test for style inheritance

CHANGELOG.md

@@ -5,6 +5,19 @@ Versioned according to [Semantic Versioning](http://semver.org/).
 
 ## Unreleased
 
+## [2.28.0] - 2021-11-30
+
+Added:
+
+  * Store parameterization of processors in METS for provenance, #747
+  * `ocrd workspace find --download`: Add a `--wait` option to wait between downloads, #745
+  * bashlib: Check fileGrps when parsing CLI args, #743, OCR-D/ocrd_olena#76
+  * Dockerfile: Install `time` to have `/usr/bin/time` in the image, #748, OCR-D/ocrd_all#271
+
+Fixed:
+
+  * `ocrd-dummy`: Also set pcGtsId, v0.0.2, #739
+
 ## [2.27.0] - 2021-11-09
 
 Fixed:
@@ -1381,6 +1394,7 @@ Fixed
 Initial Release
 
 <!-- link-labels -->
+[2.28.0]: ../../compare/v2.28.0..v2.27.0
 [2.27.0]: ../../compare/v2.27.0..v2.26.1
 [2.26.1]: ../../compare/v2.26.1..v2.26.0
 [2.26.0]: ../../compare/v2.26.0..v2.25.1

Dockerfile

@@ -23,6 +23,7 @@ RUN apt-get update && apt-get -y install --no-install-recommends \
     python3-pip \
     make \
     wget \
+    time \
     curl \
     sudo \
     git \

