fnp
/
librarian.git
/ blobdiff
commit
grep
author
committer
pickaxe
?
search:
re
summary
|
shortlog
|
log
|
commit
|
commitdiff
|
tree
raw
|
inline
| side by side
swap endlines inside stanzas only
[librarian.git]
/
librarian
/
pdf.py
diff --git
a/librarian/pdf.py
b/librarian/pdf.py
index
5debb68
..
2d45372
100644
(file)
--- a/
librarian/pdf.py
+++ b/
librarian/pdf.py
@@
-36,7
+36,7
@@
STYLESHEETS = {
}
}
-def insert_tags(doc, split_re, tagname):
+def insert_tags(doc, split_re, tagname
, exclude=None
):
""" inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree
>>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>');
""" inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree
>>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>');
@@
-46,6
+46,8
@@
def insert_tags(doc, split_re, tagname):
"""
for elem in doc.iter(tag=etree.Element):
"""
for elem in doc.iter(tag=etree.Element):
+ if exclude and elem.tag in exclude:
+ continue
if elem.text:
chunks = split_re.split(elem.text)
while len(chunks) > 1:
if elem.text:
chunks = split_re.split(elem.text)
while len(chunks) > 1:
@@
-67,13
+69,17
@@
def insert_tags(doc, split_re, tagname):
def substitute_hyphens(doc):
insert_tags(doc,
re.compile("(?<=[^-\s])-(?=[^-\s])"),
def substitute_hyphens(doc):
insert_tags(doc,
re.compile("(?<=[^-\s])-(?=[^-\s])"),
- "dywiz")
+ "dywiz",
+ exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+ )
def fix_hanging(doc):
insert_tags(doc,
re.compile("(?<=\s\w)\s+"),
def fix_hanging(doc):
insert_tags(doc,
re.compile("(?<=\s\w)\s+"),
- "nbsp")
+ "nbsp",
+ exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+ )
def move_motifs_inside(doc):
def move_motifs_inside(doc):
@@
-277,8
+283,7
@@
def load_including_children(provider, slug=None, uri=None, file_path=None):
raise ValueError('Neither slug, URI nor file path provided for a book.')
document = WLDocument.from_file(f, True,
raise ValueError('Neither slug, URI nor file path provided for a book.')
document = WLDocument.from_file(f, True,
- parse_dublincore=True,
- preserve_lines=False)
+ parse_dublincore=True)
f.close()
f.close()