Merge heads

Doc/library/bz2.rst

@@ -37,14 +37,18 @@ All of the classes in this module may safely be accessed from multiple threads.
    *fileobj*), or operate directly on a named file (named by *filename*).
    Exactly one of these two parameters should be provided.
 
-   The *mode* argument can be either ``'r'`` for reading (default), or ``'w'``
-   for writing.
+   The *mode* argument can be either ``'r'`` for reading (default), ``'w'`` for
+   overwriting, or ``'a'`` for appending. If *fileobj* is provided, a mode of
+   ``'w'`` does not truncate the file, and is instead equivalent to ``'a'``.
 
    The *buffering* argument is ignored. Its use is deprecated.
 
-   If *mode* is ``'w'``, *compresslevel* can be a number between ``1`` and
-   ``9`` specifying the level of compression: ``1`` produces the least
-   compression, and ``9`` (default) produces the most compression.
+   If *mode* is ``'w'`` or ``'a'``, *compresslevel* can be a number between
+   ``1`` and ``9`` specifying the level of compression: ``1`` produces the
+   least compression, and ``9`` (default) produces the most compression.
+
+   If *mode* is ``'r'``, the input file may be the concatenation of multiple
+   compressed streams.
 
    :class:`BZ2File` provides all of the members specified by the
    :class:`io.BufferedIOBase`, except for :meth:`detach` and :meth:`truncate`.
@@ -70,6 +74,10 @@ All of the classes in this module may safely be accessed from multiple threads.
    .. versionchanged:: 3.3
       The *fileobj* argument to the constructor was added.
 
+   .. versionchanged:: 3.3
+      The ``'a'`` (append) mode was added, along with support for reading
+      multi-stream files.
+
 
 Incremental (de)compression
 ---------------------------
@@ -106,14 +114,20 @@ Incremental (de)compression
    incrementally. For one-shot compression, use the :func:`decompress` function
    instead.
 
+   .. note::
+      This class does not transparently handle inputs containing multiple
+      compressed streams, unlike :func:`decompress` and :class:`BZ2File`. If
+      you need to decompress a multi-stream input with :class:`BZ2Decompressor`,
+      you must use a new decompressor for each stream.
+
    .. method:: decompress(data)
 
       Provide data to the decompressor object. Returns a chunk of decompressed
       data if possible, or an empty byte string otherwise.
 
-      Attempting to decompress data after the end of stream is reached raises
-      an :exc:`EOFError`. If any data is found after the end of the stream, it
-      is ignored and saved in the :attr:`unused_data` attribute.
+      Attempting to decompress data after the end of the current stream is
+      reached raises an :exc:`EOFError`. If any data is found after the end of
+      the stream, it is ignored and saved in the :attr:`unused_data` attribute.
 
 
    .. attribute:: eof
@@ -127,6 +141,9 @@ Incremental (de)compression
 
       Data found after the end of the compressed stream.
 
+      If this attribute is accessed before the end of the stream has been
+      reached, its value will be ``b''``.
+
 
 One-shot (de)compression
 ------------------------
@@ -145,5 +162,11 @@ One-shot (de)compression
 
    Decompress *data*.
 
+   If *data* is the concatenation of multiple compressed streams, decompress
+   all of the streams.
+
    For incremental decompression, use a :class:`BZ2Decompressor` instead.
 
+   .. versionchanged:: 3.3
+      Support for multi-stream inputs was added.
+

Lib/bz2.py

@@ -76,6 +76,10 @@ class BZ2File(io.BufferedIOBase):
             mode = "wb"
             mode_code = _MODE_WRITE
             self._compressor = BZ2Compressor()
+        elif mode in ("a", "ab"):
+            mode = "ab"
+            mode_code = _MODE_WRITE
+            self._compressor = BZ2Compressor()
         else:
             raise ValueError("Invalid mode: {!r}".format(mode))
 
