From 371c1bee5babcbc9237a252d9d3da3480e734d51 Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Thu, 11 Dec 2008 13:47:29 -0800
Subject: [PATCH] Add support for reading metadata from LRX files (thanks to
 igorsk)

---
 src/calibre/ebooks/metadata/lrx.py  | 89 +++++++++++++++++++++++++++++
 src/calibre/ebooks/metadata/meta.py |  5 +-
 2 files changed, 92 insertions(+), 2 deletions(-)
 create mode 100644 src/calibre/ebooks/metadata/lrx.py

diff --git a/src/calibre/ebooks/metadata/lrx.py b/src/calibre/ebooks/metadata/lrx.py
new file mode 100644
index 0000000000..2106a45450
--- /dev/null
+++ b/src/calibre/ebooks/metadata/lrx.py
@@ -0,0 +1,89 @@
+#!/usr/bin/env  python
+__license__   = 'GPL v3'
+__copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
+__docformat__ = 'restructuredtext en'
+
+'''
+Read metadata from LRX files
+'''
+
+import sys, struct
+from zlib import decompress
+from lxml import etree
+
+from calibre.ebooks.metadata import MetaInformation, string_to_authors
+
+def _read(f, at, amount):
+    f.seek(at)
+    return f.read(amount)
+
+def word_be(buf):
+    return struct.unpack('>L', buf)[0]
+
+def word_le(buf):
+    return struct.unpack('<L', buf)[0]
+
+def short_le(buf):
+    return struct.unpack('<H', buf)[0]
+
+def short_be(buf):
+    return struct.unpack('>H', buf)[0]
+
+
+def get_metadata(f):
+    read = lambda at, amount: _read(f, at, amount)
+    f.seek(0)
+    buf = f.read(12) 
+    if buf[4:] == 'ftypLRX2':
+        offset = 0
+        while True:
+            offset += word_be(buf[:4])
+            try:
+                buf = read(offset, 8)
+            except:
+                raise ValueError('Not a valid LRX file')
+            if buf[4:] == 'bbeb':
+                break
+        offset += 8
+        buf = read(offset, 16)
+        if buf[:8].decode('utf-16-le') != 'LRF\x00':
+            raise ValueError('Not a valid LRX file')
+        lrf_version = word_le(buf[8:12])
+        offset += 0x4c
+        compressed_size = short_le(read(offset, 2))
+        offset += 2
+        if lrf_version >= 800:
+            offset += 6
+        compressed_size -= 4
+        uncompressed_size = word_le(read(offset, 4))
+        info = decompress(f.read(compressed_size))
+        if len(info) != uncompressed_size:
+            raise ValueError('LRX file has malformed metadata section')
+        root = etree.fromstring(info)
+        bi = root.find('BookInfo')
+        title = bi.find('Title')
+        title_sort = title.get('reading', None)
+        title = title.text
+        author = bi.find('Author')
+        author_sort = author.get('reading', None)
+        mi = MetaInformation(title, string_to_authors(author.text))
+        mi.title_sort, mi.author_sort = title_sort, author_sort
+        author = author.text
+        publisher = bi.find('Publisher')
+        mi.publisher = getattr(publisher, 'text', None)
+        mi.tags = [x.text for x in  bi.findall('Category')]
+        mi.language = root.find('DocInfo').find('Language').text
+        return mi
+        
+    elif buf[4:8] == 'LRX':
+        raise ValueError('Librie LRX format not supported')
+    else:
+        raise ValueError('Not a LRX file')
+    
+
+def main(args=sys.argv):
+    print get_metadata(open(args[1], 'rb'))
+    return 0
+
+if __name__ == '__main__':
+    sys.exit(main())
\ No newline at end of file
diff --git a/src/calibre/ebooks/metadata/meta.py b/src/calibre/ebooks/metadata/meta.py
index aff09281da..3264655f9c 100644
--- a/src/calibre/ebooks/metadata/meta.py
+++ b/src/calibre/ebooks/metadata/meta.py
@@ -16,6 +16,7 @@ from calibre.ebooks.metadata.epub import get_metadata as epub_metadata
 from calibre.ebooks.metadata.html import get_metadata as html_metadata
 from calibre.ebooks.mobi.reader   import get_metadata as mobi_metadata
 from calibre.ebooks.metadata.odt  import get_metadata as odt_metadata
+from calibre.ebooks.metadata.lrx  import get_metadata as lrx_metadata 
 from calibre.ebooks.metadata.opf2 import OPF
 from calibre.ebooks.metadata.rtf  import set_metadata as set_rtf_metadata
 from calibre.ebooks.lrf.meta      import set_metadata as set_lrf_metadata
@@ -29,12 +30,12 @@ except OSError:
 from calibre.libunzip import extract_member as zip_extract_first
 
 from calibre.ebooks.metadata import MetaInformation
-from calibre.ptempfile import TemporaryDirectory
 
 _METADATA_PRIORITIES = [
                        'html', 'htm', 'xhtml', 'xhtm',
                        'rtf', 'fb2', 'pdf', 'prc', 'odt',
-                       'epub', 'lit', 'lrf', 'mobi', 'rb', 'imp'
+                       'epub', 'lit', 'lrx', 'lrf', 'mobi', 
+                       'rb', 'imp'
                       ]
 
 # The priorities for loading metadata from different file types