epub converter: add internal links processing2

src/epub_converter.py

@@ -113,9 +113,16 @@ class EpubBook:
         self.book_api_wrapper.set_process_status()
         self.logger_object.log('Beginning of processing json output.')
 
-        json_converter = EpubPostprocessor(self.epub_path, access=self.access, logger=self.logger_object)
-        content_dict = json_converter.convert_to_dict()
-        self.book_api_wrapper.set_generate_status()
-        self.write_to_json(content_dict)
-        self.send_json_content(content_dict)
-        self.logger_object.log(f'End of the conversion to LawCarta format. Check {self.output_path}.')
+        try:
+            json_converter = EpubPostprocessor(self.epub_path, access=self.access, logger=self.logger_object)
+            content_dict = json_converter.convert_to_dict()
+            self.book_api_wrapper.set_generate_status()
+            self.write_to_json(content_dict)
+            self.send_json_content(content_dict)
+            self.logger_object.log(f'End of the conversion to LawCarta format. Check {self.output_path}.')
+
+        except Exception as exc:
+            self.logger_object.log('Error has occurred while conversion.', logging.ERROR)
+            self.logger_object.log_error_to_main_log(str(exc))
+            self.book_api_wrapper.set_error_status()
+            raise exc

src/epub_postprocessor.py

@@ -25,7 +25,7 @@ class EpubPostprocessor:
         self.access = access
         self.logger: BookLogger = logger
         self.ebooklib_book = epub.read_epub(file)  # todo: log error from ebooklib
-        self.internal_links_found = 0
+        self.internal_anchors = set()
         self.logger.log('Image processing.')
         self.href2img_bytes = {}
         self.old_image_path2_aws_path = {}
@@ -196,7 +196,7 @@ class EpubPostprocessor:
 
     @staticmethod
     def _create_unique_id(href, id_):
-        return re.sub(r'([^\w\s])|_|-', '', href) + id_
+        return re.sub(r'([^\w\s])|_|-', '', href) + re.sub(r'[_-]', '0', id_)
 
     def process_internal_links(self):
         # rebuild ids to be unique in all documents
@@ -235,15 +235,29 @@ class EpubPostprocessor:
                 anchor_tags = anchor_soup.find_all(attrs={'id': new_id})
                 if anchor_tags:
                     if len(anchor_tags) > 1:
-                        self.logger.log(f'Warning in {href}: multiple anchors: {anchor_tags} found.'
+                        self.logger.log(f'Warning in {href}: multiple anchors: {len(anchor_tags)} found.'
                                         f' While processing {internal_link_tag}')
 
                     anchor_tag = anchor_tags[0]
                     # if anchor is found we could add placeholder for link creation on server side.
                     internal_link_tag.attrs['placeholder'] = '{{tempStyleToAnchor-' + new_id + '}}'
-                    anchor_tag.attrs['class'] = 'link-anchor'
+                    # create span to have cyclic links, link has 1 type of class, anchor another
+                    if anchor_tag.name == 'a':
+                        new_anchor_already_created = soup.find_all('span',
+                                                                   attrs={'class': 'link-anchor',
+                                                                          'id': anchor_tag.attrs['id']})
+                        if not new_anchor_already_created:
+                            new_anchor_span = soup.new_tag("span")
+                            new_anchor_span.attrs['id'] = anchor_tag.attrs['id']
+                            new_anchor_span.attrs['class'] = 'link-anchor'
+                            anchor_tag.insert(0, new_anchor_span)
+                            del anchor_tag.attrs['id']
+                            self.internal_anchors.add(new_anchor_span)
+                    else:
+                        self.internal_anchors.add(anchor_tag)
+                        anchor_tag.attrs['class'] = 'link-anchor'
+
                     del internal_link_tag.attrs['href']
-                    self.internal_links_found += 1
 
                 else:
                     internal_link_tag.attrs['converter-mark'] = 'bad-link'
@@ -325,7 +339,7 @@ class EpubPostprocessor:
             top_level_chapters.append(chapter)
 
         top_level_dict_chapters = [x.to_dict() for x in top_level_chapters]
-        self.logger.log(f'Internal links found: {self.internal_links_found}.')
+        self.logger.log(f'Anchors found: {len(self.internal_anchors)}.')
         self.logger.log('End conversion.')
 
         return {

src/html_epub_preprocessor.py

@@ -241,6 +241,8 @@ def unwrap_structural_tags(body_tag):
     # should be before other tags processing, not to remove converter empty tags with id
     # not all cases, if span has <p>s and NavigableString, it won't unwrap
     for s in body_tag.find_all("span"):
+        if s.attrs.get('epub:type') == 'pagebreak':
+            continue
         if s.contents:
             is_not_struct_tag = [child.name not in structural_tags_names for child in s.contents]
             if all(is_not_struct_tag):