[library-web] hack support for .tar.xz in (code from ftpadmin)

[Frederic Peters <fpeters src gnome org>]

commit efce501b5fb248a17778beca08a33fa27aa61c23
Author: FrÃdÃric PÃters <fpeters 0d be>
Date:   Thu Nov 24 09:30:57 2011 +0100

    hack support for .tar.xz in (code from ftpadmin)

 src/lgo.py   |    4 --
 src/utils.py |   89 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 89 insertions(+), 4 deletions(-)
---
diff --git a/src/lgo.py b/src/lgo.py
index 929453d..c29be72 100755
--- a/src/lgo.py
+++ b/src/lgo.py
@@ -366,10 +366,6 @@ class Lgo(App):
                     href.startswith('http://ftp.gnome.org/')):
                 continue
             filename = self.download(href)
-            # FIXME: there's no support for .xz on the webapps vm used for
-            # library-web, fallback to the .bz2 archive.
-            if filename.endswith('.xz'):
-                filename = os.path.splitext(filename)[0] + '.bz2'
             if not filename:
                 continue
             logging.info('extracting module %s (from %s moduleset)' % (
diff --git a/src/utils.py b/src/utils.py
index c21f589..9955a4f 100644
--- a/src/utils.py
+++ b/src/utils.py
@@ -22,6 +22,9 @@ import re
 import time
 import urllib2
 import sys
+import tarfile
+import lzma # pyliblzma
+
 
 class FakeTarFile:
     '''
@@ -104,3 +107,89 @@ class LogFormatter(logging.Formatter):
             sys.stdout.flush()
         return '%c:[%s] %s' % (record.levelname[0], time.strftime('%Y%m%d %H:%M:%S'), record.msg)
 
+
+class _LZMAProxy(object):
+    """Small proxy class that enables external file object
+       support for "r:lzma" and "w:lzma" modes. This is actually
+       a workaround for a limitation in lzma module's LZMAFile
+       class which (unlike gzip.GzipFile) has no support for
+       a file object argument.
+    """
+
+    blocksize = 16 * 1024
+
+    def __init__(self, fileobj, mode):
+        self.fileobj = fileobj
+        self.mode = mode
+        self.name = getattr(self.fileobj, "name", None)
+        self.init()
+
+    def init(self):
+        self.pos = 0
+        if self.mode == "r":
+            self.lzmaobj = lzma.LZMADecompressor()
+            self.fileobj.seek(0)
+            self.buf = ""
+        else:
+            self.lzmaobj = lzma.LZMACompressor()
+
+    def read(self, size):
+        b = [self.buf]
+        x = len(self.buf)
+        while x < size:
+            raw = self.fileobj.read(self.blocksize)
+            if not raw:
+                break
+            try:
+                data = self.lzmaobj.decompress(raw)
+            except EOFError:
+                break
+            b.append(data)
+            x += len(data)
+        self.buf = "".join(b)
+
+        buf = self.buf[:size]
+        self.buf = self.buf[size:]
+        self.pos += len(buf)
+        return buf
+
+    def seek(self, pos):
+        if pos < self.pos:
+            self.init()
+        self.read(pos - self.pos)
+
+
+class XzTarFile(tarfile.TarFile):
+
+    OPEN_METH = tarfile.TarFile.OPEN_METH.copy()
+    OPEN_METH["xz"] = "xzopen"
+
+    @classmethod
+    def xzopen(cls, name, mode="r", fileobj=None, **kwargs):
+        """Open gzip compressed tar archive name for reading or writing.
+           Appending is not allowed.
+        """
+        if len(mode) > 1 or mode not in "rw":
+            raise ValueError("mode must be 'r' or 'w'")
+
+        if fileobj is not None:
+            fileobj = _LMZAProxy(fileobj, mode)
+        else:
+            fileobj = lzma.LZMAFile(name, mode)
+
+        try:
+            # lzma doesn't immediately return an error
+            # try and read a bit of data to determine if it is a valid xz file
+            fileobj.read(_LZMAProxy.blocksize)
+            fileobj.seek(0)
+            t = cls.taropen(name, mode, fileobj, **kwargs)
+        except IOError:
+            raise tarfile.ReadError("not a xz file")
+        except lzma.error:
+            raise tarfile.ReadError("not a xz file")
+        t._extfileobj = False
+        return t
+
+if not hasattr(tarfile.TarFile, 'xvopen'):
+    tarfile.open = XzTarFile.open
+