README.md

@@ -53,7 +53,7 @@ pip install ocrd
 pip install ocrd_modelfactory
 ```
 
-All python software released by [OCR-D](https://github.com/OCR-D) requires Python 3.5 or higher.
+All python software released by [OCR-D](https://github.com/OCR-D) requires Python 3.6 or higher.
 
 ## Command line tools
 
@@ -118,5 +118,6 @@ Test with local files: `make test`
 
 ## See Also
 
-  - [OCR-D Specifications](https://ocr-d.github.io) ([Repo](https://github.com/ocr-d/spec))
-  - [OCR-D Documentation](https://ocr-d.github.io/docs) ([Repo](https://github.com/ocr-d/docs))
+  - [OCR-D Specifications](https://https://ocr-d.de/en/spec/) ([Repo](https://github.com/ocr-d/spec))
+  - [OCR-D core API documentation](https://ocr-d.de/core) (built here via `make docs`)
+  - [OCR-D Website](https://ocr-d.de) ([Repo](https://github.com/ocr-d/ocrd-website))

ocrd/bashlib/Makefile

@@ -13,6 +13,7 @@ help:
 # Build the lib
 lib: lib.bash
 
+.DELETE_ON_ERROR:
 lib.bash: $(wildcard src/*.bash)
 	shinclude -c pound src/lib.bash > "$@"
 

ocrd/bashlib/src/parse_argv.bash

@@ -68,6 +68,18 @@ ocrd__parse_argv () {
         ocrd__raise "Provide --output-file-grp/-O explicitly!"
     fi
 
+    # check fileGrps
+    local _valopts=( --workspace "${ocrd__argv[working_dir]}" )
+    if [[ ${ocrd__argv[overwrite]} = true ]]; then
+        _valopts+=( --overwrite )
+    fi
+    if [[ -n "${ocrd__argv[page_id]:-}" ]]; then
+        _valopts+=( --page-id "${ocrd__argv[page_id]}" )
+    fi
+    _valopts+=( "${OCRD_TOOL_NAME#ocrd-} -I ${ocrd__argv[input_file_grp]} -O ${ocrd__argv[output_file_grp]}" )
+    ocrd validate tasks "${_valopts[@]}" || exit $?
+
+    # check parameters
     local params_parsed retval
     params_parsed="$(ocrd ocrd-tool "$OCRD_TOOL_JSON" tool $OCRD_TOOL_NAME parse-params "${__parameters[@]}" "${__parameter_overrides[@]}")" || {
         retval=$?

ocrd/ocrd/cli/validate.py

@@ -100,14 +100,16 @@ def validate_page(page, **kwargs):
 #      _inform_of_result(WorkspaceValidator.validate(Resolver(), mets_url, **kwargs))
 
 @validate_cli.command('tasks')
-@click.option('--workspace', nargs=1, required=False, help='Workspace these tasks are to be run. If omitted, only validate syntax')
+@click.option('--workspace', nargs=1, required=False, help='Workspace directory these tasks are to be run. If omitted, only validate syntax')
+@click.option('--overwrite', is_flag=True, default=False, help='When checking against a concrete workspace, simulate overwriting output or page range.')
+@click.option('-g', '--page-id', help="ID(s) of the pages to process")
 @click.argument('tasks', nargs=-1, required=True)
-def validate_process(tasks, workspace):
+def validate_process(tasks, workspace, overwrite, page_id):
     '''
     Validate a sequence of tasks passable to 'ocrd process'
     '''
     if workspace:
-        _inform_of_result(validate_tasks([ProcessorTask.parse(t) for t in tasks], Workspace(Resolver(), directory=workspace)))
+        _inform_of_result(validate_tasks([ProcessorTask.parse(t) for t in tasks], Workspace(Resolver(), directory=workspace), page_id=page_id, overwrite=overwrite))
     else:
         for t in [ProcessorTask.parse(t) for t in tasks]:
             _inform_of_result(t.validate())

ocrd/ocrd/cli/workspace.py

@@ -13,6 +13,7 @@
 import sys
 from glob import glob   # XXX pathlib.Path.glob does not support absolute globs
 import re
+import time
 
 import click
 
@@ -347,8 +348,9 @@ def workspace_cli_bulk_add(ctx, regex, mimetype, page_id, file_id, url, file_grp
             'local_filename',
         ]))
 @click.option('--download', is_flag=True, help="Download found files to workspace and change location in METS file ")
+@click.option('--wait', type=int, default=0, help="Wait this many seconds between download requests")
 @pass_workspace
-def workspace_find(ctx, file_grp, mimetype, page_id, file_id, output_field, download):
+def workspace_find(ctx, file_grp, mimetype, page_id, file_id, output_field, download, wait):
     """
     Find files.
 
@@ -367,6 +369,8 @@ def workspace_find(ctx, file_grp, mimetype, page_id, file_id, output_field, down
         if download and not f.local_filename:
             workspace.download_file(f)
             modified_mets = True
+            if wait:
+                time.sleep(wait)
         ret.append([f.ID if field == 'pageId' else getattr(f, field) or ''
                     for field in output_field])
     if modified_mets:

ocrd/ocrd/lib.bash

@@ -156,6 +156,18 @@ ocrd__parse_argv () {
         ocrd__raise "Provide --output-file-grp/-O explicitly!"