@@ -161,14 +165,25 @@ class BZ2File(io.BufferedIOBase):
     def _fill_buffer(self):
         if self._buffer:
             return True
-        if self._decompressor.eof:
-            self._mode = _MODE_READ_EOF
-            self._size = self._pos
-            return False
-        rawblock = self._fp.read(_BUFFER_SIZE)
+
+        if self._decompressor.unused_data:
+            rawblock = self._decompressor.unused_data
+        else:
+            rawblock = self._fp.read(_BUFFER_SIZE)
+
         if not rawblock:
-            raise EOFError("Compressed file ended before the "
-                           "end-of-stream marker was reached")
+            if self._decompressor.eof:
+                self._mode = _MODE_READ_EOF
+                self._size = self._pos
+                return False
+            else:
+                raise EOFError("Compressed file ended before the "
+                               "end-of-stream marker was reached")
+
+        # Continue to next stream.
+        if self._decompressor.eof:
+            self._decompressor = BZ2Decompressor()
+
         self._buffer = self._decompressor.decompress(rawblock)
         return True
 
@@ -384,9 +399,15 @@ def decompress(data):
     """
     if len(data) == 0:
         return b""
-    decomp = BZ2Decompressor()
-    result = decomp.decompress(data)
-    if not decomp.eof:
-        raise ValueError("Compressed data ended before the "
-                         "end-of-stream marker was reached")
-    return result
+
+    result = b""
+    while True:
+        decomp = BZ2Decompressor()
+        result += decomp.decompress(data)
+        if not decomp.eof:
+            raise ValueError("Compressed data ended before the "
+                             "end-of-stream marker was reached")
+        if not decomp.unused_data:
+            return result
+        # There is unused data left over. Proceed to next stream.
+        data = decomp.unused_data

Lib/test/test_bz2.py

@@ -84,9 +84,9 @@ class BZ2FileTest(BaseTest):
         else:
             return self.DATA
 
-    def createTempFile(self, crlf=False):
+    def createTempFile(self, crlf=False, streams=1):
         with open(self.filename, "wb") as f:
-            f.write(self.getData(crlf))
+            f.write(self.getData(crlf) * streams)
 
     def testRead(self):
         # "Test BZ2File.read()"
@@ -95,6 +95,26 @@ class BZ2FileTest(BaseTest):
             self.assertRaises(TypeError, bz2f.read, None)
             self.assertEqual(bz2f.read(), self.TEXT)
 
+    def testReadMultiStream(self):
+        # "Test BZ2File.read() with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            self.assertRaises(TypeError, bz2f.read, None)
+            self.assertEqual(bz2f.read(), self.TEXT * 5)
+
+    def testReadMonkeyMultiStream(self):
+        # "Test BZ2File.read() with a multi stream archive in which stream"
+        # "end is alined with internal buffer size"
+        buffer_size = bz2._BUFFER_SIZE
+        bz2._BUFFER_SIZE = len(self.DATA)
+        try:
+            self.createTempFile(streams=5)
+            with BZ2File(self.filename) as bz2f:
+                self.assertRaises(TypeError, bz2f.read, None)
+                self.assertEqual(bz2f.read(), self.TEXT * 5)
+        finally:
+            bz2._BUFFER_SIZE = buffer_size
+
     def testRead0(self):
         # "Test BBZ2File.read(0)"
         self.createTempFile()
@@ -114,6 +134,18 @@ class BZ2FileTest(BaseTest):
                 text += str
             self.assertEqual(text, self.TEXT)
 
+    def testReadChunk10MultiStream(self):
+        # "Test BZ2File.read() in chunks of 10 bytes with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            text = b''
+            while 1:
+                str = bz2f.read(10)
+                if not str:
+                    break
+                text += str
+            self.assertEqual(text, self.TEXT * 5)
+
     def testRead100(self):
         # "Test BZ2File.read(100)"
         self.createTempFile()
@@ -151,6 +183,15 @@ class BZ2FileTest(BaseTest):
             for line in sio.readlines():
                 self.assertEqual(bz2f.readline(), line)
 
