Add support for the FBZ format (zipped FB2)

Fixes #1762634 [Enhancement add support for .fb2.zip files](https://bugs.launchpad.net/calibre/+bug/1762634)

src/calibre/customize/builtins.py

@@ -197,7 +197,7 @@ class EPUBMetadataReader(MetadataReaderPlugin):
 class FB2MetadataReader(MetadataReaderPlugin):
 
     name        = 'Read FB2 metadata'
-    file_types  = set(['fb2'])
+    file_types  = {'fb2', 'fbz'}
     description = _('Read metadata from %s files')%'FB2'
 
     def get_metadata(self, stream, ftype):
@@ -476,7 +476,7 @@ class EPUBMetadataWriter(MetadataWriterPlugin):
 class FB2MetadataWriter(MetadataWriterPlugin):
 
     name = 'Set FB2 metadata'
-    file_types = set(['fb2'])
+    file_types = {'fb2', 'fbz'}
     description = _('Set metadata in %s files')%'FB2'
 
     def set_metadata(self, stream, mi, type):

src/calibre/ebooks/__init__.py

@@ -32,7 +32,7 @@ class ParserError(ValueError):
 
 BOOK_EXTENSIONS = ['lrf', 'rar', 'zip', 'rtf', 'lit', 'txt', 'txtz', 'text', 'htm', 'xhtm',
                    'html', 'htmlz', 'xhtml', 'pdf', 'pdb', 'updb', 'pdr', 'prc', 'mobi', 'azw', 'doc',
-                   'epub', 'fb2', 'djv', 'djvu', 'lrx', 'cbr', 'cbz', 'cbc', 'oebzip',
+                   'epub', 'fb2', 'fbz', 'djv', 'djvu', 'lrx', 'cbr', 'cbz', 'cbc', 'oebzip',
                    'rb', 'imp', 'odt', 'chm', 'tpz', 'azw1', 'pml', 'pmlz', 'mbp', 'tan', 'snb',
                    'xps', 'oxps', 'azw4', 'book', 'zbf', 'pobi', 'docx', 'docm', 'md',
                    'textile', 'markdown', 'ibook', 'ibooks', 'iba', 'azw3', 'ps', 'kepub', 'kfx']

src/calibre/ebooks/conversion/plugins/fb2_input.py

@@ -17,8 +17,8 @@ class FB2Input(InputFormatPlugin):
 
     name        = 'FB2 Input'
     author      = 'Anatoly Shipitsin'
-    description = 'Convert FB2 files to HTML'
+    description = 'Convert FB2 and FBZ files to HTML'
-    file_types  = set(['fb2'])
+    file_types  = {'fb2', 'fbz'}
 
     recommendations = set([
         ('level1_toc', '//h:h1', OptionRecommendation.MED),
@@ -37,14 +37,15 @@ class FB2Input(InputFormatPlugin):
     def convert(self, stream, options, file_ext, log,
                 accelerators):
         from lxml import etree
-        from calibre.ebooks.metadata.fb2 import ensure_namespace
+        from calibre.ebooks.metadata.fb2 import ensure_namespace, get_fb2_data
         from calibre.ebooks.metadata.opf2 import OPFCreator
         from calibre.ebooks.metadata.meta import get_metadata
         from calibre.ebooks.oeb.base import XLINK_NS, XHTML_NS, RECOVER_PARSER
         from calibre.ebooks.chardet import xml_to_unicode
         self.log = log
         log.debug('Parsing XML...')
-        raw = stream.read().replace('\0', '')
+        raw = get_fb2_data(stream)[0]
+        raw = raw.replace(b'\0', b'')
         raw = xml_to_unicode(raw, strip_encoding_pats=True,
             assume_utf8=True, resolve_entities=True)[0]
         try:
@@ -173,5 +174,3 @@ class FB2Input(InputFormatPlugin):
                 else:
                     with open(fname, 'wb') as f:
                         f.write(data)
-
-

src/calibre/ebooks/metadata/fb2.py

@@ -85,10 +85,27 @@ class Context(object):
                 self.create_tag(parent, 'empty-line', at_start=False)
 
 
+def get_fb2_data(stream):
+    from calibre.utils.zipfile import ZipFile, BadZipfile
+    pos = stream.tell()
+    try:
+        zf = ZipFile(stream)
+    except BadZipfile:
+        stream.seek(pos)
+        ans = stream.read()
+        zip_file_name = None
+    else:
+        names = zf.namelist()
+        names = [x for x in names if x.lower().endswith('.fb2')] or names
+        zip_file_name = names[0]
+        ans = zf.open(zip_file_name).read()
+    return ans, zip_file_name
+
+
 def get_metadata(stream):
     ''' Return fb2 metadata as a L{MetaInformation} object '''
 
-    root = _get_fbroot(stream)
+    root = _get_fbroot(get_fb2_data(stream)[0])
     ctx = Context(root)
     book_title = _parse_book_title(root, ctx)
     authors = _parse_authors(root, ctx) or [_('Unknown')]
@@ -294,9 +311,8 @@ def _parse_language(root, mi, ctx):
         mi.languages = [language]
 
 
-def _get_fbroot(stream):
+def _get_fbroot(raw):
     parser = etree.XMLParser(recover=True, no_network=True)
-    raw = stream.read()
     raw = xml_to_unicode(raw, strip_encoding_pats=True)[0]
     root = etree.fromstring(raw, parser=parser)
     return ensure_namespace(root)
@@ -386,7 +402,8 @@ def _set_cover(title_info, mi, ctx):
 
 def set_metadata(stream, mi, apply_null=False, update_timestamp=False):
     stream.seek(0)
-    root = _get_fbroot(stream)
+    raw, zip_file_name = get_fb2_data(stream)
+    root = _get_fbroot(raw)
     ctx = Context(root)
     desc = ctx.get_or_create(root, 'description')
     ti = ctx.get_or_create(desc, 'title-info')
@@ -403,14 +420,20 @@ def set_metadata(stream, mi, apply_null=False, update_timestamp=False):
     for child in ti:
         child.tail = indent
 
-    stream.seek(0)
-    stream.truncate()
     # Apparently there exists FB2 reading software that chokes on the use of
     # single quotes in xml declaration. Sigh. See
     # https://www.mobileread.com/forums/showthread.php?p=2273184#post2273184
-    stream.write(b'<?xml version="1.0" encoding="UTF-8"?>\n')
+    raw = b'<?xml version="1.0" encoding="UTF-8"?>\n'
-    stream.write(etree.tostring(root, method='xml', encoding='utf-8',
+    raw += etree.tostring(root, method='xml', encoding='utf-8', xml_declaration=False)
-        xml_declaration=False))
+
+    stream.seek(0)
+    stream.truncate()
+    if zip_file_name:
+        from calibre.utils.zipfile import ZipFile
+        with ZipFile(stream, 'w') as zf:
+            zf.writestr(zip_file_name, raw)
+    else:
+        stream.write(raw)
 
 
 def ensure_namespace(doc):