Get rid of basestring

src/calibre/customize/zipplugin.py

@@ -15,6 +15,7 @@ from functools import partial
 from calibre import as_unicode
 from calibre.customize import (Plugin, numeric_version, platform,
         InvalidPlugin, PluginNotFound)
+from polyglot.builtins import string_or_bytes
 
 # PEP 302 based plugin loading mechanism, works around the bug in zipimport in
 # python 2.x that prevents importing from zip files in locations whose paths
@@ -34,7 +35,7 @@ def get_resources(zfp, name_or_list_of_names):
                 be just the bytes of the resource or None if it wasn't found.
     '''
     names = name_or_list_of_names
-    if isinstance(names, basestring):
+    if isinstance(names, string_or_bytes):
         names = [names]
     ans = {}
     with zipfile.ZipFile(zfp) as zf:
@@ -65,11 +66,11 @@ def get_icons(zfp, name_or_list_of_names):
     from PyQt5.Qt import QIcon, QPixmap
     names = name_or_list_of_names
     ans = get_resources(zfp, names)
-    if isinstance(names, basestring):
+    if isinstance(names, string_or_bytes):
         names = [names]
     if ans is None:
         ans = {}
-    if isinstance(ans, basestring):
+    if isinstance(ans, string_or_bytes):
         ans = dict([(names[0], ans)])
 
     ians = {}

src/calibre/db/backend.py

@@ -12,7 +12,7 @@ import os, shutil, uuid, json, glob, time, hashlib, errno, sys
 from functools import partial
 
 import apsw
-from polyglot.builtins import unicode_type, reraise
+from polyglot.builtins import unicode_type, reraise, string_or_bytes
 
 from calibre import isbytestring, force_unicode, prints, as_unicode
 from calibre.constants import (iswindows, filesystem_encoding,
@@ -1348,7 +1348,7 @@ class DB(object):
     def copy_cover_to(self, path, dest, windows_atomic_move=None, use_hardlink=False, report_file_size=None):
         path = os.path.abspath(os.path.join(self.library_path, path, 'cover.jpg'))
         if windows_atomic_move is not None:
-            if not isinstance(dest, basestring):
+            if not isinstance(dest, string_or_bytes):
                 raise Exception("Error, you must pass the dest as a path when"
                         " using windows_atomic_move")
             if os.access(path, os.R_OK) and dest and not samefile(dest, path):
@@ -1438,7 +1438,7 @@ class DB(object):
         if path is None:
             return False
         if windows_atomic_move is not None:
-            if not isinstance(dest, basestring):
+            if not isinstance(dest, string_or_bytes):
                 raise Exception("Error, you must pass the dest as a path when"
                         " using windows_atomic_move")
             if dest:

src/calibre/db/cache.py

@@ -11,7 +11,7 @@ import os, traceback, random, shutil, operator
 from io import BytesIO
 from collections import defaultdict, Set, MutableSet
 from functools import wraps, partial
-from polyglot.builtins import unicode_type, zip
+from polyglot.builtins import unicode_type, zip, string_or_bytes
 from time import time
 
 from calibre import isbytestring, as_unicode
@@ -1076,7 +1076,7 @@ class Cache(object):
             bimap, simap = {}, {}
             sfield = self.fields[name + '_index']
             for k, v in book_id_to_val_map.iteritems():
-                if isinstance(v, basestring):
+                if isinstance(v, string_or_bytes):
                     v, sid = get_series_values(v)
                 else:
                     v = sid = None
@@ -1298,7 +1298,7 @@ class Cache(object):
         # force_changes has no effect on cover manipulation
         try:
             cdata = mi.cover_data[1]
-            if cdata is None and isinstance(mi.cover, basestring) and mi.cover and os.access(mi.cover, os.R_OK):
+            if cdata is None and isinstance(mi.cover, string_or_bytes) and mi.cover and os.access(mi.cover, os.R_OK):
                 with lopen(mi.cover, 'rb') as f:
                     cdata = f.read() or None
             if cdata is not None:

src/calibre/db/search.py

@@ -19,7 +19,7 @@ from calibre.utils.date import parse_date, UNDEFINED_DATE, now, dt_as_local
 from calibre.utils.icu import primary_contains, sort_key
 from calibre.utils.localization import lang_map, canonicalize_lang
 from calibre.utils.search_query_parser import SearchQueryParser, ParseException
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 CONTAINS_MATCH = 0
 EQUALS_MATCH   = 1
@@ -706,7 +706,7 @@ class Parser(SearchQueryParser):  # {{{
             if location in text_fields:
                 for val, book_ids in self.field_iter(location, current_candidates):
                     if val is not None:
-                        if isinstance(val, basestring):
+                        if isinstance(val, string_or_bytes):
                             val = (val,)
                         if _match(q, val, matchkind, use_primary_find_in_search=upf, case_sensitive=case_sensitive):
                             matches |= book_ids

src/calibre/db/utils.py

@@ -9,7 +9,7 @@ __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 import os, errno, cPickle, sys, re
 from locale import localeconv
 from collections import OrderedDict, namedtuple
-from polyglot.builtins import map, unicode_type
+from polyglot.builtins import map, unicode_type, string_or_bytes
 from threading import Lock
 
 from calibre import as_unicode, prints
@@ -45,7 +45,7 @@ def fuzzy_title_patterns():
     if _fuzzy_title_patterns is None:
         from calibre.ebooks.metadata import get_title_sort_pat
         _fuzzy_title_patterns = tuple((re.compile(pat, re.IGNORECASE) if
-            isinstance(pat, basestring) else pat, repl) for pat, repl in
+            isinstance(pat, string_or_bytes) else pat, repl) for pat, repl in
                 [
                     (r'[\[\](){}<>\'";,:#]', ''),
                     (get_title_sort_pat(), ''),

src/calibre/devices/android/driver.py

@@ -10,6 +10,7 @@ import cStringIO
 
 from calibre import fsync
 from calibre.devices.usbms.driver import USBMS
+from polyglot.builtins import string_or_bytes
 
 HTC_BCDS = [0x100, 0x0222, 0x0224, 0x0226, 0x227, 0x228, 0x229, 0x0231, 0x9999]
 
@@ -286,7 +287,7 @@ class ANDROID(USBMS):
             opts = [self.EBOOK_DIR_MAIN, '']
 
         def strtolist(x):
-            if isinstance(x, basestring):
+            if isinstance(x, string_or_bytes):
                 x = [y.strip() for y in x.split(',')]
             return x or []
 

src/calibre/devices/kobo/books.py

@@ -14,7 +14,7 @@ from calibre.devices.usbms.books import CollectionsBookList
 from calibre.utils.config_base import prefs
 from calibre.devices.usbms.driver import debug_print
 from calibre.ebooks.metadata import author_to_author_sort
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Book(Book_):
@@ -297,7 +297,7 @@ class KTCollectionsBookList(CollectionsBookList):
                 return 1
             if y is None:
                 return -1
-            if isinstance(x, basestring) and isinstance(y, basestring):
+            if isinstance(x, string_or_bytes) and isinstance(y, string_or_bytes):
                 x, y = sort_key(force_unicode(x)), sort_key(force_unicode(y))
             c = cmp(x, y)
             if c != 0:

src/calibre/devices/kobo/driver.py

@@ -32,7 +32,7 @@ from calibre import prints, fsync
 from calibre.ptempfile import PersistentTemporaryFile
 from calibre.constants import DEBUG
 from calibre.utils.config_base import prefs
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 EPUB_EXT  = '.epub'
 KEPUB_EXT = '.kepub'
@@ -3543,14 +3543,14 @@ class KOBOTOUCH(KOBO):
             # a string, so looking for that.
             start_subclass_extra_options = OPT_MODIFY_CSS
             debugging_title = ''
-            if isinstance(settings.extra_customization[OPT_MODIFY_CSS], basestring):
+            if isinstance(settings.extra_customization[OPT_MODIFY_CSS], string_or_bytes):
                 debug_print("KoboTouch::migrate_old_settings - Don't have update_series option")
                 settings.update_series = config.get_option('update_series').default
                 settings.modify_css = config.get_option('modify_css').default
                 settings.support_newer_firmware = settings.extra_customization[OPT_UPDATE_SERIES_DETAILS]
                 debugging_title = settings.extra_customization[OPT_MODIFY_CSS]
                 start_subclass_extra_options = OPT_MODIFY_CSS + 1
-            elif isinstance(settings.extra_customization[OPT_SUPPORT_NEWER_FIRMWARE], basestring):
+            elif isinstance(settings.extra_customization[OPT_SUPPORT_NEWER_FIRMWARE], string_or_bytes):
                 debug_print("KoboTouch::migrate_old_settings - Don't have modify_css option")
                 settings.update_series = settings.extra_customization[OPT_UPDATE_SERIES_DETAILS]
                 settings.modify_css = config.get_option('modify_css').default
@@ -3565,7 +3565,7 @@ class KOBOTOUCH(KOBO):
                 debugging_title = settings.extra_customization[OPT_DEBUGGING_TITLE]
                 start_subclass_extra_options = OPT_DEBUGGING_TITLE + 1
 
-            settings.debugging_title = debugging_title if isinstance(debugging_title, basestring) else ''
+            settings.debugging_title = debugging_title if isinstance(debugging_title, string_or_bytes) else ''
             settings.update_device_metadata = settings.update_series
             settings.extra_customization = settings.extra_customization[start_subclass_extra_options:]
 

src/calibre/devices/usbms/books.py

@@ -14,6 +14,7 @@ from calibre.constants import preferred_encoding
 from calibre import isbytestring, force_unicode
 from calibre.utils.config_base import tweaks
 from calibre.utils.icu import sort_key
+from polyglot.builtins import string_or_bytes
 
 
 class Book(Metadata):
@@ -291,7 +292,7 @@ class CollectionsBookList(BookList):
                 return 1
             if y is None:
                 return -1
-            if isinstance(x, basestring) and isinstance(y, basestring):
+            if isinstance(x, string_or_bytes) and isinstance(y, string_or_bytes):
                 x, y = sort_key(force_unicode(x)), sort_key(force_unicode(y))
             try:
                 c = cmp(x, y)

src/calibre/devices/usbms/device.py

@@ -23,6 +23,7 @@ from calibre.devices.errors import DeviceError
 from calibre.devices.usbms.deviceconfig import DeviceConfig
 from calibre.constants import iswindows, islinux, isosx, isfreebsd, plugins
 from calibre.utils.filenames import ascii_filename as sanitize
+from polyglot.builtins import string_or_bytes
 
 if isosx:
     usbobserver, usbobserver_err = plugins['usbobserver']
@@ -938,7 +939,7 @@ class Device(DeviceConfig, DevicePlugin):
         sanity_check(on_card, files, self.card_prefix(), self.free_space())
 
         def get_dest_dir(prefix, candidates):
-            if isinstance(candidates, basestring):
+            if isinstance(candidates, string_or_bytes):
                 candidates = [candidates]
             if not candidates:
                 candidates = ['']

src/calibre/devices/usbms/driver.py

@@ -20,7 +20,7 @@ from calibre.devices.usbms.cli import CLI
 from calibre.devices.usbms.device import Device
 from calibre.devices.usbms.books import BookList, Book
 from calibre.ebooks.metadata.book.json_codec import JsonCodec
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 BASE_TIME = None
 