+    def testReadLineMultiStream(self):
+        # "Test BZ2File.readline() with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            self.assertRaises(TypeError, bz2f.readline, None)
+            sio = BytesIO(self.TEXT * 5)
+            for line in sio.readlines():
+                self.assertEqual(bz2f.readline(), line)
+
     def testReadLines(self):
         # "Test BZ2File.readlines()"
         self.createTempFile()
@@ -159,6 +200,14 @@ class BZ2FileTest(BaseTest):
             sio = BytesIO(self.TEXT)
             self.assertEqual(bz2f.readlines(), sio.readlines())
 
+    def testReadLinesMultiStream(self):
+        # "Test BZ2File.readlines() with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            self.assertRaises(TypeError, bz2f.readlines, None)
+            sio = BytesIO(self.TEXT * 5)
+            self.assertEqual(bz2f.readlines(), sio.readlines())
+
     def testIterator(self):
         # "Test iter(BZ2File)"
         self.createTempFile()
@@ -166,6 +215,13 @@ class BZ2FileTest(BaseTest):
             sio = BytesIO(self.TEXT)
             self.assertEqual(list(iter(bz2f)), sio.readlines())
 
+    def testIteratorMultiStream(self):
+        # "Test iter(BZ2File) with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            sio = BytesIO(self.TEXT * 5)
+            self.assertEqual(list(iter(bz2f)), sio.readlines())
+
     def testClosedIteratorDeadlock(self):
         # "Test that iteration on a closed bz2file releases the lock."
         # http://bugs.python.org/issue3309
@@ -217,6 +273,17 @@ class BZ2FileTest(BaseTest):
             self.assertRaises(IOError, bz2f.write, b"a")
             self.assertRaises(IOError, bz2f.writelines, [b"a"])
 
+    def testAppend(self):
+        # "Test BZ2File.write()"
+        with BZ2File(self.filename, "w") as bz2f:
+            self.assertRaises(TypeError, bz2f.write)
+            bz2f.write(self.TEXT)
+        with BZ2File(self.filename, "a") as bz2f:
+            self.assertRaises(TypeError, bz2f.write)
+            bz2f.write(self.TEXT)
+        with open(self.filename, 'rb') as f:
+            self.assertEqual(self.decompress(f.read()), self.TEXT * 2)
+
     def testSeekForward(self):
         # "Test BZ2File.seek(150, 0)"
         self.createTempFile()
@@ -225,6 +292,14 @@ class BZ2FileTest(BaseTest):
             bz2f.seek(150)
             self.assertEqual(bz2f.read(), self.TEXT[150:])
 
+    def testSeekForwardMultiStream(self):
+        # "Test BZ2File.seek(150, 0) across stream boundaries"
+        self.createTempFile(streams=2)
+        with BZ2File(self.filename) as bz2f:
+            self.assertRaises(TypeError, bz2f.seek)
+            bz2f.seek(len(self.TEXT) + 150)
+            self.assertEqual(bz2f.read(), self.TEXT[150:])
+
     def testSeekBackwards(self):
         # "Test BZ2File.seek(-150, 1)"
         self.createTempFile()
@@ -233,6 +308,16 @@ class BZ2FileTest(BaseTest):
             bz2f.seek(-150, 1)
             self.assertEqual(bz2f.read(), self.TEXT[500-150:])
 
+    def testSeekBackwardsMultiStream(self):
+        # "Test BZ2File.seek(-150, 1) across stream boundaries"
+        self.createTempFile(streams=2)
+        with BZ2File(self.filename) as bz2f:
+            readto = len(self.TEXT) + 100
+            while readto > 0:
+                readto -= len(bz2f.read(readto))
+            bz2f.seek(-150, 1)
+            self.assertEqual(bz2f.read(), self.TEXT[100-150:] + self.TEXT)
+
     def testSeekBackwardsFromEnd(self):
         # "Test BZ2File.seek(-150, 2)"
         self.createTempFile()
@@ -240,6 +325,13 @@ class BZ2FileTest(BaseTest):
             bz2f.seek(-150, 2)
             self.assertEqual(bz2f.read(), self.TEXT[len(self.TEXT)-150:])
 
