build page by api instead template

OCR-D · Apr 24, 2018 · d35559b · d35559b
1 parent 03a8c31
commit d35559b
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 23 deletions.
diff --git a/ocrd/constants.py b/ocrd/constants.py
@@ -13,18 +13,6 @@
     'xsl': 'http://www.w3.org/1999/XSL/Transform#',
 }
 
-PAGE_XML_EMPTY = '''<?xml version="1.0" encoding="UTF-8"?>
-<PcGts xmlns="http://schema.primaresearch.org/PAGE/gts/pagecontent/2017-07-15" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://schema.primaresearch.org/PAGE/gts/pagecontent/2017-07-15 http://schema.primaresearch.org/PAGE/gts/pagecontent/2017-07-15/pagecontent.xsd">
-    <Metadata>
-        <Creator>ocrd/core</Creator>
-        <Created>2018-01-01T00:00:00Z</Created>
-        <LastChange>2018-01-01T00:00:00Z</LastChange>
-    </Metadata>
-    <Page>
-    </Page>
-</PcGts>
-'''
-
 MIMETYPE_PAGE = 'text/page+xml'
 
 DEFAULT_UPLOAD_FOLDER = '/tmp/uploads-pyocrd'

diff --git a/ocrd/model/ocrd_page.py b/ocrd/model/ocrd_page.py
@@ -1,18 +1,22 @@
 from io import StringIO
+from datetime import datetime
 
 # pylint: disable=unused-import
 from ocrd.model.ocrd_page_generateds import (
     parse,
     parseString,
     CoordsType,
     OrderedGroupType,
+    PcGtsType,
+    PageType,
+    MetadataType,
     ReadingOrderType,
     RegionRefIndexedType,
     TextEquivType,
     TextRegionType,
     TextLineType,
 )
-from ocrd.constants import PAGE_XML_EMPTY, NAMESPACES
+from ocrd.constants import NAMESPACES, VERSION
 from ocrd.model.ocrd_exif import OcrdExif
 
 def to_xml(el):
@@ -24,21 +28,27 @@ def page_from_image(input_file):
     if input_file.local_filename is None:
         raise Exception("input_file must have 'local_filename' property")
     exif = OcrdExif.from_filename(input_file.local_filename)
-    content = PAGE_XML_EMPTY.replace('<Page>', '<Page imageWidth="%d" imageHeight="%i" imageFilename="%s">' % (
-        exif.width,
-        exif.height,
-        input_file.url
-    ))
-    return content
+    now = datetime.now()
+    return PcGtsType(
+        Metadata=MetadataType(
+            Creator="OCR-D/core %s" % VERSION,
+            Created=now,
+            LastChange=now
+        ),
+        Page=PageType(
+            imageWidth=exif.width,
+            imageHeight=exif.height,
+            imageFilename=input_file.local_filename
+        )
+    )
 
 def from_file(input_file):
     """
     Create a new PAGE-XML from a METS file representing a PAGE-XML or an image.
     """
     #  print("PARSING PARSING '%s'" % input_file)
     if input_file.mimetype.startswith('image'):
-        content = page_from_image(input_file)
-        return parseString(content.encode('utf-8'), silence=True)
+        return page_from_image(input_file)
     elif input_file.mimetype == 'text/page+xml':
         return parse(input_file.local_filename, silence=True)
     else:

diff --git a/ocrd/model/ocrd_xml_base.py b/ocrd/model/ocrd_xml_base.py
@@ -1,3 +1,5 @@
+from lxml import etree as ET
+
 from ocrd.constants import (
     NAMESPACES,
     TAG_PAGE_COORDS
@@ -8,8 +10,6 @@
     coordinate_string_from_xywh
 )
 
-from lxml import etree as ET
-
 for curie in NAMESPACES:
     ET.register_namespace(curie, NAMESPACES[curie])