#14399: zipfile now correctly handles comments added to empty zipfiles.

Patch by Serhiy Storchaka.

This also moves the TypeError that results from trying to use a unicode
comment from the 'close' step to the point at which the comment is added to
the zipfile.

diff --git a/Lib/test/test_zipfile.py b/Lib/test/test_zipfile.py
index bb0d79ad4eeb53edf8a273414a818741ebb82ee8..c0675c67a4c95761bc654d48aa22717e1bb5ea27 100644 (file)
--- a/Lib/test/test_zipfile.py
+++ b/Lib/test/test_zipfile.py
@@ -972,6 +972,28 @@ class OtherTests(unittest.TestCase):
         with zipfile.ZipFile(TESTFN, mode="r") as zipfr:
             self.assertEqual(zipfr.comment, comment2)
 
+    def test_unicode_comment(self):
+        with zipfile.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf:
+            zipf.writestr("foo.txt", "O, for a Muse of Fire!")
+            with self.assertRaises(TypeError):
+                zipf.comment = "this is an error"
+
+    def test_change_comment_in_empty_archive(self):
+        with zipfile.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf:
+            self.assertFalse(zipf.filelist)
+            zipf.comment = b"this is a comment"
+        with zipfile.ZipFile(TESTFN, "r") as zipf:
+            self.assertEqual(zipf.comment, b"this is a comment")
+
+    def test_change_comment_in_nonempty_archive(self):
+        with zipfile.ZipFile(TESTFN, "w", zipfile.ZIP_STORED) as zipf:
+            zipf.writestr("foo.txt", "O, for a Muse of Fire!")
+        with zipfile.ZipFile(TESTFN, "a", zipfile.ZIP_STORED) as zipf:
+            self.assertTrue(zipf.filelist)
+            zipf.comment = b"this is a comment"
+        with zipfile.ZipFile(TESTFN, "r") as zipf:
+            self.assertEqual(zipf.comment, b"this is a comment")
+
     def check_testzip_with_bad_crc(self, compression):
         """Tests that files with bad CRCs return their name from testzip."""
         zipdata = self.zips_with_bad_crc[compression]

diff --git a/Lib/zipfile.py b/Lib/zipfile.py
index 6ca269fe175099ff1d218e1d071d269f916dcc31..435a3e95cb9bd3e41461abda2ddf4f3db733a8ee 100644 (file)
--- a/Lib/zipfile.py
+++ b/Lib/zipfile.py
@@ -698,7 +698,7 @@ class ZipFile:
         self.compression = compression  # Method of compression
         self.mode = key = mode.replace('b', '')[0]
         self.pwd = None
-        self.comment = b''
+        self._comment = b''
 
         # Check if we were passed a file-like object
         if isinstance(file, str):
@@ -774,7 +774,7 @@ class ZipFile:
             print(endrec)
         size_cd = endrec[_ECD_SIZE]             # bytes in central directory
         offset_cd = endrec[_ECD_OFFSET]         # offset of central directory
-        self.comment = endrec[_ECD_COMMENT]     # archive comment
+        self._comment = endrec[_ECD_COMMENT]    # archive comment
 
         # "concat" is zero, unless zip was concatenated to another file
         concat = endrec[_ECD_LOCATION] - size_cd - offset_cd
@@ -886,6 +886,24 @@ class ZipFile:
         else:
             self.pwd = None
 
+    @property
+    def comment(self):
+        """The comment text associated with the ZIP file."""
+        return self._comment
+
+    @comment.setter
+    def comment(self, comment):
+        if not isinstance(comment, bytes):
+            raise TypeError("comment: expected bytes, got %s" % type(comment))
+        # check for valid comment length
+        if len(comment) >= ZIP_MAX_COMMENT:
+            if self.debug:
+                print('Archive comment is too long; truncating to %d bytes'
+                        % ZIP_MAX_COMMENT)
+            comment = comment[:ZIP_MAX_COMMENT]
+        self._comment = comment
+        self._didModify = True
+
     def read(self, name, pwd=None):
         """Return file bytes (as a string) for name."""
         with self.open(name, "r", pwd) as fp:
@@ -1287,18 +1305,11 @@ class ZipFile:
                 centDirSize = min(centDirSize, 0xFFFFFFFF)
                 centDirOffset = min(centDirOffset, 0xFFFFFFFF)
 
-            # check for valid comment length
-            if len(self.comment) >= ZIP_MAX_COMMENT:
-                if self.debug > 0:
-                    msg = 'Archive comment is too long; truncating to %d bytes' \
-                          % ZIP_MAX_COMMENT
-                self.comment = self.comment[:ZIP_MAX_COMMENT]
-
             endrec = struct.pack(structEndArchive, stringEndArchive,
                                  0, 0, centDirCount, centDirCount,
-                                 centDirSize, centDirOffset, len(self.comment))
+                                 centDirSize, centDirOffset, len(self._comment))
             self.fp.write(endrec)
-            self.fp.write(self.comment)
+            self.fp.write(self._comment)
             self.fp.flush()
 
         if not self._filePassed:

diff --git a/Misc/ACKS b/Misc/ACKS
index aab5b20a599a359f34b85e7917cba8c163fba886..625689db833d19fc04995fd24c35493bd7f8000e 100644 (file)
--- a/Misc/ACKS
+++ b/Misc/ACKS
@@ -887,6 +887,7 @@ Richard Stoakley
 Peter Stoehr
 Casper Stoel
 Michael Stone
+Serhiy Storchaka
 Ken Stox
 Dan Stromberg
 Daniel Stutzbach

diff --git a/Misc/NEWS b/Misc/NEWS
index a9fbd8dc6da197b1d527aa6b7477cd43a0dbe2e8..5cccc4d046fd5b385072077fce47e2f41310d1e0 100644 (file)
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -39,6 +39,11 @@ Core and Builtins
 Library
 -------
 
+- Issue #14399: zipfile now correctly adds a comment even when the zipfile
+  being created is otherwise empty.  In addition, the TypeError that results
+  from trying to set a non-binary value as a comment is now now raised at the
+  time the comment is set rather than at the time the zipfile is written.
+
 - Issue #7978: socketserver now restarts the select() call when EINTR is
   returned.  This avoids crashing the server loop when a signal is received.
   Patch by Jerzy Kozera.