@@ -243,7 +243,7 @@ class USBMS(CLI, Device):
                     import traceback
                     traceback.print_exc()
             return changed
-        if isinstance(ebook_dirs, basestring):
+        if isinstance(ebook_dirs, string_or_bytes):
             ebook_dirs = [ebook_dirs]
         for ebook_dir in ebook_dirs:
             ebook_dir = self.path_to_unicode(ebook_dir)

src/calibre/ebooks/BeautifulSoup.py

@@ -89,7 +89,7 @@ import types
 import re
 import calibre.ebooks.sgmllib as sgmllib
 from htmlentitydefs import name2codepoint
-from polyglot.builtins import codepoint_to_chr, unicode_type
+from polyglot.builtins import codepoint_to_chr, unicode_type, string_or_bytes
 
 #This hack makes Beautiful Soup able to parse XML with namespaces
 sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*')
@@ -177,7 +177,7 @@ class PageElement:
         return lastChild
 
     def insert(self, position, newChild):
-        if (isinstance(newChild, basestring)
+        if (isinstance(newChild, string_or_bytes)
             or isinstance(newChild, unicode_type)) \
             and not isinstance(newChild, NavigableString):
             newChild = NavigableString(newChild)
@@ -937,7 +937,7 @@ def isString(s):
     """Convenience method that works with all 2.x versions of Python
     to determine whether or not something is stringlike."""
     try:
-        return isinstance(s, unicode_type) or isinstance(s, basestring)
+        return isinstance(s, unicode_type) or isinstance(s, string_or_bytes)
     except NameError:
         return isinstance(s, str)
 

src/calibre/ebooks/conversion/plumber.py

@@ -18,7 +18,7 @@ from calibre.utils.zipfile import ZipFile
 from calibre import (extract, walk, isbytestring, filesystem_encoding,
         get_types_map)
 from calibre.constants import __version__
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 DEBUG_README=u'''
 This debug directory contains snapshots of the e-book as it passes through the
@@ -1022,7 +1022,7 @@ OptionRecommendation(name='search_replace',
 
     def dump_input(self, ret, output_dir):
         out_dir = os.path.join(self.opts.debug_pipeline, 'input')
-        if isinstance(ret, basestring):
+        if isinstance(ret, string_or_bytes):
             shutil.copytree(output_dir, out_dir)
         else:
             if not os.path.exists(out_dir):
@@ -1214,7 +1214,7 @@ OptionRecommendation(name='search_replace',
         transform_css_rules = ()
         if self.opts.transform_css_rules:
             transform_css_rules = self.opts.transform_css_rules
-            if isinstance(transform_css_rules, basestring):
+            if isinstance(transform_css_rules, string_or_bytes):
                 transform_css_rules = json.loads(transform_css_rules)
         flattener = CSSFlattener(fbase=fbase, fkey=fkey,
                 lineh=line_height,

src/calibre/ebooks/covers.py

@@ -10,7 +10,7 @@ import re, random, unicodedata
 from collections import namedtuple
 from contextlib import contextmanager
 from math import ceil, sqrt, cos, sin, atan2
-from polyglot.builtins import map, zip
+from polyglot.builtins import map, zip, string_or_bytes
 from itertools import chain
 
 from PyQt5.Qt import (
@@ -275,7 +275,7 @@ def format_fields(mi, prefs):
 
 @contextmanager
 def preserve_fields(obj, fields):
-    if isinstance(fields, basestring):
+    if isinstance(fields, string_or_bytes):
         fields = fields.split()
     null = object()
     mem = {f:getattr(obj, f, null) for f in fields}

src/calibre/ebooks/docx/writer/from_html.py

@@ -19,7 +19,7 @@ from calibre.ebooks.docx.writer.lists import ListsManager
 from calibre.ebooks.oeb.stylizer import Stylizer as Sz, Style as St
 from calibre.ebooks.oeb.base import XPath, barename
 from calibre.utils.localization import lang_as_iso639_1
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 def lang_for_tag(tag):
@@ -528,7 +528,7 @@ class Convert(object):
                     self.add_block_tag(tagname, html_tag, tag_style, stylizer, float_spec=float_spec)
 
             for child in html_tag.iterchildren():
-                if isinstance(getattr(child, 'tag', None), basestring):
+                if isinstance(getattr(child, 'tag', None), string_or_bytes):
                     self.process_tag(child, stylizer, float_spec=float_spec)
                 else:  # Comment/PI/etc.
                     tail = getattr(child, 'tail', None)

src/calibre/ebooks/fb2/fb2ml.py

@@ -19,7 +19,7 @@ from calibre.constants import __appname__, __version__
 from calibre.utils.localization import lang_as_iso639_1
 from calibre.utils.img import save_cover_data_to
 from calibre.ebooks.oeb.base import urlnormalize
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class FB2MLizer(object):
@@ -396,9 +396,9 @@ class FB2MLizer(object):
         elem = elem_tree
 
         # Ensure what we are converting is not a string and that the fist tag is part of the XHTML namespace.
-        if not isinstance(elem_tree.tag, basestring) or namespace(elem_tree.tag) != XHTML_NS:
+        if not isinstance(elem_tree.tag, string_or_bytes) or namespace(elem_tree.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return []

src/calibre/ebooks/htmlz/oeb2html.py

@@ -22,7 +22,7 @@ from calibre.ebooks.oeb.base import (
     XHTML, XHTML_NS, barename, namespace, OEB_IMAGES, XLINK, rewrite_links, urlnormalize)
 from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.utils.logging import default_log
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 SELF_CLOSING_TAGS = {'area', 'base', 'basefont', 'br', 'hr', 'input', 'img', 'link', 'meta'}
 
@@ -98,7 +98,7 @@ class OEB2HTML(object):
                 for el in root.iter():
                     attribs = el.attrib
                     try:
-                        if not isinstance(el.tag, basestring):
+                        if not isinstance(el.tag, string_or_bytes):
                             continue
                     except:
                         continue
@@ -159,10 +159,10 @@ class OEB2HTMLNoCSSizer(OEB2HTML):
         '''
 
         # We can only processes tags. If there isn't a tag return any text.
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']
@@ -248,10 +248,10 @@ class OEB2HTMLInlineCSSizer(OEB2HTML):
         '''
 
         # We can only processes tags. If there isn't a tag return any text.
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']
@@ -351,10 +351,10 @@ class OEB2HTMLClassCSSizer(OEB2HTML):
         '''
 
         # We can only processes tags. If there isn't a tag return any text.
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']

src/calibre/ebooks/lit/reader.py

@@ -22,7 +22,7 @@ from calibre.ebooks.oeb.base import urlnormalize, xpath
 from calibre.ebooks.oeb.reader import OEBReader
 from calibre.ebooks import DRMError
 from calibre import plugins
-from polyglot.builtins import codepoint_to_chr, unicode_type
+from polyglot.builtins import codepoint_to_chr, unicode_type, string_or_bytes
 
 lzx, lxzerror = plugins['lzx']
 msdes, msdeserror = plugins['msdes']
@@ -281,7 +281,7 @@ class UnBinary(object):
                         attr = current_map[oc]
                     elif oc in self.attr_map:
                         attr = self.attr_map[oc]
-                    if not attr or not isinstance(attr, basestring):
+                    if not attr or not isinstance(attr, string_or_bytes):
                         raise LitError(
                             'Unknown attribute %d in tag %s' % (oc, tag_name))
                     if attr.startswith('%'):

src/calibre/ebooks/lit/writer.py

@@ -31,7 +31,7 @@ import calibre
 from calibre import plugins
 msdes, msdeserror = plugins['msdes']
 import calibre.ebooks.lit.mssha1 as mssha1
-from polyglot.builtins import codepoint_to_chr, unicode_type
+from polyglot.builtins import codepoint_to_chr, unicode_type, string_or_bytes
 
 __all__ = ['LitWriter']
 
@@ -175,7 +175,7 @@ class ReBinary(object):
 
     def tree_to_binary(self, elem, nsrmap=NSRMAP, parents=[],
                        inhead=False, preserve=False):
-        if not isinstance(elem.tag, basestring):
+        if not isinstance(elem.tag, string_or_bytes):
             # Don't emit any comments or raw entities
             return
         nsrmap = copy.copy(nsrmap)

src/calibre/ebooks/lrf/html/table.py

@@ -7,6 +7,7 @@ from calibre.ebooks.lrf.fonts import get_font
 from calibre.ebooks.lrf.pylrs.pylrs import TextBlock, Text, CR, Span, \
                                              CharButton, Plot, Paragraph, \
                                              LrsTextTag
+from polyglot.builtins import string_or_bytes
 
 
 def ceil(num):
@@ -38,7 +39,7 @@ def tokens(tb):
             yield 2, None
         elif isinstance(x, Text):
             yield x.text, cattrs(attrs, {})
-        elif isinstance(x, basestring):
+        elif isinstance(x, string_or_bytes):
             yield x, cattrs(attrs, {})
         elif isinstance(x, (CharButton, LrsTextTag)):
             if x.contents:
@@ -382,7 +383,3 @@ class Table(object):
 
                     yield tb, xpos[c], sypos, delta, None
                     sypos += tb.blockStyle.attrs['blockheight']
-
-
-
-

src/calibre/ebooks/lrf/lrs/convert_from.py

@@ -17,6 +17,7 @@ from calibre.ebooks.lrf.pylrs.pylrs import Book, PageStyle, TextStyle, \
             Italic, Sup, Sub, Bold, EmpLine, JumpButton, CharButton, Plot, \
             DropCaps, Footer, RuledLine
 from calibre.ebooks.chardet import xml_to_unicode
+from polyglot.builtins import string_or_bytes
 
 
 class LrsParser(object):
@@ -95,7 +96,7 @@ class LrsParser(object):
             if isinstance(contents[0], NavigableString):
                 contents[0] = contents[0].string.lstrip()
             for item in contents:
-                if isinstance(item, basestring):
+                if isinstance(item, string_or_bytes):
                     p.append(item)
                 elif isinstance(item, NavigableString):
                     p.append(item.string)

src/calibre/ebooks/lrf/objects.py

@@ -7,7 +7,7 @@ from calibre.ebooks.lrf import LRFParseError, PRS500_PROFILE
 from calibre.constants import ispy3
 from calibre import entity_to_unicode, prepare_string_for_xml
 from calibre.ebooks.lrf.tags import Tag
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 ruby_tags = {
         0xF575: ['rubyAlignAndAdjust', 'W'],
@@ -121,7 +121,7 @@ class LRFContentObject(LRFObject):
     def handle_tag(self, tag):
         if tag.id in self.tag_map:
             action = self.tag_map[tag.id]
-            if isinstance(action, basestring):
+            if isinstance(action, string_or_bytes):
                 func, args = action, tuple([])
             else:
                 func, args = action[0], (action[1],)
@@ -880,7 +880,7 @@ class Text(LRFStream):
                 self.add_text(stream.read(tag.word))
             elif tag.id in self.__class__.text_tags:  # A Text tag
                 action = self.__class__.text_tags[tag.id]
-                if isinstance(action, basestring):
+                if isinstance(action, string_or_bytes):
                     getattr(self, action)(tag, stream)
                 else:
                     getattr(self, action[0])(tag, action[1])
@@ -904,7 +904,7 @@ class Text(LRFStream):
         s = u''
         open_containers = collections.deque()
         for c in self.content:
-            if isinstance(c, basestring):
+            if isinstance(c, string_or_bytes):
                 s += prepare_string_for_xml(c).replace('\0', '')
             elif c is None:
                 if open_containers:
@@ -930,7 +930,7 @@ class Text(LRFStream):
         open_containers = collections.deque()
         in_p = False
         for c in self.content:
-            if isinstance(c, basestring):
+            if isinstance(c, string_or_bytes):
                 s += c
             elif c is None:
                 if c.name == 'P':

src/calibre/ebooks/lrf/pylrs/elements.py

@@ -1,6 +1,6 @@
 """ elements.py -- replacements and helpers for ElementTree """
 
