Added htmlz and pdf formats.

Added HTML reader/writer and PDF reader.

README.rst

@@ -53,6 +53,8 @@ Currently, I've tested following input formats:
 - rtf
 - mobi
 - fb2
+- html
+- pdf
 
 Note, that old Microsoft doc format is not supported, although old documents
 can be fairly easy converted using text processors programs, lik Word or
@@ -68,6 +70,7 @@ Currently, following formats are supported:
 - epub
 - mobi
 - docx
+- htmlz (zipped HTML file with additional assets, like images)
 
 
 Installation

ebook_converter/ebooks/conversion/plugins/pdf_input.py

@@ -41,8 +41,7 @@ class PDFInput(InputFormatPlugin):
         PDFDocument(xml, self.opts, self.log)
         return os.path.join(getcwd(), 'metadata.opf')
 
-    def convert(self, stream, options, file_ext, log,
-                accelerators):
+    def convert(self, stream, options, file_ext, log, accelerators):
         from ebook_converter.ebooks.metadata.opf2 import OPFCreator
         from ebook_converter.ebooks.pdf.pdftohtml import pdftohtml
 

ebook_converter/ebooks/conversion/preprocess.py

@@ -471,7 +471,7 @@ class HTMLPreProcessor(object):
         return re.search('<H2[^><]*id=BookTitle', raw) is not None
 
     def is_pdftohtml(self, src):
-        return '<!-- created by calibre\'s pdftohtml -->' in src[:1000]
+        return '<!-- created by ebook-converter\'s pdftohtml -->' in src[:1000]
 
     def __call__(self, html, remove_special_chars=None,
             get_preprocess_html=False):
@@ -627,7 +627,7 @@ class HTMLPreProcessor(object):
             html = preprocessor(html)
 
         if is_pdftohtml:
-            html = html.replace('<!-- created by calibre\'s pdftohtml -->', '')
+            html = html.replace('<!-- created by ebook-converter\'s pdftohtml -->', '')
 
         if getattr(self.extra_opts, 'smarten_punctuation', False):
             html = smarten_punctuation(html, self.log)

ebook_converter/ebooks/conversion/utils.py

@@ -43,7 +43,7 @@ class HeuristicProcessor(object):
         self.common_in_text_beginnings = '[\\w\'\"“‘‛]'
 
     def is_pdftohtml(self, src):
-        return '<!-- created by calibre\'s pdftohtml -->' in src[:1000]
+        return '<!-- created by ebook-converter\'s pdftohtml -->' in src[:1000]
 
     def is_abbyy(self, src):
         return '<meta name="generator" content="ABBYY FineReader' in src[:1000]

ebook_converter/ebooks/docx/writer/styles.py

@@ -15,7 +15,7 @@ from ebook_converter.ebooks import parse_css_length
 from ebook_converter.ebooks.docx.writer.utils import convert_color, int_or_zero
 from ebook_converter.utils.localization import lang_as_iso639_1
 from ebook_converter.polyglot.builtins import iteritems, filter, unicode_type
-from tinycss.css21 import CSS21Parser
+from ebook_converter.tinycss.css21 import CSS21Parser
 
 css_parser = CSS21Parser()
 

ebook_converter/ebooks/docx/writer/utils.py

@@ -5,7 +5,7 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 __license__ = 'GPL v3'
 __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 
-from tinycss.color3 import parse_color_string
+from ebook_converter.tinycss.color3 import parse_color_string
 
 
 def int_or_zero(raw):

ebook_converter/ebooks/metadata/pdf.py

@@ -7,17 +7,17 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 import os, subprocess, shutil, re
 from functools import partial
 