+    def testSeekBackwardsFromEndMultiStream(self):
+        # "Test BZ2File.seek(-1000, 2) across stream boundaries"
+        self.createTempFile(streams=2)
+        with BZ2File(self.filename) as bz2f:
+            bz2f.seek(-1000, 2)
+            self.assertEqual(bz2f.read(), (self.TEXT * 2)[-1000:])
+
     def testSeekPostEnd(self):
         # "Test BZ2File.seek(150000)"
         self.createTempFile()
@@ -248,6 +340,14 @@ class BZ2FileTest(BaseTest):
             self.assertEqual(bz2f.tell(), len(self.TEXT))
             self.assertEqual(bz2f.read(), b"")
 
+    def testSeekPostEndMultiStream(self):
+        # "Test BZ2File.seek(150000)"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            bz2f.seek(150000)
+            self.assertEqual(bz2f.tell(), len(self.TEXT) * 5)
+            self.assertEqual(bz2f.read(), b"")
+
     def testSeekPostEndTwice(self):
         # "Test BZ2File.seek(150000) twice"
         self.createTempFile()
@@ -257,6 +357,15 @@ class BZ2FileTest(BaseTest):
             self.assertEqual(bz2f.tell(), len(self.TEXT))
             self.assertEqual(bz2f.read(), b"")
 
+    def testSeekPostEndTwiceMultiStream(self):
+        # "Test BZ2File.seek(150000) twice with a multi stream archive"
+        self.createTempFile(streams=5)
+        with BZ2File(self.filename) as bz2f:
+            bz2f.seek(150000)
+            bz2f.seek(150000)
+            self.assertEqual(bz2f.tell(), len(self.TEXT) * 5)
+            self.assertEqual(bz2f.read(), b"")
+
     def testSeekPreStart(self):
         # "Test BZ2File.seek(-150, 0)"
         self.createTempFile()
@@ -265,6 +374,14 @@ class BZ2FileTest(BaseTest):
             self.assertEqual(bz2f.tell(), 0)
             self.assertEqual(bz2f.read(), self.TEXT)
 
+    def testSeekPreStartMultiStream(self):
+        # "Test BZ2File.seek(-150, 0) with a multi stream archive"
+        self.createTempFile(streams=2)
+        with BZ2File(self.filename) as bz2f:
+            bz2f.seek(-150)
+            self.assertEqual(bz2f.tell(), 0)
+            self.assertEqual(bz2f.read(), self.TEXT * 2)
+
     def testFileno(self):
         # "Test BZ2File.fileno()"
         self.createTempFile()
@@ -510,6 +627,11 @@ class FuncTest(BaseTest):
         # "Test decompress() function with incomplete data"
         self.assertRaises(ValueError, bz2.decompress, self.DATA[:-10])
 
+    def testDecompressMultiStream(self):
+        # "Test decompress() function for data with multiple streams"
+        text = bz2.decompress(self.DATA * 5)
+        self.assertEqual(text, self.TEXT * 5)
+
 def test_main():
     support.run_unittest(
         BZ2FileTest,

Misc/NEWS

@@ -161,6 +161,9 @@ Core and Builtins
 Library
 -------
 
+- Issue #1625: BZ2File and bz2.decompress() now support multi-stream files.
+  Initial patch by Nir Aides.
+
 - Issue #8796: codecs.open() calls the builtin open() function instead of using
   StreamReaderWriter. Deprecate StreamReader, StreamWriter, StreamReaderWriter,
   StreamRecoder and EncodedFile() of the codec module. Use the builtin open()