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class ElementWriter(object):
@@ -24,7 +24,7 @@ class ElementWriter(object):
 
     def _writeAttribute(self, f, name, value):
         f.write(u' %s="' % unicode_type(name))
-        if not isinstance(value, basestring):
+        if not isinstance(value, string_or_bytes):
             value = unicode_type(value)
         value = self._encodeCdata(value)
         value = value.replace('"', '"')

src/calibre/ebooks/lrf/pylrs/pylrs.py

@@ -58,6 +58,7 @@ DEFAULT_GENREADING      = "fs"          # default is yes to both lrf and lrs
 
 from calibre import __appname__, __version__
 from calibre import entity_to_unicode
+from polyglot.builtins import string_or_bytes, unicode_type
 
 
 class LrsError(Exception):
@@ -96,7 +97,7 @@ def ElementWithReading(tag, text, reading=False):
 
     if text is None:
         readingText = ""
-    elif isinstance(text, basestring):
+    elif isinstance(text, string_or_bytes):
         readingText = text
     else:
         # assumed to be a sequence of (name, sortas)
@@ -155,7 +156,7 @@ class Delegator(object):
 
             """
             for setting in d.getSettings():
-                if isinstance(setting, basestring):
+                if isinstance(setting, string_or_bytes):
                     setting = (d, setting)
                 delegates = \
                         self.delegatedSettingsDict.setdefault(setting[1], [])
@@ -293,7 +294,7 @@ class LrsContainer(object):
                     (content.__class__.__name__,
                     self.__class__.__name__))
 
-        if convertText and isinstance(content, basestring):
+        if convertText and isinstance(content, string_or_bytes):
             content = Text(content)
 
         content.setParent(self)
@@ -587,15 +588,15 @@ class Book(Delegator):
             ts.attrs['baselineskip'] = rescale(ts.attrs['baselineskip'])
 
     def renderLrs(self, lrsFile, encoding="UTF-8"):
-        if isinstance(lrsFile, basestring):
+        if isinstance(lrsFile, string_or_bytes):
             lrsFile = codecs.open(lrsFile, "wb", encoding=encoding)
         self.render(lrsFile, outputEncodingName=encoding)
         lrsFile.close()
 
     def renderLrf(self, lrfFile):
         self.appendReferencedObjects(self)
-        if isinstance(lrfFile, basestring):
+        if isinstance(lrfFile, string_or_bytes):
-            lrfFile = file(lrfFile, "wb")
+            lrfFile = open(lrfFile, "wb")
         lrfWriter = LrfWriter(self.sourceencoding)
 
         lrfWriter.optimizeTags = self.optimizeTags
@@ -1493,9 +1494,9 @@ class Paragraph(LrsContainer):
 
     def __init__(self, text=None):
         LrsContainer.__init__(self, [Text, CR, DropCaps, CharButton,
-                                     LrsSimpleChar1, basestring])
+                                     LrsSimpleChar1, bytes, unicode_type])
         if text is not None:
-            if isinstance(text, basestring):
+            if isinstance(text, string_or_bytes):
                 text = Text(text)
             self.append(text)
 
@@ -1528,7 +1529,7 @@ class Paragraph(LrsContainer):
 class LrsTextTag(LrsContainer):
 
     def __init__(self, text, validContents):
-        LrsContainer.__init__(self, [Text, basestring] + validContents)
+        LrsContainer.__init__(self, [Text, bytes, unicode_type] + validContents)
         if text is not None:
             self.append(text)
 
@@ -1792,7 +1793,7 @@ class Box(LrsSimpleChar1, LrsContainer):
     """
 
     def __init__(self, linetype="solid"):
-        LrsContainer.__init__(self, [Text, basestring])
+        LrsContainer.__init__(self, [Text, bytes, unicode_type])
         if linetype not in LINE_TYPE_ENCODING:
             raise LrsError(linetype + " is not a valid line type")
         self.linetype = linetype
@@ -1812,9 +1813,9 @@ class Box(LrsSimpleChar1, LrsContainer):
 class Span(LrsSimpleChar1, LrsContainer):
 
     def __init__(self, text=None, **attrs):
-        LrsContainer.__init__(self, [LrsSimpleChar1, Text, basestring])
+        LrsContainer.__init__(self, [LrsSimpleChar1, Text, bytes, unicode_type])
         if text is not None:
-            if isinstance(text, basestring):
+            if isinstance(text, string_or_bytes):
                 text = Text(text)
             self.append(text)
 
@@ -1956,7 +1957,7 @@ class CharButton(LrsSimpleChar1, LrsContainer):
     """
 
     def __init__(self, button, text=None):
-        LrsContainer.__init__(self, [basestring, Text, LrsSimpleChar1])
+        LrsContainer.__init__(self, [bytes, unicode_type, Text, LrsSimpleChar1])
         self.button = None
         if button is not None:
             self.setButton(button)

src/calibre/ebooks/lrf/tags.py

@@ -5,7 +5,7 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 import struct
 
 from calibre.ebooks.lrf import LRFParseError
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Tag(object):
@@ -196,7 +196,7 @@ class Tag(object):
         self.id = 0xF500 + tag_id[0]
 
         size, self.name = self.__class__.tags[tag_id[0]]
-        if isinstance(size, basestring):
+        if isinstance(size, string_or_bytes):
             parser = getattr(self, size + '_parser')
             self.contents = parser(stream)
         else:

src/calibre/ebooks/metadata/odt.py

@@ -33,6 +33,7 @@ from calibre.ebooks.metadata import MetaInformation, string_to_authors, check_is
 from calibre.utils.imghdr import identify
 from calibre.utils.date import parse_date
 from calibre.utils.localization import canonicalize_lang
+from polyglot.builtins import string_or_bytes
 
 whitespace = re.compile(r'\s+')
 
@@ -125,7 +126,7 @@ class odfmetaparser(xml.sax.saxutils.XMLGenerator):
         if name == (OFFICENS,u'meta'):
             for k,v in self.addfields.items():
                 if len(v) > 0:
-                    if isinstance(k, basestring):
+                    if isinstance(k, string_or_bytes):
                         xml.sax.saxutils.XMLGenerator.startElementNS(self,(METANS,u'user-defined'),None,{(METANS,u'name'):k})
                         xml.sax.saxutils.XMLGenerator.characters(self, v)
                         xml.sax.saxutils.XMLGenerator.endElementNS(self, (METANS,u'user-defined'),None)
@@ -270,4 +271,3 @@ def read_cover(stream, zin, mi, opfmeta, extract_cover):
                 else:
                     cover_data = (fmt, raw)
             mi.cover_data = cover_data
-

src/calibre/ebooks/metadata/rtf.py

@@ -8,7 +8,7 @@ import re, cStringIO, codecs
 
 from calibre import force_unicode
 from calibre.ebooks.metadata import MetaInformation, string_to_authors
-from polyglot.builtins import codepoint_to_chr, unicode_type
+from polyglot.builtins import codepoint_to_chr, unicode_type, string_or_bytes
 
 title_pat    = re.compile(r'\{\\info.*?\{\\title(.*?)(?<!\\)\}', re.DOTALL)
 author_pat   = re.compile(r'\{\\info.*?\{\\author(.*?)(?<!\\)\}', re.DOTALL)
@@ -146,7 +146,7 @@ def create_metadata(stream, options):
         md.append(r'{\title %s}'%(title,))
     if options.authors:
         au = options.authors
-        if not isinstance(au, basestring):
+        if not isinstance(au, string_or_bytes):
             au = u', '.join(au)
         author = encode(au)
         md.append(r'{\author %s}'%(author,))

src/calibre/ebooks/metadata/xmp.py

@@ -19,6 +19,7 @@ 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 string_or_bytes
 
 _xml_declaration = re.compile(r'<\?xml[^<>]+encoding\s*=\s*[\'"](.*?)[\'"][^<>]*>', re.IGNORECASE)
 
@@ -49,6 +50,7 @@ def expand(name):
     prefix, name = name.partition(':')[::2]
     return '{%s}%s' % (NS_MAP[prefix], name)
 
+
 xpath_cache = {}
 
 
@@ -478,7 +480,7 @@ def metadata_to_xmp_packet(mi):
         'authors':('dc:creator', True), 'tags':('dc:subject', False), 'publisher':('dc:publisher', False),
     }.iteritems():
         val = mi.get(prop) or ()
-        if isinstance(val, basestring):
+        if isinstance(val, string_or_bytes):
             val = [val]
         create_sequence_property(dc, tag, val, ordered)
     if not mi.is_null('pubdate'):
@@ -636,10 +638,10 @@ def merge_xmp_packet(old, new):
 
     return serialize_xmp_packet(root)
 
+
 if __name__ == '__main__':
     from calibre.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)
-    print (merge_xmp_packet(xmp_packet, np))
+    print(merge_xmp_packet(xmp_packet, np))
-

src/calibre/ebooks/mobi/mobiml.py

@@ -16,7 +16,7 @@ from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.ebooks.oeb.transforms.flatcss import KeyMapper
 from calibre.ebooks.mobi.utils import convert_color_for_font_tag
 from calibre.utils.imghdr import identify
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 MBP_NS = 'http://mobipocket.com/ns/mbp'
 
