Add Hebrew support in pdfs.

[librarian.git] / src / librarian / html.py

diff --git a/src/librarian/html.py b/src/librarian/html.py
index c2f41c0..d6482dc 100644 (file)
--- a/src/librarian/html.py
+++ b/src/librarian/html.py
@@ -9,7 +9,7 @@ import urllib.parse
 import urllib.request
 
 from lxml import etree
 import urllib.request
 
 from lxml import etree

-from librarian import XHTMLNS, ParseError, OutputFile
+from librarian import XHTMLNS, DCNS, ParseError, OutputFile

 from librarian import functions
 from PIL import Image
 
 from librarian import functions
 from PIL import Image
 


@@ -39,7 +39,7 @@ def transform_abstrakt(abstrakt_element):
     style = etree.parse(style_filename)
     xml = etree.tostring(abstrakt_element, encoding='unicode')
     document = etree.parse(io.StringIO(
     style = etree.parse(style_filename)
     xml = etree.tostring(abstrakt_element, encoding='unicode')
     document = etree.parse(io.StringIO(

-        xml.replace('abstrakt', 'dlugi_cytat')
+        xml.replace('<abstrakt', '<dlugi_cytat').replace('</abstrakt', '</dlugi_cytat')

     ))  # HACK
     result = document.xslt(style)
     html = re.sub('<a name="sec[0-9]*"/>', '',
     ))  # HACK
     result = document.xslt(style)
     html = re.sub('<a name="sec[0-9]*"/>', '',


@@ -49,6 +49,10 @@ def transform_abstrakt(abstrakt_element):
 
 def add_image_sizes(tree, gallery_path, gallery_url, base_url):
     widths = [360, 600, 1200, 1800, 2400]
 
 def add_image_sizes(tree, gallery_path, gallery_url, base_url):
     widths = [360, 600, 1200, 1800, 2400]

+    try:
+        os.makedirs(gallery_path)
+    except:
+        pass

 
     for i, ilustr in enumerate(tree.findall('//ilustr')):
         rel_path = ilustr.attrib['src']
 
     for i, ilustr in enumerate(tree.findall('//ilustr')):
         rel_path = ilustr.attrib['src']


@@ -108,9 +112,14 @@ def transform(wldoc, stylesheet='legacy', options=None, flags=None, css=None, ga
             for flag in flags:
                 document.edoc.getroot().set(flag, 'yes')
 
             for flag in flags:
                 document.edoc.getroot().set(flag, 'yes')
 

+        ltag = document.edoc.find('//' + DCNS('language'))
+        lang = functions.lang_code_3to2(ltag.text) or 'pl'
+        document.edoc.getroot().set('lang', lang)
+

         document.clean_ed_note()
         document.clean_ed_note('abstrakt')
         document.fix_pa_akap()
         document.clean_ed_note()
         document.clean_ed_note('abstrakt')
         document.fix_pa_akap()

+        document.hebr_protect()

         
         if not options:
             options = {}
         
         if not options:
             options = {}


@@ -307,7 +316,8 @@ def any_ancestor(element, test):
 
 
 def add_anchors(root):
 
 
 def add_anchors(root):

-    counter = 1
+    link_prefix = "f"
+    counter = {"f": 1}

     visible_counter = 1
     for element in root.iterdescendants():
         def f(e):
     visible_counter = 1
     for element in root.iterdescendants():
         def f(e):


@@ -325,18 +335,21 @@ def add_anchors(root):
                 visible_counter = int(element.get('data-start'))
             except ValueError:
                 visible_counter = 1
                 visible_counter = int(element.get('data-start'))
             except ValueError:
                 visible_counter = 1

+            if element.get("data-link"):
+                link_prefix = element.get("data-link")
+                counter[link_prefix] = 1

 
         if any_ancestor(element, f):
             continue
 
         if element.tag == 'div' and 'verse' in element.get('class', ''):
             if visible_counter == 1 or visible_counter % 5 == 0:
 
         if any_ancestor(element, f):
             continue
 
         if element.tag == 'div' and 'verse' in element.get('class', ''):
             if visible_counter == 1 or visible_counter % 5 == 0:

-                add_anchor(element, "f%d" % counter, link_text=visible_counter)
-            counter += 1
+                add_anchor(element, "%s%d" % (link_prefix, counter[link_prefix]), link_text=visible_counter)
+            counter[link_prefix] += 1

             visible_counter += 1
         elif 'paragraph' in element.get('class', ''):
             visible_counter += 1
         elif 'paragraph' in element.get('class', ''):

-            add_anchor(element, "f%d" % counter, link_text=visible_counter)
-            counter += 1
+            add_anchor(element, "%s%d" % (link_prefix, counter[link_prefix]), link_text=visible_counter)
+            counter[link_prefix] += 1

             visible_counter += 1
 
 
             visible_counter += 1