-from calibre import prints
-from calibre.constants import iswindows, ispy3
-from calibre.ptempfile import TemporaryDirectory
-from calibre.ebooks.metadata import (
+from ebook_converter import prints
+from ebook_converter.constants import iswindows, ispy3
+from ebook_converter.ptempfile import TemporaryDirectory
+from ebook_converter.ebooks.metadata import (
     MetaInformation, string_to_authors, check_isbn, check_doi)
-from calibre.utils.ipc.simple_worker import fork_job, WorkerError
-from polyglot.builtins import iteritems, unicode_type
+from ebook_converter.utils.ipc.simple_worker import fork_job, WorkerError
+from ebook_converter.polyglot.builtins import iteritems, unicode_type
 
 
 def get_tools():
-    from calibre.ebooks.pdf.pdftohtml import PDFTOHTML
+    from ebook_converter.ebooks.pdf.pdftohtml import PDFTOHTML
     base = os.path.dirname(PDFTOHTML)
     suffix = '.exe' if iswindows else ''
     pdfinfo = os.path.join(base, 'pdfinfo') + suffix
@@ -32,12 +32,14 @@ def read_info(outputdir, get_cover):
     way to pass unicode paths via command line arguments. This also ensures
     that if poppler crashes, no stale file handles are left for the original
     file, only for src.pdf.'''
-    os.chdir(outputdir)
     pdfinfo, pdftoppm = get_tools()
+    source_file = os.path.join(outputdir, 'src.pdf')
+    cover_file = os.path.join(outputdir, 'cover')
     ans = {}
 
     try:
-        raw = subprocess.check_output([pdfinfo, '-enc', 'UTF-8', '-isodates', 'src.pdf'])
+        raw = subprocess.check_output([pdfinfo, '-enc', 'UTF-8', '-isodates',
+                                       source_file])
     except subprocess.CalledProcessError as e:
         prints('pdfinfo errored out with return code: %d'%e.returncode)
         return None
@@ -61,7 +63,7 @@ def read_info(outputdir, get_cover):
     # https://cgit.freedesktop.org/poppler/poppler/commit/?id=c91483aceb1b640771f572cb3df9ad707e5cad0d
     # we can no longer rely on it.
     try:
-        raw = subprocess.check_output([pdfinfo, '-meta', 'src.pdf']).strip()
+        raw = subprocess.check_output([pdfinfo, '-meta', source_file]).strip()
     except subprocess.CalledProcessError as e:
         prints('pdfinfo failed to read XML metadata with return code: %d'%e.returncode)
     else:
@@ -74,8 +76,8 @@ def read_info(outputdir, get_cover):
 
     if get_cover:
         try:
-            subprocess.check_call([pdftoppm, '-singlefile', '-jpeg', '-cropbox',
-                'src.pdf', 'cover'])
+            subprocess.check_call([pdftoppm, '-singlefile', '-jpeg',
+                                   '-cropbox', source_file, cover_file])
         except subprocess.CalledProcessError as e:
             prints('pdftoppm errored out with return code: %d'%e.returncode)
 
@@ -114,17 +116,7 @@ def get_metadata(stream, cover=True):
         stream.seek(0)
         with open(os.path.join(pdfpath, 'src.pdf'), 'wb') as f:
             shutil.copyfileobj(stream, f)
-        try:
-            res = fork_job('calibre.ebooks.metadata.pdf', 'read_info',
-                    (pdfpath, bool(cover)))
-        except WorkerError as e:
-            prints(e.orig_tb)
-            raise RuntimeError('Failed to run pdfinfo')
-        info = res['result']
-        with open(res['stdout_stderr'], 'rb') as f:
-            raw = f.read().strip()
-            if raw:
-                prints(raw)
+        info = read_info(pdfpath, bool(cover))
         if info is None:
             raise ValueError('Could not read info dict from PDF')
         covpath = os.path.join(pdfpath, 'cover.jpg')
@@ -140,8 +132,6 @@ def get_metadata(stream, cover=True):
     else:
         au = string_to_authors(au)
     mi = MetaInformation(title, au)
-    # if isbn is not None:
-    #    mi.isbn = isbn
 
     creator = info.get('Creator', None)
     if creator:
@@ -161,7 +151,7 @@ def get_metadata(stream, cover=True):
         mi.tags.insert(0, subject)
 
     if 'xmp_metadata' in info:
-        from calibre.ebooks.metadata.xmp import consolidate_metadata
+        from ebook_converter.ebooks.metadata.xmp import consolidate_metadata
         mi = consolidate_metadata(mi, info)
 