@@ -144,7 +144,7 @@ class MobiMLizer(object):
         return self.fnums[self.fmap[ptsize]]
 
     def mobimlize_measure(self, ptsize):
-        if isinstance(ptsize, basestring):
+        if isinstance(ptsize, string_or_bytes):
             return ptsize
         embase = self.profile.fbase
         if round(ptsize) < embase:
@@ -187,7 +187,7 @@ class MobiMLizer(object):
             parent = bstate.nested[-1] if bstate.nested else bstate.body
             indent = istate.indent
             left = istate.left
-            if isinstance(indent, basestring):
+            if isinstance(indent, string_or_bytes):
                 indent = 0
             if indent < 0 and abs(indent) < left:
                 left += indent
@@ -308,7 +308,7 @@ class MobiMLizer(object):
         inline = bstate.inline
         content = self.preize_text(text, pre_wrap=istate.pre_wrap) if istate.preserve or istate.pre_wrap else [text]
         for item in content:
-            if isinstance(item, basestring):
+            if isinstance(item, string_or_bytes):
                 if len(inline) == 0:
                     inline.text = (inline.text or '') + item
                 else:
@@ -319,7 +319,7 @@ class MobiMLizer(object):
 
     def mobimlize_elem(self, elem, stylizer, bstate, istates,
             ignore_valign=False):
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             return
         style = stylizer.style(elem)

src/calibre/ebooks/mobi/writer2/serializer.py

@@ -18,7 +18,7 @@ from calibre.ebooks.mobi.utils import is_guide_ref_start
 from calibre.ebooks.oeb.base import (
     OEB_DOCS, XHTML, XHTML_NS, XML_NS, namespace, prefixname, urlnormalize
 )
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Buf(BytesIO):
@@ -310,7 +310,7 @@ class Serializer(object):
 
     def serialize_elem(self, elem, item, nsrmap=NSRMAP):
         buf = self.buf
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
             or namespace(elem.tag) not in nsrmap:
             return
         tag = prefixname(elem.tag, nsrmap)

src/calibre/ebooks/odt/input.py

@@ -19,7 +19,7 @@ from odf.namespaces import TEXTNS as odTEXTNS
 
 from calibre import CurrentDir, walk
 from calibre.ebooks.oeb.base import _css_logger
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Extract(ODF2XHTML):
@@ -249,7 +249,7 @@ class Extract(ODF2XHTML):
         # first load the odf structure
         self.lines = []
         self._wfunc = self._wlines
-        if isinstance(odffile, basestring) \
+        if isinstance(odffile, string_or_bytes) \
                 or hasattr(odffile, 'read'):  # Added by Kovid
             self.document = odLoad(odffile)
         else:

src/calibre/ebooks/oeb/base.py

@@ -22,7 +22,7 @@ from calibre.ebooks.oeb.parse_utils import (barename, XHTML_NS, RECOVER_PARSER,
         namespace, XHTML, parse_html, NotHTML)
 from calibre.utils.cleantext import clean_xml_chars
 from calibre.utils.short_uuid import uuid4
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 XML_NS       = 'http://www.w3.org/XML/1998/namespace'
 OEB_DOC_NS   = 'http://openebook.org/namespaces/oeb-document/1.0/'
@@ -1055,7 +1055,7 @@ class Manifest(object):
                     mt = self.media_type.lower()
                 except Exception:
                     mt = 'application/octet-stream'
-                if not isinstance(data, basestring):
+                if not isinstance(data, string_or_bytes):
                     pass  # already parsed
                 elif mt in OEB_DOCS:
                     data = self._parse_xhtml(data)
@@ -1318,7 +1318,7 @@ class Spine(object):
         self.page_progression_direction = None
 
     def _linear(self, linear):
-        if isinstance(linear, basestring):
+        if isinstance(linear, string_or_bytes):
             linear = linear.lower()
         if linear is None or linear in ('yes', 'true'):
             linear = True

src/calibre/ebooks/oeb/normalize_css.py

@@ -6,7 +6,7 @@ from __future__ import (unicode_literals, division, absolute_import,
 __license__ = 'GPL v3'
 __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 
-from polyglot.builtins import zip
+from polyglot.builtins import zip, string_or_bytes
 from functools import wraps
 
 from css_parser.css import PropertyValue
@@ -127,10 +127,10 @@ def normalize_font(cssvalue, font_family_as_list=False):
         ans = {k:DEFAULTS[k] for k in composition}
         ans.update(parse_font(val))
     if font_family_as_list:
-        if isinstance(ans['font-family'], basestring):
+        if isinstance(ans['font-family'], string_or_bytes):
             ans['font-family'] = [x.strip() for x in ans['font-family'].split(',')]
     else:
-        if not isinstance(ans['font-family'], basestring):
+        if not isinstance(ans['font-family'], string_or_bytes):
             ans['font-family'] = serialize_font_family(ans['font-family'])
     return ans
 

src/calibre/ebooks/oeb/parse_utils.py

@@ -14,7 +14,7 @@ from lxml import etree, html
 from calibre import xml_replace_entities, force_unicode
 from calibre.constants import filesystem_encoding
 from calibre.ebooks.chardet import xml_to_unicode, strip_encoding_declarations
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 RECOVER_PARSER = etree.XMLParser(recover=True, no_network=True)
 XHTML_NS     = 'http://www.w3.org/1999/xhtml'
@@ -99,7 +99,7 @@ def html5_parse(data, max_nesting_depth=100):
     # Check that the asinine HTML 5 algorithm did not result in a tree with
     # insane nesting depths
     for x in data.iterdescendants():
-        if isinstance(x.tag, basestring) and not len(x):  # Leaf node
+        if isinstance(x.tag, string_or_bytes) and not len(x):  # Leaf node
             depth = node_depth(x)
             if depth > max_nesting_depth:
                 raise ValueError('HTML 5 parsing resulted in a tree with nesting'
@@ -310,7 +310,7 @@ def parse_html(data, log=None, decoder=None, preprocessor=None,
         nroot = etree.Element(XHTML('html'),
             nsmap={None: XHTML_NS}, attrib=attrib)
         for elem in data.iterdescendants():
-            if isinstance(elem.tag, basestring) and \
+            if isinstance(elem.tag, string_or_bytes) and \
                 namespace(elem.tag) == ns:
                 elem.tag = XHTML(barename(elem.tag))
         for elem in data:

src/calibre/ebooks/oeb/polish/embed.py

@@ -15,13 +15,14 @@ from lxml import etree
 from calibre import prints
 from calibre.ebooks.oeb.base import XHTML
 from calibre.utils.filenames import ascii_filename
+from polyglot.builtins import string_or_bytes
 
 props = {'font-family':None, 'font-weight':'normal', 'font-style':'normal', 'font-stretch':'normal'}
 
 
 def matching_rule(font, rules):
     ff = font['font-family']
-    if not isinstance(ff, basestring):
+    if not isinstance(ff, string_or_bytes):
         ff = tuple(ff)[0]
     family = icu_lower(ff)
     wt = font['font-weight']
@@ -31,7 +32,7 @@ def matching_rule(font, rules):
     for rule in rules:
         if rule['font-style'] == style and rule['font-stretch'] == stretch and rule['font-weight'] == wt:
             ff = rule['font-family']
-            if not isinstance(ff, basestring):
+            if not isinstance(ff, string_or_bytes):
                 ff = tuple(ff)[0]
             if icu_lower(ff) == family:
                 return rule
@@ -162,7 +163,7 @@ def do_embed(container, font, report):
 def embed_font(container, font, all_font_rules, report, warned):
     rule = matching_rule(font, all_font_rules)
     ff = font['font-family']
-    if not isinstance(ff, basestring):
+    if not isinstance(ff, string_or_bytes):
         ff = ff[0]
     if rule is None:
         from calibre.utils.fonts.scanner import font_scanner, NoFonts

src/calibre/ebooks/oeb/polish/split.py

@@ -7,7 +7,7 @@ __license__ = 'GPL v3'
 __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
 
 import copy, os, re
-from polyglot.builtins import map
+from polyglot.builtins import map, string_or_bytes
 from urlparse import urlparse
 
 from calibre.ebooks.oeb.base import barename, XPNSMAP, XPath, OPF, XHTML, OEB_DOCS
@@ -395,9 +395,9 @@ def merge_html(container, names, master):
 
         first_child = ''
         for first_child in children:
-            if not isinstance(first_child, basestring):
+            if not isinstance(first_child, string_or_bytes):
                 break
-        if isinstance(first_child, basestring):
+        if isinstance(first_child, string_or_bytes):
             # body contained only text, no tags
             first_child = body.makeelement(XHTML('p'))
             first_child.text, children[0] = children[0], first_child
@@ -429,7 +429,7 @@ def merge_html(container, names, master):
                 a.set('href', '#' + amap[q])
 
         for child in children:
-            if isinstance(child, basestring):
+            if isinstance(child, string_or_bytes):
                 add_text(master_body, child)
             else:
                 master_body.append(copy.deepcopy(child))

src/calibre/ebooks/oeb/transforms/flatcss.py

@@ -21,7 +21,7 @@ from calibre.ebooks.oeb.base import (XHTML, XHTML_NS, CSS_MIME, OEB_STYLES,
 from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.utils.filenames import ascii_filename, ascii_text
 from calibre.utils.icu import numeric_sort_key
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 COLLAPSE = re.compile(r'[ \t\r\n\v]+')
 STRIPNUM = re.compile(r'[-0-9]+$')
@@ -367,7 +367,7 @@ class CSSFlattener(object):
                     cssdict[property] = "%0.5fem" % (value / fsize)
 
     def flatten_node(self, node, stylizer, names, styles, pseudo_styles, psize, item_id):
-        if not isinstance(node.tag, basestring) \
+        if not isinstance(node.tag, string_or_bytes) \
            or namespace(node.tag) != XHTML_NS:
             return
         tag = barename(node.tag)

src/calibre/ebooks/oeb/transforms/manglecase.py

@@ -11,6 +11,7 @@ from calibre.ebooks.oeb.base import XHTML, XHTML_NS
 from calibre.ebooks.oeb.base import CSS_MIME
 from calibre.ebooks.oeb.base import namespace
 from calibre.ebooks.oeb.stylizer import Stylizer
+from polyglot.builtins import string_or_bytes
 
 CASE_MANGLER_CSS = """
 .calibre_lowercase {
@@ -95,7 +96,7 @@ class CaseMangler(object):
                 last = child
 
     def mangle_elem(self, elem, stylizer):
-        if not isinstance(elem.tag, basestring) or \
+        if not isinstance(elem.tag, string_or_bytes) or \
            namespace(elem.tag) != XHTML_NS:
             return
         children = list(elem)

src/calibre/ebooks/pml/pmlml.py

@@ -14,7 +14,7 @@ from lxml import etree
 
 from calibre.ebooks.pdb.ereader import image_name
 from calibre.ebooks.pml import unipmlcode
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 TAG_MAP = {
     'b'       : 'B',
@@ -224,9 +224,9 @@ class PMLMLizer(object):
     def dump_text(self, elem, stylizer, page, tag_stack=[]):
         from calibre.ebooks.oeb.base import XHTML_NS, barename, namespace
 
-        if not isinstance(elem.tag, basestring) or namespace(elem.tag) != XHTML_NS:
+        if not isinstance(elem.tag, string_or_bytes) or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return []

src/calibre/ebooks/rb/rbml.py

@@ -12,6 +12,7 @@ import re
 
 from calibre import prepare_string_for_xml
 from calibre.ebooks.rb import unique_name
+from polyglot.builtins import string_or_bytes
 
