ocrd_mets: add get_physical_pages(for_pageIds=...)

ocrd/ocrd/cli/workspace.py

@@ -370,21 +370,22 @@ def workspace_cli_bulk_add(ctx, regex, mimetype, page_id, file_id, url, file_grp
 @workspace_cli.command('find')
 @mets_find_options
 @click.option('-k', '--output-field', help="Output field. Repeat for multiple fields, will be joined with tab",
-        default=['url'],
-        multiple=True,
-        type=click.Choice([
-            'url',
-            'mimetype',
-            'page_id',
-            'pageId',
-            'file_id',
-            'ID',
-            'file_grp',
-            'fileGrp',
-            'basename',
-            'basename_without_extension',
-            'local_filename',
-        ]))
+              default=['url'],
+              show_default=True,
+              multiple=True,
+              type=click.Choice([
+                  'url',
+                  'mimetype',
+                  'page_id',
+                  'pageId',
+                  'file_id',
+                  'ID',
+                  'file_grp',
+                  'fileGrp',
+                  'basename',
+                  'basename_without_extension',
+                  '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
@@ -535,13 +536,41 @@ def list_groups(ctx):
 # ----------------------------------------------------------------------
 
 @workspace_cli.command('list-page')
+@click.option('-g', '--page-id', help="Page ID", metavar='FILTER')
+@click.option('-k', '--output-field', help="Output field. Repeat for multiple fields, will be joined with tab",
+              default=['ID'],
+              show_default=True,
+              multiple=True,
+              type=click.Choice([
+                  'ID',
+                  'ORDER',
+                  'ORDERLABEL',
+                  'LABEL',
+              ]))
 @pass_workspace
-def list_pages(ctx):
+def list_pages(ctx, page_id, output_field):
     """
     List physical page IDs
+
+    (If any ``FILTER`` starts with ``//``, then its remainder
+     will be interpreted as a regular expression.)
     """
     workspace = Workspace(ctx.resolver, directory=ctx.directory, mets_basename=ctx.mets_basename)
-    print("\n".join(workspace.mets.physical_pages))
+    if page_id is None:
+        pages = workspace.mets.physical_pages
+    else:
+        pages = workspace.mets.get_physical_pages(for_pageIds=page_id)
+    if output_field == ['ID']:
+        print("\n".join(pages))
+    else:
+        labels = workspace.mets.physical_pages_labels
+        def field2label(page, field):
+            if field == 'ID':
+                return page
+            return labels[page][['ORDER', 'ORDERLABEL', 'LABEL'].index(field)]
+        for page in pages:
+            print("\t".join(field2label(page, field) or ''
+                            for field in output_field))
 
 # ----------------------------------------------------------------------
 # ocrd workspace get-id

ocrd_models/ocrd_models/ocrd_mets.py

@@ -571,15 +571,40 @@ def physical_pages(self):
             'mets:structMap[@TYPE="PHYSICAL"]/mets:div[@TYPE="physSequence"]/mets:div[@TYPE="page"]/@ID',
             namespaces=NS)
 
-    def get_physical_pages(self, for_fileIds=None):
+    def get_physical_pages(self, for_fileIds=None, for_pageIds=None):
         """
         List all page IDs (the ``@ID`` of each physical ``mets:structMap`` ``mets:div``),
-        optionally for a subset of ``mets:file`` ``@ID`` :py:attr:`for_fileIds`.
+        optionally for a subset of ``mets:file`` ``@ID`` :py:attr:`for_fileIds`,
+        or for a subset selector expression (comma-separated, range, and/or regex) :py:attr:`for_pageIds`.
         """
-        if for_fileIds is None:
+        if for_fileIds is None and for_pageIds is None:
             return self.physical_pages
+        if for_pageIds is not None:
+            ret = []
+            pageId_patterns = []
+            for pageId_token in re.split(r',', for_pageIds):
+                if pageId_token.startswith(REGEX_PREFIX):
+                    pageId_patterns.append(re.compile(pageId_token[REGEX_PREFIX_LEN:]))
+                elif '..' in pageId_token:
+                    pageId_patterns += generate_range(*pageId_token.split('..', 1))
+                else:
+                    pageId_patterns += [pageId_token]
+            if self._cache_flag:
+                for page_id in self._page_cache.keys():
+                    if page_id in pageId_patterns or \
+                        any([isinstance(p, typing.Pattern) and p.fullmatch(page_id) for p in pageId_patterns]):
+                        ret.append(page_id)
+            else:
+                for page in self._tree.getroot().xpath(
+                    'mets:structMap[@TYPE="PHYSICAL"]/mets:div[@TYPE="physSequence"]/mets:div[@TYPE="page"]',
+                        namespaces=NS):
+                    page_id = page.get('ID')
+                    if page_id in pageId_patterns or \
+                        any([isinstance(p, typing.Pattern) and p.fullmatch(page_id) for p in pageId_patterns]):
+                        ret.append(page_id)
+            return ret
+
         ret = [None] * len(for_fileIds)
-
         if self._cache_flag:
             for pageId in self._fptr_cache.keys():
                 for fptr in self._fptr_cache[pageId].keys():
@@ -728,6 +753,18 @@ def remove_physical_page_fptr(self, fileId):
             mets_div.remove(mets_fptr)
         return ret
 
+    @property
+    def physical_pages_labels(self):
+        """
+        Map all page IDs (the ``@ID`` of each physical ``mets:structMap`` ``mets:div``) to their
+        ``@ORDER``, ``@ORDERLABEL`` and ``@LABEL`` attributes, if any.
+        """
+        divs = self._tree.getroot().xpath(
+            'mets:structMap[@TYPE="PHYSICAL"]/mets:div[@TYPE="physSequence"]/mets:div[@TYPE="page"]',
+            namespaces=NS)
+        return {div.get('ID'): (div.get('ORDER', None), div.get('ORDERLABEL', None), div.get('LABEL', None))
+                for div in divs}
+
     def merge(self, other_mets, force=False, fileGrp_mapping=None, fileId_mapping=None, pageId_mapping=None, after_add_cb=None, **kwargs):
         """
         Add all files from other_mets.