Localization works.

gragir/book.py

@@ -9,7 +9,7 @@ class Item(object):
         self.content_type = content_type
         self.payload = payload
         self.needed_by = set()
-        self.needed_by_elem = set()
+        self.needed_by_elem = []
         self.needs = set()
         self.soup = None
 
@@ -23,7 +23,7 @@ class Item(object):
         if directory[-1] != '/':
             directory += '/'
         file_name = directory + self.url
-        logger.info("Saved {}".format(file_name))
+        logger.info("Saveing {}".format(file_name))
         #
         #   Ensure directory exist.
         #
@@ -44,6 +44,13 @@ class Item(object):
         defrag,_ =urllib.parse.urldefrag(link)
         return urllib.parse.urljoin(self.url, defrag)
 
+    def refersTo(self, ref_item):
+        self.needs.add(ref_item)
+
+    def referencedBy(self, item, element):
+        self.needed_by.add(item)
+        self.needed_by_elem.append(element)
+
 
 class Book(object):
 
@@ -61,15 +68,6 @@ class Book(object):
         for item in self.content.values():
             item.save_file(directory)
 
-    def insertDependency(self, item, element, url):
-        logger = logging.getLogger(__name__)
-        if url in self.content:
-            item.needs.add(self.content[url])
-            self.content[url].needed_by.add(item)
-            self.content[url].needed_by_elem.add(element)
-        elif url:
-            logger.info("   refered but no item exist: {}".format(url))
-
     def print(self):
         logger = logging.getLogger(__name__)
         logger.info("Book Structure:")

gragir/enrich_html.py

@@ -41,6 +41,7 @@ class EnrichHtml(object):
             if item.soup is not None:
                 logger.info("Create DAG {}".format(item.url))
                 cls.normalizeUrlAndSetDependecy(book, item, 'a', 'href')
+                cls.normalizeUrlAndSetDependecy(book, item, 'link', 'href')
                 cls.normalizeUrlAndSetDependecy(book, item, 'img', 'src')
                 cls.normalizeUrlAndSetDependecy(book, item, 'img', 'data-mfp-src')
 
@@ -53,8 +54,14 @@ class EnrichHtml(object):
             if not url:
                 continue
             normal_url = item.getAbsoluteUrl(url)
-            logger.info("   depends on: {}".format(normal_url))
+            logger.info("  refers to: {}".format(normal_url))
-            book.insertDependency(item, element, normal_url)
+            if normal_url in book.content:
+                ref_item = book.content[normal_url]
+                item.refersTo(ref_item)
+                ref_item.referencedBy(item, element)
+            else:
+                logger.info("  refered but no item exist: {}".format(url))
+                element[attr] = ''
 
 
     @classmethod
@@ -71,7 +78,7 @@ class EnrichHtml(object):
                     body.clear()
                     body.append(content)
                 else:
-                    logger.error("  No content found: {}".format(item.url))
+                    logger.warn("  No content found: {}".format(item.url))
                     remove.append(item)
 
         for item in remove:

gragir/parse_mhtml.py

@@ -14,16 +14,16 @@ def parseMht(mht, book):
 
     parts = mhtContent.get_payload()
     # Multiple parts, usually? If single 'str' part, then convert to a list.
-    if not type(parts) is list: 
+    if not type(parts) is list:
-        parts = [mhtContent] 
+        parts = [mhtContent]
 
     logger.info('   Number of parts: {}'.format(len(parts)))
 
     # Save all parts to files.
     for p in parts: # walk() for a tree, but I'm guessing MHT is never nested?
-            #??? cs = p.get_charset() # Expecting "utf-8" for root HTML, None for all other parts.						
+            #??? cs = p.get_charset() # Expecting "utf-8" for root HTML, None for all other parts.
             ct = p.get_content_type()
-                 # String coerced to lower case of the form maintype/subtype, else get_default_type().			
+                 # String coerced to lower case of the form maintype/subtype, else get_default_type().
             fp = p.get("content-location") or "index.html" # File path. Expecting root HTML is only part with no location.
 
             logger.info('       Content type: {}, Location: {}, Size: {}'
@@ -34,9 +34,9 @@ def parseMht(mht, book):
 
 def parseMhtFile(zip, mhtInfo, book):
     logger = logging.getLogger(__name__)
-    logger.info('Parsing {}, size: {}, csize: {} '
+    logger.info('Reading {}, size: {}, csize: {} '
                 .format(mhtInfo.filename,
-                        mhtInfo.file_size, 
+                        mhtInfo.file_size,
                         mhtInfo.compress_size))
 
     with zip.open(mhtInfo) as mht:

gragir/prepare_epub.py

@@ -86,9 +86,13 @@ class PrepareEpub(object):
         for ref_elem in item.needed_by_elem:
             if ref_elem.name == 'a':
                 _,fragment = urllib.parse.urldefrag(ref_elem['href'])
-                ref_elem['href'] = local_url + "#" + fragment
+                if fragment:
-            if ref_elem.name == 'img':
+                    ref_elem['href'] = local_url + "#" + fragment
+                else:
+                    ref_elem['href'] = local_url
+            elif ref_elem.name == 'img':
                 ref_elem['src'] = local_url
-            if ref_elem.name == 'img':
                 ref_elem['data-mfp-src'] = local_url
+            else:
+                logger.info("Renaming {} -> {}".format(item.url, local_url))
         item.url = local_url