 TAGS = [
     'b',
@@ -142,9 +143,9 @@ class RBMLizer(object):
     def dump_text(self, elem, stylizer, page, tag_stack=[]):
         from calibre.ebooks.oeb.base import XHTML_NS, barename, namespace
 
-        if not isinstance(elem.tag, basestring) or namespace(elem.tag) != XHTML_NS:
+        if not isinstance(elem.tag, string_or_bytes) or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return [u'']

src/calibre/ebooks/rtf/rtfml.py

@@ -17,7 +17,7 @@ from lxml import etree
 from calibre.ebooks.metadata import authors_to_string
 from calibre.utils.img import save_cover_data_to
 from calibre.utils.imghdr import identify
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 TAGS = {
     'b': '\\b',
@@ -224,10 +224,10 @@ class RTFMLizer(object):
         from calibre.ebooks.oeb.base import (XHTML_NS, namespace, barename,
                 urlnormalize)
 
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return elem.tail
             return u''

src/calibre/ebooks/snb/snbml.py

@@ -12,7 +12,7 @@ import os
 import re
 
 from lxml import etree
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 def ProcessFileName(fileName):
@@ -212,10 +212,10 @@ class SNBMLizer(object):
     def dump_text(self, subitems, elem, stylizer, end='', pre=False, li=''):
         from calibre.ebooks.oeb.base import XHTML_NS, barename, namespace
 
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']

src/calibre/ebooks/txt/markdownml.py

@@ -15,7 +15,7 @@ from functools import partial
 from calibre.ebooks.htmlz.oeb2html import OEB2HTML
 from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace, rewrite_links
 from calibre.ebooks.oeb.stylizer import Stylizer
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class MarkdownMLizer(OEB2HTML):
@@ -111,10 +111,10 @@ class MarkdownMLizer(OEB2HTML):
         '''
 
         # We can only processes tags. If there isn't a tag return any text.
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']

src/calibre/ebooks/txt/textileml.py

@@ -16,6 +16,7 @@ from calibre.ebooks.oeb.base import XHTML, XHTML_NS, barename, namespace, rewrit
 from calibre.ebooks.oeb.stylizer import Stylizer
 from calibre.ebooks import unit_convert
 from calibre.ebooks.textile.unsmarten import unsmarten
+from polyglot.builtins import string_or_bytes
 
 
 class TextileMLizer(OEB2HTML):
@@ -72,9 +73,9 @@ class TextileMLizer(OEB2HTML):
                 # I'm not checking for duplicated spans '%' as any that follow each other were being incorrectly merged
                 txt = '%s' % t
                 if txt != '%':
-                    text = re.sub(r'([^'+t+'|^\n])'+t+'\]\['+t+'([^'+t+'])', r'\1\2', text)
+                    text = re.sub(r'([^'+t+'|^\n])'+t+r'\]\['+t+'([^'+t+'])', r'\1\2', text)
                     text = re.sub(r'([^'+t+'|^\n])'+t+t+'([^'+t+'])', r'\1\2', text)
-                text = re.sub(r'(\s|[*_\'"])\[('+t+'[a-zA-Z0-9 \'",.*_]+'+t+')\](\s|[*_\'"?!,.])', r'\1\2\3', text)
+                text = re.sub(r'(\s|[*_\'"])\[('+t+'[a-zA-Z0-9 \'",.*_]+'+t+r')\](\s|[*_\'"?!,.])', r'\1\2\3', text)
             return text
 
         # Now tidyup links and ids - remove ones that don't have a correponding opposite
@@ -82,13 +83,13 @@ class TextileMLizer(OEB2HTML):
             for i in self.our_links:
                 if i[0] == '#':
                     if i not in self.our_ids:
-                        text = re.sub(r'"(.+)":'+i+'(\s)', r'\1\2', text)
+                        text = re.sub(r'"(.+)":'+i+r'(\s)', r'\1\2', text)
             for i in self.our_ids:
                 if i not in self.our_links:
-                    text = re.sub(r'%?\('+i+'\)\xa0?%?', r'', text)
+                    text = re.sub(r'%?\('+i+'\\)\xa0?%?', r'', text)
 
         # Remove obvious non-needed escaping, add sub/sup-script ones
-        text = check_escaping(text, ['\*', '_', '\*'])
+        text = check_escaping(text, [r'\*', '_', r'\*'])
         # escape the super/sub-scripts if needed
         text = re.sub(r'(\w)([~^]\w+[~^])', r'\1[\2]', text)
         # escape the super/sub-scripts if needed
@@ -111,16 +112,16 @@ class TextileMLizer(OEB2HTML):
 
         # reduce blank lines
         text = re.sub(r'\n{3}', r'\n\np. \n\n', text)
-        text = re.sub(u'%\n(p[<>=]{1,2}\.|p\.)', r'%\n\n\1', text)
+        text = re.sub(u'%\n(p[<>=]{1,2}\\.|p\\.)', r'%\n\n\1', text)
         # Check span following blank para
         text = re.sub(r'\n+ +%', r' %', text)
-        text = re.sub(u'p[<>=]{1,2}\.\n\n?', r'', text)
+        text = re.sub(u'p[<>=]{1,2}\\.\n\n?', r'', text)
         # blank paragraph
         text = re.sub(r'\n(p.*\.)\n', r'\n\1 \n\n', text)
         # blank paragraph
         text = re.sub(u'\n\xa0', r'\np. ', text)
         # blank paragraph
-        text = re.sub(u'\np[<>=]{1,2}?\. \xa0', r'\np. ', text)
+        text = re.sub(u'\np[<>=]{1,2}?\\. \xa0', r'\np. ', text)
         text = re.sub(r'(^|\n)(p.*\. ?\n)(p.*\.)', r'\1\3', text)
         text = re.sub(r'\n(p\. \n)(p.*\.|h.*\.)', r'\n\2', text)
         # sort out spaces in tables
@@ -225,10 +226,10 @@ class TextileMLizer(OEB2HTML):
         '''
 
         # We can only processes tags. If there isn't a tag return any text.
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']

src/calibre/ebooks/txt/txtml.py

@@ -11,7 +11,7 @@ Transform OEB content into plain text
 import re
 
 from lxml import etree
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 BLOCK_TAGS = [
@@ -192,10 +192,10 @@ class TXTMLizer(object):
         '''
         from calibre.ebooks.oeb.base import XHTML_NS, barename, namespace
 
-        if not isinstance(elem.tag, basestring) \
+        if not isinstance(elem.tag, string_or_bytes) \
            or namespace(elem.tag) != XHTML_NS:
             p = elem.getparent()
-            if p is not None and isinstance(p.tag, basestring) and namespace(p.tag) == XHTML_NS \
+            if p is not None and isinstance(p.tag, string_or_bytes) and namespace(p.tag) == XHTML_NS \
                     and elem.tail:
                 return [elem.tail]
             return ['']

src/calibre/gui2/__init__.py

@@ -35,7 +35,7 @@ from calibre.utils.config import Config, ConfigProxy, JSONConfig, dynamic
 from calibre.utils.date import UNDEFINED_DATE
 from calibre.utils.file_type_icons import EXT_MAP
 from calibre.utils.localization import get_lang
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 try:
     NO_URL_FORMATTING = QUrl.None_
@@ -1125,7 +1125,7 @@ def open_url(qurl):
     # Qt 5 requires QApplication to be constructed before trying to use
     # QDesktopServices::openUrl()
     ensure_app()
-    if isinstance(qurl, basestring):
+    if isinstance(qurl, string_or_bytes):
         qurl = QUrl(qurl)
     with sanitize_env_vars():
         QDesktopServices.openUrl(qurl)

src/calibre/gui2/actions/__init__.py

@@ -15,7 +15,7 @@ from calibre import prints
 from calibre.constants import isosx
 from calibre.gui2 import Dispatcher
 from calibre.gui2.keyboard import NameConflict
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 def menu_action_unique_name(plugin, unique_name):
@@ -177,7 +177,7 @@ class InterfaceAction(QObject):
         if attr == 'qaction':
             shortcut_action = ma
         if shortcut is not None:
-            keys = ((shortcut,) if isinstance(shortcut, basestring) else
+            keys = ((shortcut,) if isinstance(shortcut, string_or_bytes) else
                     tuple(shortcut))
             if shortcut_name is None and spec[0]:
                 shortcut_name = unicode_type(spec[0])
@@ -249,7 +249,7 @@ class InterfaceAction(QObject):
             ac.setIcon(icon)
         keys = ()
         if shortcut is not None and shortcut is not False:
-            keys = ((shortcut,) if isinstance(shortcut, basestring) else
+            keys = ((shortcut,) if isinstance(shortcut, string_or_bytes) else
                     tuple(shortcut))
         unique_name = menu_action_unique_name(self, unique_name)
         if description is not None:

src/calibre/gui2/actions/add.py

@@ -25,6 +25,7 @@ from calibre.gui2.actions import InterfaceAction
 from calibre.gui2 import question_dialog
 from calibre.ebooks.metadata import MetaInformation
 from calibre.ptempfile import PersistentTemporaryFile
+from polyglot.builtins import string_or_bytes
 
 
 def get_filters():
@@ -436,7 +437,7 @@ class AddAction(InterfaceAction):
             self.gui.refresh_cover_browser()
 
     def __add_filesystem_book(self, paths, allow_device=True):
-        if isinstance(paths, basestring):
+        if isinstance(paths, string_or_bytes):
             paths = [paths]
         books = [path for path in map(os.path.abspath, paths) if os.access(path,
             os.R_OK)]
@@ -591,7 +592,7 @@ class AddAction(InterfaceAction):
             self.gui.device_job_exception(job)
             return
         paths = job.result
-        ok_paths = [x for x in paths if isinstance(x, basestring)]
+        ok_paths = [x for x in paths if isinstance(x, string_or_bytes)]
         failed_paths = [x for x in paths if isinstance(x, tuple)]
         if failed_paths:
             if not ok_paths:

src/calibre/gui2/actions/similar_books.py

@@ -9,6 +9,7 @@ __docformat__ = 'restructuredtext en'
 from PyQt5.Qt import QToolButton
 
 from calibre.gui2.actions import InterfaceAction
+from polyglot.builtins import string_or_bytes
 
 
 class SimilarBooksAction(InterfaceAction):
@@ -74,7 +75,7 @@ class SimilarBooksAction(InterfaceAction):
         if not val:
             return
 