     fi
 
+    # check fileGrps
+    local _valopts=( --workspace "${ocrd__argv[working_dir]}" )
+    if [[ ${ocrd__argv[overwrite]} = true ]]; then
+        _valopts+=( --overwrite )
+    fi
+    if [[ -n "${ocrd__argv[page_id]:-}" ]]; then
+        _valopts+=( --page-id "${ocrd__argv[page_id]}" )
+    fi
+    _valopts+=( "${OCRD_TOOL_NAME#ocrd-} -I ${ocrd__argv[input_file_grp]} -O ${ocrd__argv[output_file_grp]}" )
+    ocrd validate tasks "${_valopts[@]}" || exit $?
+
+    # check parameters
     local params_parsed retval
     params_parsed="$(ocrd ocrd-tool "$OCRD_TOOL_JSON" tool $OCRD_TOOL_NAME parse-params "${__parameters[@]}" "${__parameter_overrides[@]}")" || {
         retval=$?

ocrd/ocrd/processor/builtin/dummy_processor.py

@@ -21,7 +21,7 @@
 
 class DummyProcessor(Processor):
     """
-    Bare-bones processor that copies mets:file from input group to output group.
+    Bare-bones processor that only copies mets:file from input group to output group.
     """
 
     def process(self):
@@ -34,6 +34,7 @@ def process(self):
             ext = MIME_TO_EXT.get(input_file.mimetype, '')
             local_filename = join(self.output_file_grp, file_id + ext)
             pcgts = page_from_file(self.workspace.download_file(input_file))
+            pcgts.set_pcGtsId(file_id)
             self.add_metadata(pcgts)
             LOG.info("cp %s %s # %s -> %s", input_file.url, local_filename, input_file.ID, file_id)
             if input_file.mimetype == MIMETYPE_PAGE:
@@ -75,7 +76,7 @@ def process(self):
 
     def __init__(self, *args, **kwargs):
         kwargs['ocrd_tool'] = OCRD_TOOL
-        kwargs['version'] = '0.0.1'
+        kwargs['version'] = '0.0.2'
         super(DummyProcessor, self).__init__(*args, **kwargs)
 
 @click.command()

ocrd/ocrd/processor/helpers.py

@@ -88,20 +88,25 @@ def run_processor(
     processor.process()
     t1_wall = perf_counter() - t0_wall
     t1_cpu = process_time() - t0_cpu
-    logProfile.info("Executing processor '%s' took %fs (wall) %fs (CPU)( [--input-file-grp='%s' --output-file-grp='%s' --parameter='%s']" % (
+    logProfile.info("Executing processor '%s' took %fs (wall) %fs (CPU)( [--input-file-grp='%s' --output-file-grp='%s' --parameter='%s' --page-id='%s']" % (
         ocrd_tool['executable'],
         t1_wall,
         t1_cpu,
-        input_file_grp if input_file_grp else '',
-        output_file_grp if output_file_grp else '',
-        json.dumps(parameter) if parameter else {}
+        input_file_grp or '',
+        output_file_grp or '',
+        json.dumps(parameter) or '',
+        page_id or ''
     ))
     workspace.mets.add_agent(
         name=name,
         _type='OTHER',
         othertype='SOFTWARE',
         role='OTHER',
-        otherrole=otherrole
+        otherrole=otherrole,
+        notes=[({'option': 'input-file-grp'}, input_file_grp or ''),
+               ({'option': 'output-file-grp'}, output_file_grp or ''),
+               ({'option': 'parameter'}, json.dumps(parameter or '')),
+               ({'option': 'page-id'}, page_id or '')]
     )
     workspace.save_mets()
     return processor

ocrd_models/ocrd_models/constants.py

@@ -17,6 +17,7 @@
     'TAG_METS_FLOCAT',
     'TAG_METS_METSHDR',
     'TAG_METS_NAME',
+    'TAG_METS_NOTE',
     'TAG_METS_STRUCTMAP',
     'TAG_MODS_IDENTIFIER',
     'TAG_PAGE_ALTERNATIVEIMAGE',
@@ -39,6 +40,7 @@
     'xlink': "http://www.w3.org/1999/xlink",
     'page': "http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15",
     'xsl': 'http://www.w3.org/1999/XSL/Transform#',
+    'ocrd': 'https://ocr-d.de',
 }
 
 # pylint: disable=bad-whitespace
@@ -51,6 +53,7 @@
 TAG_METS_FLOCAT           = '{%s}FLocat' % NAMESPACES['mets']
 TAG_METS_METSHDR          = '{%s}metsHdr' % NAMESPACES['mets']
 TAG_METS_NAME             = '{%s}name' % NAMESPACES['mets']
+TAG_METS_NOTE             = '{%s}note' % NAMESPACES['mets']
 TAG_METS_STRUCTMAP        = '{%s}structMap' % NAMESPACES['mets']
 
 TAG_MODS_IDENTIFIER       = '{%s}identifier' % NAMESPACES['mods']

ocrd_models/ocrd_models/ocrd_agent.py

@@ -2,7 +2,7 @@
 API to ``mets:agent``
 """
 #  import os
-from .constants import NAMESPACES as NS, TAG_METS_AGENT, TAG_METS_NAME
+from .constants import NAMESPACES as NS, TAG_METS_AGENT, TAG_METS_NAME, TAG_METS_NOTE
 from .ocrd_xml_base import ET
 
 class OcrdAgent():
@@ -20,7 +20,8 @@ class OcrdAgent():
     #      #  version = name_parts[1][1:]     # v0.0.1 => 0.0.1
     #      return OcrdAgent(el, name, role, _type, otherrole)
 
