' not in self.skip:
                     self._validate_mets_files()
                 if 'pixel_density' not in self.skip:
@@ -192,7 +194,11 @@ class WorkspaceValidator():
             self.workspace.download_file(f)
             page = page_from_file(f).get_Page()
             imageFilename = page.imageFilename
-            if not self.mets.find_files(url=imageFilename, **self.find_kwargs):
+            if is_local_filename(imageFilename):
+                kwargs = dict(local_filename=imageFilename, **self.find_kwargs)
+            else:
+                kwargs = dict(url=imageFilename, **self.find_kwargs)
+            if not self.mets.find_files(**kwargs):
                 self.report.add_error(f"PAGE '{f.ID}': imageFilename '{imageFilename}' not found in METS")
             if is_local_filename(imageFilename) and not Path(imageFilename).exists():
                 self.report.add_warning(f"PAGE '{f.ID}': imageFilename '{imageFilename}' points to non-existent local file")
@@ -295,6 +301,9 @@ class WorkspaceValidator():
             if f.url and 'url' not in self.skip:
                 if re.match(r'^file:/[^/]', f.url):
                     self.report.add_error(f"File '{f.ID}' has an invalid (Java-specific) file URL '{f.url}'")
+                elif ':' not in f.url:
+                    self.report.add_error(f"File '{f.ID}' has an invalid (non-URI) file URL '{f.url}'")
+                    continue
                 scheme = f.url[0:f.url.index(':')]
                 if scheme not in ('http', 'https', 'file'):
                     self.report.add_warning(f"File '{f.ID}' has non-HTTP, non-file URL '{f.url}'")
@@ -321,17 +330,43 @@ class WorkspaceValidator():
             pcgts = page_from_file(f)
             page = pcgts.get_Page()
             if 'dimension' in self.page_checks:
-                _, _, exif = self.workspace.image_from_page(page, f.pageId)
-                if page.imageHeight != exif.height:
-                    self.report.add_error(f"PAGE '{f.ID}': @imageHeight != image's actual height ({page.imageHeight} != {exif.height})")
-                if page.imageWidth != exif.width:
-                    self.report.add_error(f"PAGE '{f.ID}': @imageWidth != image's actual width ({page.imageWidth} != {exif.width})")
+                img = self.workspace._resolve_image_as_pil(page.imageFilename)
+                if page.imageHeight != img.height:
+                    self.report.add_error(f"PAGE '{f.ID}': @imageHeight != image's actual height ({page.imageHeight} != {img.height})")
+                if page.imageWidth != img.width:
+                    self.report.add_error(f"PAGE '{f.ID}': @imageWidth != image's actual width ({page.imageWidth} != {img.width})")
             if 'imagefilename' in self.page_checks:
                 imageFilename = page.imageFilename
-                if not self.mets.find_files(url=imageFilename):
+                if is_local_filename(imageFilename):
+                    kwargs = dict(local_filename=imageFilename, **self.find_kwargs)
+                else:
+                    kwargs = dict(url=imageFilename, **self.find_kwargs)
+                if not self.mets.find_files(**kwargs):
                     self.report.add_error(f"PAGE '{f.ID}': imageFilename '{imageFilename}' not found in METS")
                 if is_local_filename(imageFilename) and not Path(imageFilename).exists():
                     self.report.add_warning(f"PAGE '{f.ID}': imageFilename '{imageFilename}' points to non-existent local file")
+            if 'alternativeimage_filename' in self.page_checks:
+                for altimg in page.get_AllAlternativeImages():
+                    if is_local_filename(altimg.filename):
+                        kwargs = dict(local_filename=altimg.filename, **self.find_kwargs)
+                    else:
+                        kwargs = dict(url=altimg.filename, **self.find_kwargs)
+                    if not self.mets.find_files(**kwargs):
+                        self.report.add_error(f"PAGE '{f.ID}': {altimg.parent_object_.id} AlternativeImage "
+                                              f"'{altimg.filename}' not found in METS")
+                    if is_local_filename(altimg.filename) and not Path(altimg.filename).exists():
+                        self.report.add_warning(f"PAGE '{f.ID}': {altimg.parent_object_.id} AlternativeImage "
+                                                f"'{altimg.filename}' points to non-existent local file")
+            if 'alternativeimage_comments' in self.page_checks:
+                for altimg in page.get_AllAlternativeImages():
+                    if altimg.comments is None:
+                        self.report.add_error(f"PAGE '{f.ID}': {altimg.parent_object_.id} AlternativeImage "
+                                              f"'{altimg.filename}' features not specified in PAGE")
+                    else:
+                        for feature in altimg.comments.split(','):
+                            if feature not in PAGE_ALTIMG_FEATURES:
+                                self.report.add_error(f"PAGE '{f.ID}': {altimg.parent_object_.id} AlternativeImage "
+                                                      f"'{altimg.filename}' feature '{feature}' not standardized for PAGE")
             if 'mets_fileid_page_pcgtsid' in self.page_checks and pcgts.pcGtsId != f.ID:
                 self.report.add_warning('pc:PcGts/@pcGtsId differs from mets:file/@ID: "%s" !== "%s"' % (pcgts.pcGtsId or '', f.ID or ''))