Merge branch 'master' of github.com:mikegerber/my_ocrd_workflow

Dockerfile

@@ -70,10 +70,13 @@ COPY data/textline_detection /var/lib/textline_detection
 
 
 # Install requirements
+# Using pipdeptree here to get more info than from pip3 check
 COPY requirements.txt /tmp/
 RUN pip3 install --no-cache-dir --upgrade pip && \
     pip3 install --no-cache-dir -r /tmp/requirements.txt && \
-    pip3 check
+    pip3 install --no-cache-dir pipdeptree && \
+    pipdeptree -w fail
+
 
 COPY my_ocrd_workflow /usr/bin/
 COPY xsd/*            /usr/share/xml/

README.md

@@ -72,9 +72,6 @@ This produces a workspace directory `PPN77164308X` with the OCR results in it;
 the results are viewable as explained above.
 
 ppn2ocr requires a working Docker setup and properly set up environment
-variables for the proxy configuration. At SBB, this means:
-~~~
-export HTTP_PROXY=http://http-proxy.sbb.spk-berlin.de:3128/
-export HTTPS_PROXY=$HTTP_PROXY; export http_proxy=$HTTP_PROXY; export https_proxy=$HTTP_PROXY
-export no_proxy=localhost,digital.staatsbibliothek-berlin.de,content.staatsbibliothek-berlin.de
-~~~
+variables for the proxy configuration. At SBB, this following
+`howto/docker-proxy.md` and `howto/proxy-settings-for-shell+python.md`
+(in qurator's mono-repo).

check-FULLTEXT-Page-dimensions-vs-BEST-dimensions.py

@@ -1,47 +0,0 @@
-"""Check FULLTEXT ALTO page dimensions against BEST image dimensions"""
-
-import PIL.Image
-import sys
-from ocrd.workspace import Workspace
-from ocrd.resolver import Resolver
-from lxml import etree as ET
-
-
-def alto_namespace(tree):
-    """
-    Return the ALTO namespace used in the given ElementTree.
-
-    This relies on the assumption that, in any given ALTO file, the root
-    element has the local name "alto". We do not check if the files uses any
-    valid ALTO namespace.
-    """
-    root_name = ET.QName(tree.getroot().tag)
-    if root_name.localname == 'alto':
-        return root_name.namespace
-    else:
-        raise ValueError('Not an ALTO tree')
-
-
-exit_code = 0
-workspace = Workspace(Resolver(), '.')
-
-for n, page_id in enumerate(workspace.mets.physical_pages):
-    gt_file = workspace.mets.find_files(fileGrp='FULLTEXT', pageId=page_id)[0]
-    img_file = workspace.mets.find_files(fileGrp='BEST', pageId=page_id)[0]
-    gt_file = workspace.download_file(gt_file)
-    img_file = workspace.download_file(img_file)
-
-    tree = ET.parse(gt_file.local_filename)
-    nsmap = {'alto': alto_namespace(tree)}
-    alto_page = tree.find('//alto:Page', namespaces=nsmap)  # one page assumed
-    gt_size = int(alto_page.attrib['WIDTH']), int(alto_page.attrib['HEIGHT'])
-
-    img_size = PIL.Image.open(img_file.local_filename).size
-
-    if gt_size == img_size:
-        print('OK', page_id)
-    else:
-        print('ERR', page_id, gt_size, '!=', img_size)
-        exit_code = 1
-
-sys.exit(exit_code)

my_ocrd_workflow

@@ -134,6 +134,8 @@ page_upgrade_to_2019() {
 
 
 main() {
+  do_validate
+
   do_binarization
   do_validate
 

ppn2ocr

@@ -74,9 +74,10 @@ def make_workspace(ppn, workspace):
     mets = oai_mets(ppn)
 
     # XXX
-    # Delete PRESENTATION file group
-    # (local file:/// links, not handled well by "ocrd workspace")
+    # Delete PRESENTATION + LOCAL file groups
+    # (local file:/// or file:/ links, not handled well by "ocrd workspace")
     remove_file_grp(mets, 'PRESENTATION')
+    remove_file_grp(mets, 'LOCAL')
 
 
     # Duplicate DEFAULT file group into a new file group BEST