def anchorNCXUpdates(data, originating_filename, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary
id_dict = {}
for i in range(0, len(keylist)):
id_dict[keylist[i]] = valuelist[i]
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
original_filename_with_relative_path = TEXT_FOLDER_NAME + "/" + originating_filename
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
if (parts is not None) and (len(parts) > 1) and (
parts[0] == original_filename_with_relative_path) and (
parts[1] != ""):
fragment_id = parts[1]
if fragment_id in id_dict:
attribute_value = TEXT_FOLDER_NAME + "/" + quoteurl(
id_dict[fragment_id]) + "#" + fragment_id
tag["src"] = attribute_value
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=" ")
return newdata
def performPageMapUpdates(data, currentdir, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary of xml_updates properly adjusted
updates = {}
for i in range(0, len(keylist)):
updates[ keylist[i] ] = "../" + valuelist[i]
xml_empty_tags = ["page"]
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=xml_empty_tags)
soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
for tag in soup.find_all(["page"]):
for att in ["href"]:
if att in tag.attrs :
ref = tag[att]
if ref.find(":") == -1 :
parts = ref.split('#')
url = parts[0]
fragment = ""
if len(parts) > 1:
fragment = parts[1]
bookrelpath = os.path.join(currentdir, unquoteurl(url))
bookrelpath = os.path.normpath(bookrelpath)
bookrelpath = bookrelpath.replace(os.sep, "/")
if bookrelpath in updates:
attribute_value = updates[bookrelpath]
if fragment != "":
attribute_value = attribute_value + "#" + fragment
attribute_value = quoteurl(attribute_value)
tag[att] = attribute_value
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ")
return newdata
def anchorNCXUpdatesAfterMerge(data, ncx_bookpath, sink_bookpath,
merged_bookpaths):
data = _remove_xml_header(data)
startdir = startingDir(ncx_bookpath)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
if parts is not None:
ahref = unquoteurl(parts[0])
target_bookpath = buildBookPath(ahref, startdir)
if target_bookpath in merged_bookpaths:
attribute_value = buildRelativePath(
ncx_bookpath, sink_bookpath)
if len(parts) > 1 and parts[1] != "":
attribute_value += "#" + parts[1]
tag["src"] = quoteurl(attribute_value)
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=" ")
return newdata
def performOPFSourceUpdates(data, newbkpath, oldbkpath, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary
updates = {}
for i in range(0, len(keylist)):
updates[keylist[i]] = valuelist[i]
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
for tag in soup.find_all(["link", "item", "reference", "site"]):
if "href" in tag.attrs:
href = tag["href"]
if href.find(":") == -1:
parts = href.split('#')
ahref = unquoteurl(parts[0])
fragment = ""
if len(parts) > 1:
fragment = parts[1]
oldtarget = buildBookPath(ahref, startingDir(oldbkpath))
newtarget = updates.get(oldtarget, oldtarget)
attribute_value = buildRelativePath(newbkpath, newtarget)
if fragment != "":
attribute_value = attribute_value + "#" + fragment
attribute_value = quoteurl(attribute_value)
tag["href"] = attribute_value
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=" ")
return newdata
def anchorNCXUpdatesAfterMerge(data, sink_filename, merged_names):
data = _remove_xml_header(data)
# build list of urls to rreplace
namelist = []
for fn in merged_names:
namelist.append(TEXT_FOLDER_NAME + "/" + fn)
# and build url of file to replace them with
sink_filename_with_relative_path = TEXT_FOLDER_NAME + "/" + sink_filename
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
if (parts is not None) and (parts[0] in namelist):
attribute_value = sink_filename_with_relative_path
if len(parts) > 1 and parts[1] != "":
attribute_value += "#" + parts[1]
tag["src"] = attribute_value
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=" ")
return newdata
def repairXML(data, mtype="", indent_chars=" "):
newdata = _remove_xml_header(data)
# if well-formed - don't mess with it
if _well_formed(newdata):
return data
newdata = _make_it_sane(newdata)
if not _well_formed(newdata):
newdata = _reformat(newdata)
if mtype == "application/oebps-package+xml":
newdata = newdata.decode('utf-8')
newdata = Opf_Parser(newdata).rebuild_opfxml()
# lxml requires utf-8 on Mac, won't work with unicode
if isinstance(newdata, str):
newdata = newdata.encode('utf-8')
voidtags = get_void_tags(mtype)
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=voidtags)
soup = BeautifulSoup(newdata,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=indent_chars)
return newdata
def anchorNCXUpdates(data, ncx_bookpath, originating_bookpath, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary
id_dict = OrderedDict()
for i in range(0, len(keylist)):
id_dict[ keylist[i] ] = valuelist[i]
startdir = startingDir(ncx_bookpath)
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
apath = urldecodepart(parts[0])
# convert this path to its target bookpath
target_bookpath = buildBookPath(apath, startdir)
if (parts is not None) and (len(parts) > 1) and (target_bookpath == originating_bookpath) and (parts[1] != ""):
fragment_id = urldecodepart(parts[1])
if fragment_id in id_dict:
target_bookpath = id_dict[fragment_id]
attribute_value = urlencodepart(buildRelativePath(ncx_bookpath, target_bookpath))
attribute_value = attribute_value + "#" + urlencodepart(fragment_id)
tag["src"] = attribute_value;
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ")
return newdata
def performPageMapUpdates(data, newbkpath, oldbkpath, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary of xml_updates properly adjusted