-        if isinstance(val, basestring):
+        if isinstance(val, string_or_bytes):
             val = [val]
         search = [col + ':"='+t.replace('"', '\\"')+'"' for t in val]
         if search:

src/calibre/gui2/add.py

@@ -11,7 +11,7 @@ from threading import Thread
 from collections import OrderedDict
 from Queue import Empty
 from io import BytesIO
-from polyglot.builtins import map, unicode_type
+from polyglot.builtins import map, unicode_type, string_or_bytes
 
 from PyQt5.Qt import QObject, Qt, pyqtSignal
 
@@ -33,7 +33,7 @@ from calibre.utils.ipc.pool import Pool, Failure
 
 
 def validate_source(source, parent=None):  # {{{
-    if isinstance(source, basestring):
+    if isinstance(source, string_or_bytes):
         if not os.path.exists(source):
             error_dialog(parent, _('Cannot add books'), _(
                 'The path %s does not exist') % source, show=True)
@@ -173,7 +173,7 @@ class Adder(QObject):
             return tdir
 
         try:
-            if isinstance(self.source, basestring):
+            if isinstance(self.source, string_or_bytes):
                 find_files(self.source)
                 self.ignore_opf = True
             else:

src/calibre/gui2/device.py

@@ -34,7 +34,7 @@ from calibre.utils.config import tweaks, device_prefs
 from calibre.utils.img import scale_image
 from calibre.library.save_to_disk import find_plugboard
 from calibre.ptempfile import PersistentTemporaryFile, force_unicode as filename_to_unicode
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 # }}}
 
 
@@ -471,7 +471,7 @@ class DeviceManager(Thread):  # {{{
         info = self.device.get_device_information(end_session=False)
         if len(info) < 5:
             info = tuple(list(info) + [{}])
-        info = [i.replace('\x00', '').replace('\x01', '') if isinstance(i, basestring) else i
+        info = [i.replace('\x00', '').replace('\x01', '') if isinstance(i, string_or_bytes) else i
                  for i in info]
         cp = self.device.card_prefix(end_session=False)
         fs = self.device.free_space()

src/calibre/gui2/ebook_download.py

@@ -19,6 +19,7 @@ from calibre.gui2.threaded_jobs import ThreadedJob
 from calibre.ptempfile import PersistentTemporaryDirectory
 from calibre.utils.filenames import ascii_filename
 from calibre.web import get_download_filename_from_response
+from polyglot.builtins import string_or_bytes
 
 
 class DownloadInfo(MessageBox):
@@ -141,7 +142,7 @@ class EbookDownloadMixin(object):
 
     def download_ebook(self, url='', cookie_file=None, filename='', save_loc='', add_to_lib=True, tags=[], create_browser=None):
         if tags:
-            if isinstance(tags, basestring):
+            if isinstance(tags, string_or_bytes):
                 tags = tags.split(',')
         start_ebook_download(Dispatcher(self.downloaded_ebook), self.job_manager, self, cookie_file, url, filename, save_loc, add_to_lib, tags, create_browser)
         self.status_bar.show_message(_('Downloading') + ' ' + filename.decode('utf-8', 'ignore') if filename else url.decode('utf-8', 'ignore'), 3000)

src/calibre/gui2/library/models.py

@@ -30,7 +30,7 @@ from calibre.constants import filesystem_encoding, DEBUG, config_dir
 from calibre.gui2.library import DEFAULT_SORT
 from calibre.utils.localization import calibre_langcode_to_name
 from calibre.library.coloring import color_row_key
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 Counts = namedtuple('Counts', 'library_total total current')
 
@@ -1282,7 +1282,7 @@ class OnDeviceSearch(SearchQueryParser):  # {{{
                     vals = accessor(row)
                     if vals is None:
                         vals = ''
-                    if isinstance(vals, basestring):
+                    if isinstance(vals, string_or_bytes):
                         vals = vals.split(',') if locvalue == 'collections' else [vals]
                     if _match(query, vals, m, use_primary_find_in_search=upf):
                         matches.add(index)

src/calibre/gui2/linux_file_dialogs.py

@@ -11,7 +11,7 @@ import sys
 import time
 from threading import Thread
 
-from polyglot.builtins import reraise, unicode_type
+from polyglot.builtins import reraise, unicode_type, string_or_bytes
 from PyQt5.Qt import QEventLoop
 
 from calibre import force_unicode
@@ -68,7 +68,7 @@ def get_initial_dir(name, title, default_dir, no_save_dir):
         return ensure_dir(process_path(default_dir))
     key = dialog_name(name, title)
     saved = dynamic.get(key)
-    if not isinstance(saved, basestring):
+    if not isinstance(saved, string_or_bytes):
         saved = None
     if saved and os.path.isdir(saved):
         return ensure_dir(process_path(saved))

src/calibre/gui2/lrf_renderer/text.py

@@ -10,7 +10,7 @@ from PyQt5.Qt import (
 from calibre.ebooks.lrf.fonts import LIBERATION_FONT_MAP
 from calibre.ebooks.BeautifulSoup import Tag
 from calibre.ebooks.hyphenate import hyphenate_word
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 WEIGHT_MAP = lambda wt : int((wt/10.)-1)
 NULL       = lambda a, b: a
@@ -223,7 +223,7 @@ class TextBlock(object):
         open_containers = collections.deque()
         self.in_para = False
         for i in tb.content:
-            if isinstance(i, basestring):
+            if isinstance(i, string_or_bytes):
                 self.process_text(i)
             elif i is None:
                 if len(open_containers) > 0:

src/calibre/gui2/open_with.py

@@ -22,6 +22,7 @@ from calibre.gui2.widgets2 import Dialog
 from calibre.gui2.progress_indicator import ProgressIndicator
 from calibre.utils.config import JSONConfig
 from calibre.utils.icu import numeric_sort_key as sort_key
+from polyglot.builtins import string_or_bytes
 
 ENTRY_ROLE = Qt.UserRole
 
@@ -197,7 +198,7 @@ else:
 
     def entry_to_item(entry, parent):
         icon_path = entry.get('Icon') or I('blank.png')
-        if not isinstance(icon_path, basestring):
+        if not isinstance(icon_path, string_or_bytes):
             icon_path = I('blank.png')
         ans = QListWidgetItem(QIcon(icon_path), entry.get('Name') or _('Unknown'), parent)
         ans.setData(ENTRY_ROLE, entry)

src/calibre/gui2/preferences/__init__.py

@@ -14,7 +14,7 @@ from PyQt5.Qt import (QWidget, pyqtSignal, QCheckBox, QAbstractSpinBox,
 from calibre.customize.ui import preferences_plugins
 from calibre.utils.config import ConfigProxy
 from calibre.gui2.complete2 import EditWithComplete
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class AbortCommit(Exception):
@@ -144,7 +144,7 @@ class Setting(object):
             else:
                 self.gui_obj.clear()
                 for x in choices:
-                    if isinstance(x, basestring):
+                    if isinstance(x, string_or_bytes):
                         x = (x, x)
                     self.gui_obj.addItem(x[0], (x[1]))
         self.set_gui_val(self.get_config_val(default=False))

src/calibre/gui2/qt_file_dialogs.py

@@ -11,7 +11,7 @@ from PyQt5.Qt import QFileDialog, QObject
 
 from calibre.gui2.linux_file_dialogs import dialog_name, image_extensions
 from calibre.utils.filenames import expanduser
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 def select_initial_dir(q):
@@ -71,7 +71,7 @@ class FileDialog(QObject):
         else:
             initial_dir = dynamic.get(self.dialog_name,
                     expanduser(default_dir))
-        if not isinstance(initial_dir, basestring):
+        if not isinstance(initial_dir, string_or_bytes):
             initial_dir = expanduser(default_dir)
         if not initial_dir or (not os.path.exists(initial_dir) and not (
                 mode == QFileDialog.AnyFile and (no_save_dir or combine_file_and_saved_dir))):

src/calibre/gui2/search_box.py

@@ -20,7 +20,7 @@ from calibre.gui2.dialogs.confirm_delete import confirm
 from calibre.gui2.dialogs.saved_search_editor import SavedSearchEditor
 from calibre.gui2.dialogs.search import SearchDialog
 from calibre.utils.icu import primary_sort_key
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 QT_HIDDEN_CLEAR_ACTION = '_q_qlineeditclearaction'
 
@@ -183,7 +183,7 @@ class SearchBox2(QComboBox):  # {{{
         self.setFocus(Qt.OtherFocusReason)
 
     def search_done(self, ok):
-        if isinstance(ok, basestring):
+        if isinstance(ok, string_or_bytes):
             self.setToolTip(ok)
             ok = False
         if not unicode_type(self.currentText()).strip():

src/calibre/gui2/tweak_book/boss.py

@@ -70,7 +70,7 @@ from calibre.utils.config import JSONConfig
 from calibre.utils.icu import numeric_sort_key
 from calibre.utils.imghdr import identify
 from calibre.utils.tdir_in_cache import tdir_in_cache
-from polyglot.builtins import iteritems
+from polyglot.builtins import iteritems, string_or_bytes
 
 _diff_dialogs = []
 last_used_transform_rules = []
@@ -1337,7 +1337,7 @@ class Boss(QObject):
 
     @in_thread_job
     def export_requested(self, name_or_names, path):
-        if isinstance(name_or_names, basestring):
+        if isinstance(name_or_names, string_or_bytes):
             return self.export_file(name_or_names, path)
         for name in name_or_names:
             dest = os.path.abspath(os.path.join(path, name))

src/calibre/gui2/tweak_book/editor/widget.py

@@ -24,11 +24,11 @@ from calibre.gui2.tweak_book.editor import SPELL_PROPERTY, LINK_PROPERTY, TAG_NA
 from calibre.gui2.tweak_book.editor.help import help_url
 from calibre.gui2.tweak_book.editor.text import TextEdit
 from calibre.utils.icu import utf16_length
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 def create_icon(text, palette=None, sz=None, divider=2, fill='white'):
-    if isinstance(fill, basestring):
+    if isinstance(fill, string_or_bytes):
         fill = QColor(fill)
     sz = sz or int(math.ceil(tprefs['toolbar_icon_size'] * QApplication.instance().devicePixelRatio()))
     if palette is None:

src/calibre/gui2/ui.py

@@ -53,7 +53,7 @@ from calibre.gui2.dbus_export.widgets import factory
 from calibre.gui2.open_with import register_keyboard_shortcuts
 from calibre.library import current_library_name
 from calibre.srv.library_broker import GuiLibraryBroker
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Listener(Thread):  # {{{
@@ -615,7 +615,7 @@ class Main(MainWindow, MainWindowMixin, DeviceMixin, EmailMixin,  # {{{
         self.tags_view.recount()
 
     def handle_cli_args(self, args):
-        if isinstance(args, basestring):
+        if isinstance(args, string_or_bytes):
             args = [args]
         files = [os.path.abspath(p) for p in args if not os.path.isdir(p) and os.access(p, os.R_OK)]
         if files:

src/calibre/gui2/viewer/documentview.py

@@ -7,7 +7,7 @@ __docformat__ = 'restructuredtext en'
 import math, json
 from base64 import b64encode
 from functools import partial
-from polyglot.builtins import map, unicode_type
+from polyglot.builtins import map, unicode_type, string_or_bytes
 
