gryf/ebook-converter
1
0
Fork 0
mirror of https://github.com/gryf/ebook-converter.git synced 2026-04-24 07:01:30 +02:00
Code Issues Projects Releases Wiki Activity

Use the real constants module.

Browse Source 
This is progressing refactor of the calibre code to make it more
readable, and transform it to something more coherent.

In this patch, there are changes regarding imports for some modules,
instead of polluting namespace of each module with some other modules
symbols, which often were imported from other modules. Yuck.
This commit is contained in:
gryf
2020-05-29 17:04:53 +02:00
parent ee4801228f
commit ce89f5c9d1
54 changed files with 2383 additions and 2081 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ebook_converter/ebooks/docx/container.py
+54 -41
View File
@@ -1,9 +1,12 @@
import os, sys, shutil
import os
import shutil
import sys
from lxml import etree
from ebook_converter import walk, guess_type
from ebook_converter.ebooks.metadata import string_to_authors, authors_to_sort_string
from ebook_converter.ebooks.metadata import authors_to_sort_string
from ebook_converter.ebooks.metadata import string_to_authors
from ebook_converter.ebooks.metadata.book.base import Metadata
from ebook_converter.ebooks.docx import InvalidDOCX
from ebook_converter.ebooks.docx.names import DOCXNamespace
@@ -11,21 +14,11 @@ from ebook_converter.ptempfile import PersistentTemporaryDirectory
from ebook_converter.utils.localization import canonicalize_lang
from ebook_converter.utils.logging import default_log
from ebook_converter.utils.zipfile import ZipFile
from ebook_converter.utils.xml_parse import safe_xml_fromstring
__license__ = 'GPL v3'
__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
def fromstring(raw, parser=None):
return safe_xml_fromstring(raw)
# Read metadata {{{
def read_doc_props(raw, mi, XPath):
root = fromstring(raw)
root = etree.fromstring(raw)
titles = XPath('//dc:title')(root)
if titles:
title = titles[0].text
@@ -53,29 +46,31 @@ def read_doc_props(raw, mi, XPath):
desc = XPath('//dc:description')(root)
if desc:
raw = etree.tostring(desc[0], method='text', encoding='unicode')
raw = raw.replace('_x000d_', '') # Word 2007 mangles newlines in the summary
# Word 2007 mangles newlines in the summary
raw = raw.replace('_x000d_', '')
mi.comments = raw.strip()
langs = []
for lang in XPath('//dc:language')(root):
if lang.text and lang.text.strip():
l = canonicalize_lang(lang.text)
if l:
langs.append(l)
canonic_lang = canonicalize_lang(lang.text)
if canonic_lang:
langs.append(canonic_lang)
if langs:
mi.languages = langs
def read_app_props(raw, mi):
root = fromstring(raw)
root = etree.fromstring(raw)
company = root.xpath('//*[local-name()="Company"]')
if company and company[0].text and company[0].text.strip():
mi.publisher = company[0].text.strip()
def read_default_style_language(raw, mi, XPath):
root = fromstring(raw)
for lang in XPath('/w:styles/w:docDefaults/w:rPrDefault/w:rPr/w:lang/@w:val')(root):
root = etree.fromstring(raw)
for lang in XPath('/w:styles/w:docDefaults/w:rPrDefault/w:rPr/w:lang/'
'@w:val')(root):
lang = canonicalize_lang(lang)
if lang:
mi.languages = [lang]
@@ -87,7 +82,9 @@ class DOCX(object):
def __init__(self, path_or_stream, log=None, extract=True):
self.docx_is_transitional = True
stream = path_or_stream if hasattr(path_or_stream, 'read') else open(path_or_stream, 'rb')
stream = path_or_stream
if not hasattr(path_or_stream, 'read'):
stream = open(path_or_stream, 'rb')
self.name = getattr(stream, 'name', None) or '<stream>'
self.log = log or default_log
if extract:
@@ -107,9 +104,9 @@ class DOCX(object):
try:
zf = ZipFile(stream)
zf.extractall(self.tdir)
except:
except Exception:
self.log.exception('DOCX appears to be invalid ZIP file, trying a'
' more forgiving ZIP parser')
' more forgiving ZIP parser')
from ebook_converter.utils.localunzip import extractall
stream.seek(0)
extractall(stream, self.tdir)
@@ -133,13 +130,17 @@ class DOCX(object):
try:
raw = self.read('[Content_Types].xml')
except KeyError:
raise InvalidDOCX('The file %s docx file has no [Content_Types].xml' % self.name)
root = fromstring(raw)
raise InvalidDOCX('The file %s docx file has no '
'[Content_Types].xml' % self.name)
root = etree.fromstring(raw)
self.content_types = {}
self.default_content_types = {}
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Default" and @Extension and @ContentType]'):
self.default_content_types[item.get('Extension').lower()] = item.get('ContentType')
for item in root.xpath('//*[local-name()="Types"]/*[local-name()="Override" and @PartName and @ContentType]'):
for item in root.xpath('//*[local-name()="Types"]/*[local-name()='