     # Look for recognizable identifiers in the info dict, if they were not
@@ -182,9 +172,8 @@ def get_metadata(stream, cover=True):
 
 get_quick_metadata = partial(get_metadata, cover=False)
 
-from calibre.utils.podofo import set_metadata as podofo_set_metadata
+#from ebook_converter.utils.podofo import set_metadata as podofo_set_metadata
 
 
 def set_metadata(stream, mi):
-    stream.seek(0)
-    return podofo_set_metadata(stream, mi)
+    return None

ebook_converter/ebooks/metadata/xmp.py

@@ -12,14 +12,14 @@ from collections import defaultdict
 from lxml import etree
 from lxml.builder import ElementMaker
 
-from calibre import prints
-from calibre.ebooks.metadata import check_isbn, check_doi
-from calibre.utils.xml_parse import safe_xml_fromstring
-from calibre.ebooks.metadata.book.base import Metadata
-from calibre.ebooks.metadata.opf2 import dump_dict
-from calibre.utils.date import parse_date, isoformat, now
-from calibre.utils.localization import canonicalize_lang, lang_as_iso639_1
-from polyglot.builtins import iteritems, string_or_bytes, filter
+from ebook_converter import prints
+from ebook_converter.ebooks.metadata import check_isbn, check_doi
+from ebook_converter.utils.xml_parse import safe_xml_fromstring
+from ebook_converter.ebooks.metadata.book.base import Metadata
+from ebook_converter.ebooks.metadata.opf2 import dump_dict
+from ebook_converter.utils.date import parse_date, isoformat, now
+from ebook_converter.utils.localization import canonicalize_lang, lang_as_iso639_1
+from ebook_converter.polyglot.builtins import iteritems, string_or_bytes, filter
 
 _xml_declaration = re.compile(r'<\?xml[^<>]+encoding\s*=\s*[\'"](.*?)[\'"][^<>]*>', re.IGNORECASE)
 
@@ -173,8 +173,8 @@ def read_series(root):
 
 
 def read_user_metadata(mi, root):
-    from calibre.utils.config import from_json
-    from calibre.ebooks.metadata.book.json_codec import decode_is_multiple
+    from ebook_converter.utils.config import from_json
+    from ebook_converter.ebooks.metadata.book.json_codec import decode_is_multiple
     fields = set()
     for item in XPath('//calibre:custom_metadata')(root):
         for li in XPath('./rdf:Bag/rdf:li')(item):
@@ -436,8 +436,8 @@ def create_series(calibre, series, series_index):
 
 
 def create_user_metadata(calibre, all_user_metadata):
-    from calibre.utils.config import to_json
-    from calibre.ebooks.metadata.book.json_codec import object_to_unicode, encode_is_multiple
+    from ebook_converter.utils.config import to_json
+    from ebook_converter.ebooks.metadata.book.json_codec import object_to_unicode, encode_is_multiple
 
     s = calibre.makeelement(expand('calibre:custom_metadata'))
     calibre.append(s)
@@ -640,7 +640,7 @@ def merge_xmp_packet(old, new):
 
 
 if __name__ == '__main__':
-    from calibre.utils.podofo import get_xmp_metadata
+    from ebook_converter.utils.podofo import get_xmp_metadata
     xmp_packet = get_xmp_metadata(sys.argv[-1])
     mi = metadata_from_xmp_packet(xmp_packet)
     np = metadata_to_xmp_packet(mi)

ebook_converter/ebooks/mobi/utils.py

@@ -14,7 +14,7 @@ from ebook_converter.utils.img import save_cover_data_to, scale_image, image_to_
 from ebook_converter.utils.imghdr import what
 from ebook_converter.ebooks import normalize
 from ebook_converter.polyglot.builtins import unicode_type, range, as_bytes, map
-from tinycss.color3 import parse_color_string
+from ebook_converter.tinycss.color3 import parse_color_string
 