 from PyQt5.Qt import (
     QSize, QSizePolicy, QUrl, Qt, QPainter, QPalette, QBrush,
@@ -1169,12 +1169,12 @@ class DocumentView(QWebView):  # {{{
         old_pos = (self.document.xpos if self.document.in_paged_mode else
                 self.document.ypos)
         if self.document.in_paged_mode:
-            if isinstance(pos, basestring):
+            if isinstance(pos, string_or_bytes):
                 self.document.jump_to_anchor(pos)
             else:
                 self.document.scroll_fraction = pos
         else:
-            if isinstance(pos, basestring):
+            if isinstance(pos, string_or_bytes):
                 self.document.jump_to_anchor(pos)
             else:
                 if pos >= 1:

src/calibre/gui2/win_file_dialogs.py

@@ -9,6 +9,7 @@ from threading import Thread
 from uuid import uuid4
 
 from PyQt5.Qt import pyqtSignal, QEventLoop, Qt
+from polyglot.builtins import string_or_bytes
 
 is64bit = sys.maxsize > (1 << 32)
 base = sys.extensions_location if hasattr(sys, 'new_app_layout') else os.path.dirname(sys.executable)
@@ -76,7 +77,7 @@ def serialize_file_types(file_types):
         buf.append(struct.pack(b'=H%ds' % len(x), len(x), x))
     for name, extensions in file_types:
         add(name or _('Files'))
-        if isinstance(extensions, basestring):
+        if isinstance(extensions, string_or_bytes):
             extensions = extensions.split()
         add('; '.join('*.' + ext.lower() for ext in extensions))
     return b''.join(buf)

src/calibre/library/caches.py

@@ -20,7 +20,7 @@ from calibre.db.search import CONTAINS_MATCH, EQUALS_MATCH, REGEXP_MATCH, _match
 from calibre.ebooks.metadata import title_sort, author_to_author_sort
 from calibre.ebooks.metadata.opf2 import metadata_to_opf
 from calibre import prints, force_unicode
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class MetadataBackup(Thread):  # {{{
@@ -793,7 +793,7 @@ class ResultCache(SearchQueryParser):  # {{{
                         continue
 
                     if q == 'true' and matchkind == CONTAINS_MATCH:
-                        if isinstance(item[loc], basestring):
+                        if isinstance(item[loc], string_or_bytes):
                             if item[loc].strip() == '':
                                 continue
                         matches.add(item[0])

src/calibre/library/catalogs/bibtex.py

@@ -15,6 +15,7 @@ from calibre.library.catalogs import FIELDS, TEMPLATE_ALLOWED_FIELDS
 from calibre.customize.conversion import DummyReporter
 from calibre.constants import preferred_encoding
 from calibre.ebooks.metadata import format_isbn
+from polyglot.builtins import string_or_bytes
 
 
 class BIBTEX(CatalogPlugin):
@@ -212,13 +213,13 @@ class BIBTEX(CatalogPlugin):
                     bibtex_entry.append(u'year = "%s"' % item.year)
                     bibtex_entry.append(u'month = "%s"' % bibtexdict.utf8ToBibtex(strftime("%b", item)))
 
-                elif field.startswith('#') and isinstance(item, basestring):
+                elif field.startswith('#') and isinstance(item, string_or_bytes):
                     bibtex_entry.append(u'custom_%s = "%s"' % (field[1:],
                         bibtexdict.utf8ToBibtex(item)))
 
-                elif isinstance(item, basestring):
+                elif isinstance(item, string_or_bytes):
                     # elif field in ['title', 'publisher', 'cover', 'uuid', 'ondevice',
-                        # 'author_sort', 'series', 'title_sort'] :
+                    # 'author_sort', 'series', 'title_sort'] :
                     bibtex_entry.append(u'%s = "%s"' % (field, bibtexdict.utf8ToBibtex(item)))
 
             bibtex_entry = u',\n    '.join(bibtex_entry)

src/calibre/library/database2.py

@@ -48,7 +48,7 @@ from calibre.db.lazy import FormatMetadata, FormatsList
 from calibre.db.categories import Tag, CATEGORY_SORTS
 from calibre.utils.localization import (canonicalize_lang,
         calibre_langcode_to_name)
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 copyfile = os.link if hasattr(os, 'link') else shutil.copyfile
 SPOOL_SIZE = 30*1024*1024
@@ -1115,7 +1115,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
 
     def find_identical_books(self, mi):
         fuzzy_title_patterns = [(re.compile(pat, re.IGNORECASE) if
-            isinstance(pat, basestring) else pat, repl) for pat, repl in
+            isinstance(pat, string_or_bytes) else pat, repl) for pat, repl in
                 [
                     (r'[\[\](){}<>\'";,:#]', ''),
                     (get_title_sort_pat(), ''),
@@ -1398,7 +1398,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
             id_ = index if index_is_id else self.id(index)
             raise NoSuchFormat('Record %d has no %s file'%(id_, fmt))
         if windows_atomic_move is not None:
-            if not isinstance(dest, basestring):
+            if not isinstance(dest, string_or_bytes):
                 raise Exception("Error, you must pass the dest as a path when"
                         " using windows_atomic_move")
             if dest:
@@ -1454,7 +1454,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         id = index if index_is_id else self.id(index)
         path = os.path.join(self.library_path, self.path(id, index_is_id=True), 'cover.jpg')
         if windows_atomic_move is not None:
-            if not isinstance(dest, basestring):
+            if not isinstance(dest, string_or_bytes):
                 raise Exception("Error, you must pass the dest as a path when"
                         " using windows_atomic_move")
             if os.access(path, os.R_OK) and dest and not samefile(dest, path):
@@ -2334,7 +2334,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         # force_changes has no effect on cover manipulation
         if mi.cover_data[1] is not None:
             doit(self.set_cover, id, mi.cover_data[1], commit=False)
-        elif isinstance(mi.cover, basestring) and mi.cover:
+        elif isinstance(mi.cover, string_or_bytes) and mi.cover:
             if os.access(mi.cover, os.R_OK):
                 with lopen(mi.cover, 'rb') as f:
                     raw = f.read()
@@ -2641,7 +2641,7 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
     def set_pubdate(self, id, dt, notify=True, commit=True):
         if not dt:
             dt = UNDEFINED_DATE
-        if isinstance(dt, basestring):
+        if isinstance(dt, string_or_bytes):
             dt = parse_only_date(dt)
         self.conn.execute('UPDATE books SET pubdate=? WHERE id=?', (dt, id))
         self.data.set(id, self.FIELD_MAP['pubdate'], dt, row_is_id=True)

src/calibre/srv/ajax.py

@@ -7,7 +7,7 @@ __license__ = 'GPL v3'
 __copyright__ = '2015, Kovid Goyal <kovid at kovidgoyal.net>'
 
 from functools import partial
-from polyglot.builtins import unicode_type, zip
+from polyglot.builtins import unicode_type, zip, string_or_bytes
 from itertools import cycle
 
 from calibre import force_unicode
@@ -107,7 +107,7 @@ def book_to_json(ctx, rd, db, book_id,
                 if (fm and fm['is_category'] and not fm['is_csp'] and
                         key != 'formats' and fm['datatype'] != 'rating'):
                     categories = mi.get(key) or []
-                    if isinstance(categories, basestring):
+                    if isinstance(categories, string_or_bytes):
                         categories = [categories]
                     category_urls[key] = dbtags = {}
                     for category in categories:

src/calibre/srv/legacy.py

@@ -2,29 +2,31 @@
 # vim:fileencoding=utf-8
 # License: GPLv3 Copyright: 2016, Kovid Goyal <kovid at kovidgoyal.net>
 
-from __future__ import (unicode_literals, division, absolute_import,
+from __future__ import absolute_import, division, print_function, unicode_literals
-                        print_function)
+
 from functools import partial
+from urllib import urlencode
+
 from lxml.html import tostring
 from lxml.html.builder import E as E_
-from urllib import urlencode
 
 from calibre import strftime
 from calibre.constants import __appname__
 from calibre.db.view import sanitize_sort_field_name
 from calibre.ebooks.metadata import authors_to_string
-from calibre.srv.content import get, book_filename
+from calibre.srv.content import book_filename, get
-from calibre.srv.errors import HTTPRedirect, HTTPBadRequest
+from calibre.srv.errors import HTTPBadRequest, HTTPRedirect
 from calibre.srv.routes import endpoint
 from calibre.srv.utils import get_library_data, http_date
 from calibre.utils.cleantext import clean_xml_chars
-from calibre.utils.date import timestampfromdt, dt_as_local, is_date_undefined
+from calibre.utils.date import dt_as_local, is_date_undefined, timestampfromdt
+from polyglot.builtins import string_or_bytes
 
 # /mobile {{{
 
 
 def clean(x):
-    if isinstance(x, basestring):
+    if isinstance(x, string_or_bytes):
         x = clean_xml_chars(x)
     return x
 

src/calibre/utils/config_base.py

@@ -234,7 +234,7 @@ class OptionSet(object):
 
     def serialize_opt(self, val):
         if val is val is True or val is False or val is None or \
-           isinstance(val, (int, float, long, basestring)):
+           isinstance(val, (int, float, long, bytes, unicode_type)):
             return repr(val)
         pickle = cPickle.dumps(val, -1)
         return 'cPickle.loads(%s)'%repr(pickle)

src/calibre/utils/img.py

@@ -23,6 +23,7 @@ from calibre.ptempfile import TemporaryDirectory
 from calibre.utils.config_base import tweaks
 from calibre.utils.filenames import atomic_rename
 from calibre.utils.imghdr import what
+from polyglot.builtins import string_or_bytes
 
 # Utilities {{{
 imageops, imageops_err = plugins['imageops']
@@ -436,7 +437,7 @@ def quantize_image(img, max_colors=256, dither=True, palette=''):
     img = image_from_data(img)
     if img.hasAlphaChannel():
         img = blend_image(img)
-    if palette and isinstance(palette, basestring):
+    if palette and isinstance(palette, string_or_bytes):
         palette = palette.split()
     return imageops.quantize(img, max_colors, dither, [QColor(x).rgb() for x in palette])
 

src/calibre/utils/imghdr.py

@@ -7,6 +7,7 @@ from __future__ import (unicode_literals, division, absolute_import,
 from struct import unpack, error
 import os
 from calibre.utils.speedups import ReadOnlyFileBuffer
+from polyglot.builtins import string_or_bytes
 
 """ Recognize image file formats and sizes based on their first few bytes."""
 