-    def __init__(self, el=None, name=None, _type=None, othertype=None, role=None, otherrole=None):
+    def __init__(self, el=None, name=None, _type=None, othertype=None, role=None, otherrole=None,
+                 notes=None):
         """
         Args:
             el (LxmlElement):
@@ -29,6 +30,7 @@ def __init__(self, el=None, name=None, _type=None, othertype=None, role=None, ot
             othertype (string):
             role (string):
             otherrole (string):
+            notes (dict):
         """
         if el is None:
             el = ET.Element(TAG_METS_AGENT)
@@ -38,6 +40,7 @@ def __init__(self, el=None, name=None, _type=None, othertype=None, role=None, ot
         self.othertype = othertype
         self.role = role
         self.otherrole = otherrole
+        self.notes = notes
 
     def __str__(self):
         """
@@ -105,7 +108,7 @@ def otherrole(self):
     @otherrole.setter
     def otherrole(self, otherrole):
         """
-        Get the ``OTHERROLE`` attribute value.
+        Set the ``OTHERROLE`` attribute value.
         """
         if otherrole is not None:
             self._el.set('ROLE', 'OTHER')
@@ -116,17 +119,44 @@ def name(self):
         """
         Get the ``mets:name`` element value.
         """
-        el_name = self._el.find('mets:name', NS)
+        el_name = self._el.find(TAG_METS_NAME)
         if el_name is not None:
             return el_name.text
 
     @name.setter
     def name(self, name):
         """
-        Get the ``mets:name`` element value.
+        Set the ``mets:name`` element value.
         """
         if name is not None:
-            el_name = self._el.find('mets:name', NS)
+            el_name = self._el.find(TAG_METS_NAME)
             if el_name is None:
                 el_name = ET.SubElement(self._el, TAG_METS_NAME)
             el_name.text = name
+
+    @property
+    def notes(self):
+        """
+        Get the ``mets:note`` element values (as tuples of attributes and text).
+        """
+        el_notes = self._el.findall(TAG_METS_NOTE)
+        if el_notes is not None:
+            return [(note.attrib, note.text)
+                    for note in el_notes]
+
+    @notes.setter
+    def notes(self, notes):
+        """
+        Set the ``mets:note`` element values.
+        """
+        el_notes = self._el.findall(TAG_METS_NOTE)
+        if el_notes:
+            for el_note in el_notes:
+                self._el.remove(el_note)
+        if notes:
+            for note in notes:
+                el_note = ET.SubElement(self._el, TAG_METS_NOTE, nsmap={'ocrd': NS['ocrd']})
+                attrib, text = note
+                el_note.text = text
+                for name, value in attrib.items():
+                    el_note.set('{%s}' % NS["ocrd"] + name, value)

ocrd_utils/setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name='ocrd_utils',
-    version='2.27.0',
+    version='2.28.0',
     description='OCR-D framework - shared code, helpers, constants',
     long_description=open('README.md').read(),
     long_description_content_type='text/markdown',

tests/model/test_ocrd_page.py

@@ -8,6 +8,8 @@
     PageType,
     TextRegionType,
     TextLineType,
+    TextEquivType,
+    TextStyleType,
     OrderedGroupIndexedType,
     UnorderedGroupIndexedType,
     ReadingOrderType,
@@ -341,5 +343,33 @@ def test_id(self):
         assert pcgts.id == 'PAGE_0017_PAGE'
         assert pcgts.get_Page().id == 'OCR-D-IMG/INPUT_0017.tif'
 
+    def test_style_inheritance_from_scratch(self):
+        """https://github.com/OCR-D/core/pull/700"""
+        pcgts = PcGtsType(pcGtsId="foo")
+        page = PageType()
+        region = TextRegionType()
+        textstyle = TextStyleType(fontFamily='Schwabacher')
+        textline = TextLineType(TextEquiv=[TextEquivType(Unicode='foo')])
+
+        pcgts.set_Page(page)
+        page.add_TextRegion(region)
+        region.add_TextLine(textline)
+
+        assert not textline.get_TextStyle()
+        assert not region.get_TextStyle()
+
+        region.set_TextStyle(textstyle)
+
+        # TODO doesn't work that way
+        # assert region.get_TextStyle() == textstyle
+        # assert textline.get_TextStyle() == textstyle
+
+        pcgts = parseString(to_xml(pcgts, skip_declaration=True))
+        region = pcgts.get_Page().get_TextRegion()[0]
+        textline = region.get_TextLine()[0]
+        assert region.get_TextStyle().get_fontFamily() == 'Schwabacher'
+        assert textline.get_TextStyle().get_fontFamily() == 'Schwabacher'
+
+
 if __name__ == '__main__':
     main(__file__)