 IMAGE_MAX_SIZE = 10 * 1024 * 1024
 RECORD_SIZE = 0x1000  # 4096 (Text record size (uncompressed))

ebook_converter/ebooks/oeb/stylizer.py

@@ -500,7 +500,6 @@ class Style(object):
         background shortcut properties. Note that inheritance/default values
         are not used. None is returned if no background color is set.
         '''
-
         def validate_color(col):
             return cssprofiles.validateWithProfile('color',
                         col,

ebook_converter/ebooks/pdf/pdftohtml.py

@@ -98,7 +98,7 @@ def pdftohtml(output_dir, pdf_path, no_images, as_xml=False):
             with lopen(index, 'r+b') as i:
                 raw = i.read().decode('utf-8', 'replace')
                 raw = flip_images(raw)
-                raw = raw.replace('<head', '<!-- created by ebook_converter\'s pdftohtml -->\n  <head', 1)
+                raw = raw.replace('<head', '<!-- created by ebook-converter\'s pdftohtml -->\n  <head', 1)
                 i.seek(0)
                 i.truncate()
                 # versions of pdftohtml >= 0.20 output self closing <br> tags, this

ebook_converter/tinycss/__init__.py

@@ -12,9 +12,9 @@
 from .version import VERSION
 __version__ = VERSION
 
-from tinycss.css21 import CSS21Parser
-from tinycss.page3 import CSSPage3Parser
-from tinycss.fonts3 import CSSFonts3Parser
+from ebook_converter.tinycss.css21 import CSS21Parser
+from ebook_converter.tinycss.page3 import CSSPage3Parser
+from ebook_converter.tinycss.fonts3 import CSSFonts3Parser
 from ebook_converter.tinycss.media3 import CSSMedia3Parser
 
 

ebook_converter/tinycss/css21.py

@@ -13,10 +13,10 @@
 from __future__ import unicode_literals
 from itertools import chain, islice
 
-from tinycss.decoding import decode
-from tinycss.token_data import TokenList
-from tinycss.tokenizer import tokenize_grouped
-from tinycss.parsing import (
+from ebook_converter.tinycss.decoding import decode
+from ebook_converter.tinycss.token_data import TokenList
+from ebook_converter.tinycss.tokenizer import tokenize_grouped
+from ebook_converter.tinycss.parsing import (
     strip_whitespace, remove_whitespace, split_on_comma, validate_value,
     validate_any, ParseError)
 

ebook_converter/tinycss/decoding.py

@@ -15,7 +15,7 @@ from __future__ import unicode_literals
 import operator
 import re
 
-from polyglot.binary import from_hex_bytes
+from ebook_converter.polyglot.binary import from_hex_bytes
 
 
 __all__ = ['decode']  # Everything else is implementation detail

ebook_converter/tinycss/fonts3.py

@@ -8,8 +8,8 @@ __copyright__ = '2014, Kovid Goyal <kovid at kovidgoyal.net>'
 
 import re
 from ebook_converter.polyglot.builtins import map
-from tinycss.css21 import CSS21Parser, ParseError
-from tinycss.tokenizer import tokenize_grouped
+from ebook_converter.tinycss.css21 import CSS21Parser, ParseError
+from ebook_converter.tinycss.tokenizer import tokenize_grouped
 
 
 def parse_font_family_tokens(tokens):

ebook_converter/tinycss/media3.py

@@ -5,8 +5,8 @@ from __future__ import absolute_import, division, print_function, unicode_litera
 __license__ = 'GPL v3'
 __copyright__ = '2014, Kovid Goyal <kovid at kovidgoyal.net>'
 
-from tinycss.css21 import CSS21Parser
-from tinycss.parsing import remove_whitespace, split_on_comma, ParseError
+from ebook_converter.tinycss.css21 import CSS21Parser
+from ebook_converter.tinycss.parsing import remove_whitespace, split_on_comma, ParseError
 from ebook_converter.polyglot.builtins import error_message
 
 

ebook_converter/tinycss/tokenizer.py

@@ -14,7 +14,7 @@
 
 from __future__ import unicode_literals
 
-from tinycss import token_data
+from ebook_converter.tinycss import token_data
 
 
 def tokenize_flat(css_source, ignore_comments=True,