Reuse 'concat' handling for is_zipfile

Author: thatch

Lib/zipfile/__init__.py

@@ -239,8 +239,9 @@ def _check_zipfile(fp):
             if endrec[_ECD_ENTRIES_TOTAL] == 0 and endrec[_ECD_SIZE] == 0 and endrec[_ECD_OFFSET] == 0:
                 return True     # Empty zipfiles are still zipfiles
             elif endrec[_ECD_DISK_NUMBER] == endrec[_ECD_DISK_START]:
-                fp.seek(endrec[_ECD_OFFSET])    # Central directory is on the same disk
-                if fp.tell() == endrec[_ECD_OFFSET] and endrec[_ECD_SIZE] >= sizeCentralDir:
+                # Central directory is on the same disk
+                fp.seek(sum(_handle_prepended_data(endrec)))
+                if endrec[_ECD_SIZE] >= sizeCentralDir:
                     data = fp.read(sizeCentralDir)   # CD is where we expect it to be
                     if len(data) == sizeCentralDir:
                         centdir = struct.unpack(structCentralDir, data) # CD is the right size
@@ -268,6 +269,22 @@ def is_zipfile(filename):
         pass
     return result
 
+def _handle_prepended_data(endrec, debug=0):
+    size_cd = endrec[_ECD_SIZE]             # bytes in central directory
+    offset_cd = endrec[_ECD_OFFSET]         # offset of central directory
+
+    # "concat" is zero, unless zip was concatenated to another file
+    concat = endrec[_ECD_LOCATION] - size_cd - offset_cd
+    if endrec[_ECD_SIGNATURE] == stringEndArchive64:
+        # If Zip64 extension structures are present, account for them
+        concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator)
+
+    if debug > 2:
+        inferred = concat + offset_cd
+        print("given, inferred, offset", offset_cd, inferred, concat)
+
+    return offset_cd, concat
+
 def _EndRecData64(fpin, offset, endrec):
     """
     Read the ZIP64 end-of-archive records and use that to update endrec
@@ -1511,28 +1528,21 @@ def _RealGetContents(self):
             raise BadZipFile("File is not a zip file")
         if self.debug > 1:
             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
 
-        # "concat" is zero, unless zip was concatenated to another file
-        concat = endrec[_ECD_LOCATION] - size_cd - offset_cd
-        if endrec[_ECD_SIGNATURE] == stringEndArchive64:
-            # If Zip64 extension structures are present, account for them
-            concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator)
+        offset_cd, concat = _handle_prepended_data(endrec, self.debug)
+
+        # self.start_dir:  Position of start of central directory
+        self.start_dir = offset_cd + concat
 
         # store the offset to the beginning of data for the
         # .data_offset property
         self._data_offset = concat
 
-        if self.debug > 2:
-            inferred = concat + offset_cd
-            print("given, inferred, offset", offset_cd, inferred, concat)
-        # self.start_dir:  Position of start of central directory
-        self.start_dir = offset_cd + concat
         if self.start_dir < 0:
             raise BadZipFile("Bad offset for central directory")
         fp.seek(self.start_dir, 0)
+        size_cd = endrec[_ECD_SIZE]
         data = fp.read(size_cd)
         fp = io.BytesIO(data)
         total = 0