'"Default" and @Extension and @ContentType]'):
self.default_content_types[item.get('Extension').lower()] = \
item.get('ContentType')
for item in root.xpath('//*[local-name()="Types"]/*[local-name()='
'"Override" and @PartName and @ContentType]'):
name = item.get('PartName').lstrip('/')
self.content_types[name] = item.get('ContentType')
@@ -155,15 +156,19 @@ class DOCX(object):
try:
raw = self.read('_rels/.rels')
except KeyError:
raise InvalidDOCX('The file %s docx file has no _rels/.rels' % self.name)
root = fromstring(raw)
raise InvalidDOCX('The file %s docx file has no _rels/.rels' %
self.name)
root = etree.fromstring(raw)
self.relationships = {}
self.relationships_rmap = {}
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name'
'()="Relationship" and @Type and @Target]'):
target = item.get('Target').lstrip('/')
typ = item.get('Type')
if target == 'word/document.xml':
self.docx_is_transitional = typ != 'http://purl.oclc.org/ooxml/officeDocument/relationships/officeDocument'
self.docx_is_transitional = (typ != 'http://purl.oclc.org/'
'ooxml/officeDocument/'
'relationships/officeDocument')
self.relationships[typ] = target
self.relationships_rmap[target] = typ
@@ -171,15 +176,17 @@ class DOCX(object):
def document_name(self):
name = self.relationships.get(self.namespace.names['DOCUMENT'], None)
if name is None:
names = tuple(n for n in self.names if n == 'document.xml' or n.endswith('/document.xml'))
names = tuple(n for n in self.names if n == 'document.xml' or
n.endswith('/document.xml'))
if not names:
raise InvalidDOCX('The file %s docx file has no main document' % self.name)
raise InvalidDOCX('The file %s docx file has no main '
'document' % self.name)
name = names[0]
return name
@property
def document(self):
return fromstring(self.read(self.document_name))
return etree.fromstring(self.read(self.document_name))
@property
def document_relationships(self):
@@ -195,10 +202,13 @@ class DOCX(object):
except KeyError:
pass
else:
root = fromstring(raw)
for item in root.xpath('//*[local-name()="Relationships"]/*[local-name()="Relationship" and @Type and @Target]'):
root = etree.fromstring(raw)
for item in root.xpath('//*[local-name()="Relationships"]/*'
'[local-name()="Relationship" and @Type '
'and @Target]'):
target = item.get('Target')
if item.get('TargetMode', None) != 'External' and not target.startswith('#'):
if (item.get('TargetMode', None) != 'External' and not
target.startswith('#')):
target = '/'.join((base, target.lstrip('/')))
typ = item.get('Type')
Id = item.get('Id')
@@ -209,13 +219,15 @@ class DOCX(object):
def get_document_properties_names(self):
name = self.relationships.get(self.namespace.names['DOCPROPS'], None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/core.xml')
names = tuple(n for n in self.names
if n.lower() == 'docprops/core.xml')
if names:
name = names[0]
yield name
name = self.relationships.get(self.namespace.names['APPPROPS'], None)
if name is None:
names = tuple(n for n in self.names if n.lower() == 'docprops/app.xml')
names = tuple(n for n in self.names
if n.lower() == 'docprops/app.xml')
if names:
name = names[0]
yield name
@@ -239,7 +251,8 @@ class DOCX(object):
else:
read_default_style_language(raw, mi, self.namespace.XPath)
ap_name = self.relationships.get(self.namespace.names['APPPROPS'], None)
ap_name = self.relationships.get(self.namespace.names['APPPROPS'],
None)
if ap_name:
try:
raw = self.read(ap_name)
ebook_converter/ebooks/docx/to_html.py
+8 -7
View File
@@ -1,12 +1,13 @@
import sys, os, re, math, errno, uuid, numbers
from collections import OrderedDict, defaultdict
from lxml import etree
from lxml import html
from lxml.html.builder import (
HTML, HEAD, TITLE, BODY, LINK, META, P, SPAN, BR, DIV, A, DT, DL, DD, H1)
from ebook_converter import guess_type
from ebook_converter.ebooks.docx.container import DOCX, fromstring
from ebook_converter.ebooks.docx.container import DOCX
from ebook_converter.ebooks.docx.names import XML, generate_anchor
from ebook_converter.ebooks.docx.styles import Styles, inherit, PageProperties
from ebook_converter.ebooks.docx.numbering import Numbering
@@ -311,7 +312,7 @@ class Convert(object):
raise
self.log.warn('Settings %s file missing' % sename)
else:
self.settings(fromstring(seraw))
self.settings(etree.fromstring(seraw))
if foname is not None:
try:
@@ -327,7 +328,7 @@ class Convert(object):
self.log.warn('Endnotes %s do not exist' % enname)
else:
enrel = self.docx.get_relationships(enname)
footnotes(fromstring(foraw) if foraw else None, forel, fromstring(enraw) if enraw else None, enrel)
