rel

[librarian.git] / src / librarian / elements / base.py

diff --git a/src/librarian/elements/base.py b/src/librarian/elements/base.py
index 646067e..b0d16ed 100644 (file)
--- a/src/librarian/elements/base.py
+++ b/src/librarian/elements/base.py
@@ -1,3 +1,6 @@
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Wolne Lektury. See NOTICE for more information.
+#
 import copy
 import re
 from lxml import etree
@@ -39,11 +42,11 @@ class WLElement(etree.ElementBase):
     STRIP = False
 
     text_substitutions = [
-        (u'---', u'—'),
-        (u'--', u'–'),
-        #(u'...', u'…'),  # Temporary turnoff for epub
-        (u',,', u'„'),
-        (u'"', u'”'),
+        ('---', '—'),
+        ('--', '–'),
+        #('...', '…'),  # Temporary turnoff for epub
+        (',,', '„'),
+        ('"', '”'),
         ('\ufeff', ''),
 
         ("'", "\u2019"),    # This was enabled for epub.
@@ -113,11 +116,11 @@ class WLElement(etree.ElementBase):
             newt = ''
             wlist = re.compile(r'\w+|[^\w]', re.UNICODE).findall(text)
             for w in wlist:
-                newt += builder.hyphenator.inserted(w, u'\u00AD')
+                newt += builder.hyphenator.inserted(w, '\u00AD')
             text = newt
 
         if builder.orphans:
-            text = re.sub(r'(?<=\s\w)\s+', u'\u00A0', text)
+            text = re.sub(r'(?<=\s\w)\s+', '\u00A0', text)
 
         return text
 
@@ -133,6 +136,8 @@ class WLElement(etree.ElementBase):
         for i, child in enumerate(self):
             if isinstance(child, WLElement):
                 getattr(child, build_method)(builder)
+            elif getattr(builder, 'debug') and child.tag is etree.Comment:
+                builder.process_comment(child)
             if self.CAN_HAVE_TEXT and child.tail:
                 text = self.normalize_text(child.tail, builder)
                 if self.STRIP and i == child_count - 1:
@@ -195,7 +200,7 @@ class WLElement(etree.ElementBase):
         # TEMPORARY
         self.CAN_HAVE_TEXT = True
         self.STRIP = False
-       
+
         start_chunk = self.EPUB_START_CHUNK and isinstance(self.getparent(), Master)
 
         if start_chunk:
@@ -217,6 +222,11 @@ class WLElement(etree.ElementBase):
             attr = self.get_epub_attr(builder)
             if fragment:
                 attr['id'] = fragment
+            if builder.debug:
+                chunkno, sourceline = 0, self.sourceline
+                if builder.splits:
+                    chunkno, sourceline = len(builder.splits), sourceline - builder.splits[-1]
+                attr['data-debug'] = f'{chunkno}:{sourceline}'
             builder.start_element(
                 self.EPUB_TAG,
                 attr
@@ -278,8 +288,6 @@ class WLElement(etree.ElementBase):
             # do we dare go up?
             parent = self.getparent()
             if parent is not None and parent.CAN_HAVE_TEXT:
-                print(etree.tostring(self, encoding='unicode'))
-                assert False
                 words, parsnip = parent.snip(words, before=self)
                 return words, parsnip[:-1] + snippet + parsnip[-1:]