updates = OrderedDict()
for i in range(0, len(keylist)):
updates[ keylist[i] ] = valuelist[i]
xml_empty_tags = ["page"]
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=xml_empty_tags)
soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
for tag in soup.find_all(["page"]):
for att in ["href"]:
if att in tag.attrs :
ref = tag[att]
if ref.find(":") == -1 :
parts = ref.split('#')
apath = urldecodepart(parts[0])
fragment = ""
if len(parts) > 1:
fragment = urldecodepart(parts[1])
oldtarget = buildBookPath(apath, startingDir(oldbkpath))
newtarget = updates.get(oldtarget, oldtarget)
attribute_value = urlencodepart(buildRelativePath(newbkpath, newtarget))
if fragment != "":
attribute_value = attribute_value + "#" + urlencodepart(fragment)
tag[att] = attribute_value
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ")
return newdata
def performNCXSourceUpdates(data, currentdir, keylist, valuelist):
# rebuild serialized lookup dictionary
updates = {}
for i in range(0, len(keylist)):
updates[ keylist[i] ] = valuelist[i]
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
url = parts[0]
fragment = ""
if len(parts) > 1:
fragment = parts[1]
bookrelpath = os.path.join(currentdir, unquoteurl(url))
bookrelpath = os.path.normpath(bookrelpath)
bookrelpath = bookrelpath.replace(os.sep, "/")
if bookrelpath in updates:
attribute_value = updates[bookrelpath]
if fragment != "":
attribute_value = attribute_value + "#" + fragment
attribute_value = quoteurl(attribute_value)
tag["src"] = attribute_value
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ")
return newdata
def repairXML(data, self_closing_tags=ebook_xml_empty_tags, indent_chars=" "):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=self_closing_tags)
soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
return newdata
def repairXML(data, mtype="", indent_chars=" "):
data = _remove_xml_header(data)
data = _make_it_sane(data)
voidtags = get_void_tags(mtype)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=voidtags)
soup = BeautifulSoup(data, features=None, from_encoding="utf-8", builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
return newdata
def repairXML(data, self_closing_tags=ebook_xml_empty_tags, indent_chars=" "):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=self_closing_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=indent_chars)
return newdata
def repairXML(data, mtype="", indent_chars=" "):
newdata = _remove_xml_header(data)
# if well-formed - don't mess with it
if _well_formed(newdata):
return data
newdata = _make_it_sane(newdata)
if not _well_formed(newdata):
newdata = _reformat(newdata)
if mtype == "application/oebps-package+xml":
newdata = newdata.decode('utf-8')
newdata = Opf_Parser(newdata).rebuild_opfxml()
# lxml requires utf-8 on Mac, won't work with unicode
if isinstance(newdata, str):
newdata = newdata.encode('utf-8')
voidtags = get_void_tags(mtype)
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=voidtags)
soup = BeautifulSoup(newdata, features=None, from_encoding="utf-8", builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
return newdata
def anchorNCXUpdates(data, originating_filename, keylist, valuelist):
# rebuild serialized lookup dictionary
id_dict = {}
for i in range(0, len(keylist)):
id_dict[ keylist[i] ] = valuelist[i]
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=ebook_xml_empty_tags)
soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
original_filename_with_relative_path = TEXT_FOLDER_NAME + "/" + originating_filename
for tag in soup.find_all("content"):
if "src" in tag.attrs:
src = tag["src"]
if src.find(":") == -1:
parts = src.split('#')
if (parts is not None) and (len(parts) > 1) and (parts[0] == original_filename_with_relative_path) and (parts[1] != ""):
fragment_id = parts[1]
if fragment_id in id_dict:
attribute_value = TEXT_FOLDER_NAME + "/" + quoteurl(id_dict[fragment_id]) + "#" + fragment_id
tag["src"] = attribute_value
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=" ")
return newdata
def performPageMapUpdates(data, currentdir, keylist, valuelist):
data = _remove_xml_header(data)
# lxml on a Mac does not seem to handle full unicode properly, so encode as utf-8
data = data.encode('utf-8')
# rebuild serialized lookup dictionary of xml_updates properly adjusted
updates = {}
for i in range(0, len(keylist)):
updates[keylist[i]] = "../" + valuelist[i]
xml_empty_tags = ["page"]
xmlbuilder = LXMLTreeBuilderForXML(parser=None,
empty_element_tags=xml_empty_tags)
soup = BeautifulSoup(data,
features=None,
from_encoding="utf-8",
builder=xmlbuilder)
for tag in soup.find_all(["page"]):
for att in ["href"]:
if att in tag.attrs:
ref = tag[att]
if ref.find(":") == -1:
parts = ref.split('#')
url = parts[0]
fragment = ""
if len(parts) > 1:
fragment = parts[1]
bookrelpath = os.path.join(currentdir, unquoteurl(url))
bookrelpath = os.path.normpath(bookrelpath)
bookrelpath = bookrelpath.replace(os.sep, "/")
if bookrelpath in updates:
attribute_value = updates[bookrelpath]
if fragment != "":
attribute_value = attribute_value + "#" + fragment
attribute_value = quoteurl(attribute_value)
tag[att] = attribute_value
newdata = soup.decodexml(indent_level=0,
formatter='minimal',
indent_chars=" ")
return newdata
def repairXML(data, self_closing_tags=ebook_xml_empty_tags, indent_chars=" "):
xmlbuilder = LXMLTreeBuilderForXML(parser=None, empty_element_tags=self_closing_tags)
soup = BeautifulSoup(data, features=None, builder=xmlbuilder)
newdata = soup.decodexml(indent_level=0, formatter='minimal', indent_chars=indent_chars)
return newdata