footnotes(etree.fromstring(foraw) if foraw else None, forel, etree.fromstring(enraw) if enraw else None, enrel)
if fname is not None:
embed_relationships = self.docx.get_relationships(fname)[0]
@@ -336,7 +337,7 @@ class Convert(object):
except KeyError:
self.log.warn('Fonts table %s does not exist' % fname)
else:
fonts(fromstring(raw), embed_relationships, self.docx, self.dest_dir)
fonts(etree.fromstring(raw), embed_relationships, self.docx, self.dest_dir)
if tname is not None:
try:
@@ -344,7 +345,7 @@ class Convert(object):
except KeyError:
self.log.warn('Styles %s do not exist' % sname)
else:
self.theme(fromstring(raw))
self.theme(etree.fromstring(raw))
styles_loaded = False
if sname is not None:
@@ -353,7 +354,7 @@ class Convert(object):
except KeyError:
self.log.warn('Styles %s do not exist' % sname)
else:
self.styles(fromstring(raw), fonts, self.theme)
self.styles(etree.fromstring(raw), fonts, self.theme)
styles_loaded = True
if not styles_loaded:
self.styles(None, fonts, self.theme)
@@ -364,7 +365,7 @@ class Convert(object):
except KeyError:
self.log.warn('Numbering styles %s do not exist' % nname)
else:
numbering(fromstring(raw), self.styles, self.docx.get_relationships(nname)[0])
numbering(etree.fromstring(raw), self.styles, self.docx.get_relationships(nname)[0])
self.styles.resolve_numbering(numbering)
ebook_converter/ebooks/docx/writer/from_html.py
+12 -15
View File
@@ -1,22 +1,19 @@
import collections
import re
from collections import Counter
from ebook_converter.ebooks.docx.writer.container import create_skeleton, page_size, page_effective_area
from ebook_converter.ebooks.docx.writer.styles import StylesManager, FloatSpec
from ebook_converter.ebooks.docx.writer.links import LinksManager
from ebook_converter.ebooks.docx.writer.images import ImagesManager
from ebook_converter.ebooks.docx.writer.fonts import FontsManager
from ebook_converter.ebooks.docx.writer.tables import Table
from ebook_converter.ebooks.docx.writer.images import ImagesManager
from ebook_converter.ebooks.docx.writer.links import LinksManager
from ebook_converter.ebooks.docx.writer.lists import ListsManager
from ebook_converter.ebooks.docx.writer.styles import StylesManager, FloatSpec
from ebook_converter.ebooks.docx.writer.tables import Table
from ebook_converter.ebooks.oeb import base
from ebook_converter.ebooks.oeb import parse_utils
from ebook_converter.ebooks.oeb.stylizer import Stylizer as Sz, Style as St
from ebook_converter.ebooks.oeb.base import XPath, barename
from ebook_converter.utils.localization import lang_as_iso639_1
__license__ = 'GPL v3'
__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
def lang_for_tag(tag):
for attr in ('lang', '{http://www.w3.org/XML/1998/namespace}lang'):
val = lang_as_iso639_1(tag.get(attr))
@@ -140,7 +137,7 @@ class Block(object):
self.numbering_id = None
self.parent_items = None
self.html_block = html_block
self.html_tag = barename(html_block.tag)
self.html_tag = parse_utils.barename(html_block.tag)
self.float_spec = float_spec
if float_spec is not None:
float_spec.blocks.append(self)
@@ -387,7 +384,7 @@ class Blocks(object):
def resolve_language(self):
default_lang = self.styles_manager.document_lang
for block in self.all_blocks:
count = Counter()
count = collections.Counter()
for run in block.runs:
count[run.lang] += 1
if count:
@@ -473,13 +470,13 @@ class Convert(object):
self.abshref = self.images_manager.abshref = item.abshref
self.current_lang = lang_for_tag(item.data) or self.styles_manager.document_lang
for i, body in enumerate(XPath('//h:body')(item.data)):
for i, body in enumerate(base.XPath('//h:body')(item.data)):
with self.blocks:
self.blocks.top_bookmark = self.links_manager.bookmark_for_anchor(self.links_manager.top_anchor, self.current_item, body)
self.process_tag(body, stylizer, is_first_tag=i == 0)
def process_tag(self, html_tag, stylizer, is_first_tag=False, float_spec=None):
tagname = barename(html_tag.tag)
tagname = parse_utils.barename(html_tag.tag)
tag_style = stylizer.style(html_tag)
ignore_tag_contents = tagname in {'script', 'style', 'title', 'meta'} or tag_style.is_hidden
display = tag_style._get('display')
@@ -573,7 +570,7 @@ class Convert(object):
text = html_tag.text
if text:
block.add_text(text, tag_style, ignore_leading_whitespace=True, is_parent_style=True, link=self.current_link, lang=self.current_lang)
elif tagname == 'li' and len(html_tag) and barename(html_tag[0].tag) in ('ul', 'ol') and len(html_tag[0]):
elif tagname == 'li' and len(html_tag) and parse_utils.barename(html_tag[0].tag) in ('ul', 'ol') and len(html_tag[0]):
block.force_not_empty = True
def add_inline_tag(self, tagname, html_tag, tag_style, stylizer):