Have the pageXML that is parsed available to the save methods, and be…

…tter iteration methods

page_xml/xmlPAGE.py

@@ -77,6 +77,8 @@ def parse(self):
         # --- save "namespace" base
         self.base = "".join([self.root.tag.rsplit("}", 1)[0], "}"])
 
+        ET.register_namespace("", self.XMLNS["xmlns"])
+
     def get_region(self, region_name):
         """
         get all regions in PAGE which match region_name
@@ -157,6 +159,9 @@ def get_polygons(self, element_name):
     def _iter_element(self, element):
         return self.root.iterfind("".join([".//", self.base, element]))
 
+    def _iter_subelement(self, element, subelement):
+        return element.iterfind("".join([".//", self.base, subelement]))
+
     def iter_class_coords(self, element, class_dict):
         for node in self._iter_element(element):
             element_type = self.get_region_type(node)
@@ -249,13 +254,13 @@ def write_transcriptions(self, out_dir):
 
     def new_page(self, name, rows, cols):
         """create a new PAGE xml"""
-        self.xml = ET.Element("PcGts")
-        self.xml.attrib = self.XMLNS
-        self.metadata = ET.SubElement(self.xml, "Metadata")
+        self.root = ET.Element("PcGts")
+        self.root.attrib = self.XMLNS
+        self.metadata = ET.SubElement(self.root, "Metadata")
         ET.SubElement(self.metadata, "Creator").text = self.creator
         ET.SubElement(self.metadata, "Created").text = datetime.datetime.today().strftime("%Y-%m-%dT%X")
         ET.SubElement(self.metadata, "LastChange").text = datetime.datetime.today().strftime("%Y-%m-%dT%X")
-        self.page = ET.SubElement(self.xml, "Page")
+        self.page = ET.SubElement(self.root, "Page")
         self.page.attrib = {
             "imageFilename": name,
             "imageWidth": cols,
@@ -329,8 +334,8 @@ def add_baseline(self, b_coords, parent):
 
     def save_xml(self):
         """write out XML file of current PAGE data"""
-        self._indent(self.xml)
-        tree = ET.ElementTree(self.xml)
+        self._indent(self.root)
+        tree = ET.ElementTree(self.root)
         with AtomicFileName(self.filepath) as path:
             tree.write(path, encoding="UTF-8", xml_declaration=True)
 

page_xml/xml_converters/xml_to_sem_seg.py

@@ -86,8 +86,6 @@ def build_text_line(self, page: PageData, out_size: tuple[int, int]):
             self.logger.warning(f"File {page.filepath} does not contains text line sem_seg")
         return sem_seg
 
-    # TOP BOTTOM
-
     def build_top_bottom(self, page: PageData, out_size: tuple[int, int]):
         """
         Create the sem_seg version of the top bottom