@@ -16,7 +17,7 @@ HSIZE = 120
 def what(file, h=None):
     ' Recognize image headers '
     if h is None:
-        if isinstance(file, basestring):
+        if isinstance(file, string_or_bytes):
             with lopen(file, 'rb') as f:
                 h = f.read(HSIZE)
         else:

src/calibre/utils/ipc/launch.py

@@ -12,7 +12,7 @@ from functools import partial
 from calibre.constants import iswindows, isosx, isfrozen, filesystem_encoding
 from calibre.utils.config import prefs
 from calibre.ptempfile import PersistentTemporaryFile, base_dir
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 if iswindows:
     import win32process
@@ -186,7 +186,7 @@ class Worker(object):
         _cwd = cwd
         if priority is None:
             priority = prefs['worker_process_priority']
-        cmd = [exe] if isinstance(exe, basestring) else exe
+        cmd = [exe] if isinstance(exe, string_or_bytes) else exe
         args = {
                 'env' : env,
                 'cwd' : _cwd,

src/calibre/utils/ipc/server.py

@@ -21,6 +21,7 @@ from calibre.utils.ipc.worker import PARALLEL_FUNCS
 from calibre import detect_ncpus as cpu_count
 from calibre.constants import iswindows, DEBUG, islinux
 from calibre.ptempfile import base_dir
+from polyglot.builtins import string_or_bytes
 
 _counter = 0
 
@@ -216,7 +217,7 @@ class Server(Thread):
                 'CALIBRE_WORKER_RESULT' : hexlify(rfile.encode('utf-8')),
               }
         cw = self.do_launch(env, gui, redirect_output, rfile, job_name=job_name)
-        if isinstance(cw, basestring):
+        if isinstance(cw, string_or_bytes):
             raise CriticalError('Failed to launch worker process:\n'+cw)
         if DEBUG:
             print('Worker Launch took:', time.time() - start)

src/calibre/utils/ipc/simple_worker.py

@@ -16,7 +16,7 @@ from contextlib import closing
 from calibre.constants import iswindows
 from calibre.utils.ipc import eintr_retry_call
 from calibre.utils.ipc.launch import Worker
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class WorkerError(Exception):
@@ -164,7 +164,7 @@ def start_pipe_worker(command, env=None, priority='normal', **process_args):
         args['close_fds'] = True
 
     exe = w.executable
-    cmd = [exe] if isinstance(exe, basestring) else exe
+    cmd = [exe] if isinstance(exe, string_or_bytes) else exe
     p = subprocess.Popen(cmd + ['--pipe-worker', command], **args)
     return p
 

src/calibre/utils/open_with/linux.py

@@ -13,6 +13,7 @@ from calibre import walk, guess_type, prints, force_unicode
 from calibre.constants import filesystem_encoding, cache_dir
 from calibre.utils.icu import numeric_sort_key as sort_key
 from calibre.utils.localization import canonicalize_lang, get_lang
+from polyglot.builtins import string_or_bytes
 
 
 def parse_localized_key(key):
@@ -73,6 +74,7 @@ def parse_desktop_file(path):
     if 'Exec' in ans and 'MimeType' in ans and 'Name' in ans:
         return ans
 
+
 icon_data = None
 
 
@@ -197,7 +199,7 @@ def find_programs(extensions):
                     data['Icon'] = icon
                 else:
                     data.pop('Icon')
-            if not isinstance(data.get('Icon'), basestring):
+            if not isinstance(data.get('Icon'), string_or_bytes):
                 data.pop('Icon', None)
             for k in ('Name', 'GenericName', 'Comment'):
                 val = data.get(k)

src/calibre/utils/open_with/osx.py

@@ -11,7 +11,7 @@ from collections import defaultdict
 
 from calibre.ptempfile import TemporaryDirectory
 from calibre.utils.icu import numeric_sort_key
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 application_locations = ('/Applications', '~/Applications', '~/Desktop')
 
@@ -240,7 +240,7 @@ def get_extensions_from_utis(utis, plist):
         for decl in plist.get(key, ()):
             if isinstance(decl, dict):
                 uti = decl.get('UTTypeIdentifier')
-                if isinstance(uti, basestring):
+                if isinstance(uti, string_or_bytes):
                     spec = decl.get('UTTypeTagSpecification')
                     if isinstance(spec, dict):
                         ext = spec.get('public.filename-extension')
@@ -287,10 +287,10 @@ def get_bundle_data(path):
             extensions |= get_extensions_from_utis(utis, plist)
         else:
             for ext in dtype.get('CFBundleTypeExtensions', ()):
-                if isinstance(ext, basestring):
+                if isinstance(ext, string_or_bytes):
                     extensions.add(ext.lower())
             for mt in dtype.get('CFBundleTypeMIMETypes', ()):
-                if isinstance(mt, basestring):
+                if isinstance(mt, string_or_bytes):
                     for ext in mimetypes.guess_all_extensions(mt, strict=False):
                         extensions.add(ext.lower())
     return ans

src/calibre/utils/rss_gen.py

@@ -7,6 +7,7 @@ __author__ = "Andrew Dalke <dalke@dalkescientific.com>"
 _generator_name = __name__ + "-" + ".".join(map(str, __version__))
 
 import datetime
+from polyglot.builtins import string_or_bytes
 
 # Could make this the base class; will need to add 'publish'
 
@@ -32,7 +33,7 @@ class WriteXmlMixin:
 
 
 def _element(handler, name, obj, d={}):
-    if isinstance(obj, basestring) or obj is None:
+    if isinstance(obj, string_or_bytes) or obj is None:
         # special-case handling to make the API easier
         # to use for the common case.
         handler.startElement(name, d)
@@ -373,7 +374,7 @@ class RSS2(WriteXmlMixin):
         _opt_element(handler, "lastBuildDate", lastBuildDate)
 
         for category in self.categories:
-            if isinstance(category, basestring):
+            if isinstance(category, string_or_bytes):
                 category = Category(category)
             category.publish(handler)
 
@@ -454,7 +455,7 @@ class RSSItem(WriteXmlMixin):
         _opt_element(handler, "author", self.author)
 
         for category in self.categories:
-            if isinstance(category, basestring):
+            if isinstance(category, string_or_bytes):
                 category = Category(category)
             category.publish(handler)
 

src/calibre/utils/smtplib.py

@@ -52,7 +52,7 @@ from email.base64mime import encode as encode_base64
 from sys import stderr
 from functools import partial
 
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 __all__ = ["SMTPException", "SMTPServerDisconnected", "SMTPResponseException",
            "SMTPSenderRefused", "SMTPRecipientsRefused", "SMTPDataError",
@@ -766,7 +766,7 @@ class SMTP:
             self.rset()
             raise SMTPSenderRefused(code, resp, from_addr)
         senderrs = {}
-        if isinstance(to_addrs, basestring):
+        if isinstance(to_addrs, string_or_bytes):
             to_addrs = [to_addrs]
         for each in to_addrs:
             (code, resp) = self.rcpt(each, rcpt_options)

src/calibre/utils/unrar.py

@@ -13,6 +13,7 @@ from io import BytesIO
 
 from calibre.constants import filesystem_encoding, iswindows
 from calibre.ptempfile import PersistentTemporaryFile, TemporaryDirectory
+from polyglot.builtins import string_or_bytes
 
 
 def as_unicode(x):
@@ -28,7 +29,7 @@ class StreamAsPath(object):
 
     def __enter__(self):
         self.temppath = None
-        if isinstance(self.stream, basestring):
+        if isinstance(self.stream, string_or_bytes):
             return as_unicode(self.stream)
         name = getattr(self.stream, 'name', None)
         if name and os.access(name, os.R_OK):

src/calibre/utils/zipfile.py

@@ -11,7 +11,7 @@ from tempfile import SpooledTemporaryFile
 from calibre import sanitize_file_name2
 from calibre.constants import filesystem_encoding
 from calibre.ebooks.chardet import detect
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 try:
     import zlib  # We may need its compression method
@@ -747,7 +747,7 @@ class ZipFile:
         self.comment = ''
 
         # Check if we were passed a file-like object
-        if isinstance(file, basestring):
+        if isinstance(file, string_or_bytes):
             self._filePassed = 0
             self.filename = file
             modeDict = {'r' : 'rb', 'w': 'wb', 'a' : 'r+b'}

src/calibre/web/feeds/__init__.py

@@ -12,7 +12,7 @@ from calibre.utils.logging import default_log
 from calibre import entity_to_unicode, strftime, force_unicode
 from calibre.utils.date import dt_factory, utcnow, local_tz
 from calibre.utils.cleantext import clean_ascii_chars, clean_xml_chars
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class Article(object):
@@ -21,7 +21,7 @@ class Article(object):
         from lxml import html
         self.downloaded = False
         self.id = id
-        if not title or not isinstance(title, basestring):
+        if not title or not isinstance(title, string_or_bytes):
             title = _('Unknown')
         title = force_unicode(title, 'utf-8')
         self._title = clean_xml_chars(title).strip()

src/calibre/web/feeds/news.py

@@ -31,7 +31,7 @@ from calibre.utils.icu import numeric_sort_key
 from calibre.utils.img import save_cover_data_to, add_borders_to_image, image_to_data
 from calibre.utils.localization import canonicalize_lang
 from calibre.utils.logging import ThreadSafeWrapper
-from polyglot.builtins import unicode_type
+from polyglot.builtins import unicode_type, string_or_bytes
 
 
 class LoginFailed(ValueError):
@@ -1612,7 +1612,7 @@ class BasicNewsRecipe(Recipe):
         feeds = self.get_feeds()
         parsed_feeds = []
         for obj in feeds:
-            if isinstance(obj, basestring):
+            if isinstance(obj, string_or_bytes):
                 title, url = None, obj
             else:
                 title, url = obj
@@ -1659,7 +1659,7 @@ class BasicNewsRecipe(Recipe):
         '''
         if tag is None:
             return ''
-        if isinstance(tag, basestring):
+        if isinstance(tag, string_or_bytes):
             return tag
         if callable(getattr(tag, 'xpath', None)) and not hasattr(tag, 'contents'):  # a lxml tag
             from lxml.etree import tostring

src/odf/odf2xhtml.py

@@ -1580,7 +1580,7 @@ dl.notes dd:last-of-type { page-break-after: avoid }
         """
         self.lines = []
         self._wfunc = self._wlines
-        if isinstance(odffile, basestring) \
+        if isinstance(odffile, (bytes, type(u'')) \
                 or hasattr(odffile, 'read'):  # Added by Kovid
             self.document = load(odffile)
         else:

src/odf/userfield.py

@@ -62,7 +62,7 @@ class UserFields(object):
         self.document = None
 
     def loaddoc(self):
-        if isinstance(self.src_file, basestring):
+        if isinstance(self.src_file, (bytes, type(u'')):
             # src_file is a filename, check if it is a zip-file
             if not zipfile.is_zipfile(self.src_file):
                 raise TypeError("%s is no odt file." % self.src_file)
@@ -166,4 +166,3 @@ class UserFields(object):
                 else:
                     f.setAttribute('value', value)
         self.savedoc()
-

src/polyglot/builtins.py

@@ -29,6 +29,7 @@ if is_py3:
 
     codepoint_to_chr = chr
     unicode_type = str
+    string_or_bytes = str, bytes
 
     def iteritems(d):
         return iter(d.items())
@@ -53,6 +54,7 @@ else:
 
     codepoint_to_chr = unichr
     unicode_type = unicode
+    string_or_bytes = unicode, bytes
 
     def iteritems(d):
         return d.iteritems()