From 5248103ef910774628639c767b8fbcf88684e013 Mon Sep 17 00:00:00 2001
From: Fred Drake <fdrake@acm.org>
Date: Thu, 4 Dec 2008 18:25:17 +0000
Subject: Issue #1055234: cgi.parse_header(): Fixed parsing of header
 parameters to support unusual filenames (such as those containing
 semi-colons) in Content-Disposition headers.

---
 Lib/cgi.py           | 18 +++++++++++++++---
 Lib/test/test_cgi.py | 26 ++++++++++++++++++++++++++
 Misc/NEWS            |  4 ++++
 3 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/Lib/cgi.py b/Lib/cgi.py
index 33b91bf..0bb5b8ea 100755
--- a/Lib/cgi.py
+++ b/Lib/cgi.py
@@ -289,16 +289,28 @@ def parse_multipart(fp, pdict):
     return partdict
 
 
+def _parseparam(s):
+    while s[:1] == ';':
+        s = s[1:]
+        end = s.find(';')
+        while end > 0 and s.count('"', 0, end) % 2:
+            end = s.find(';', end + 1)
+        if end < 0:
+            end = len(s)
+        f = s[:end]
+        yield f.strip()
+        s = s[end:]
+
 def parse_header(line):
     """Parse a Content-type like header.
 
     Return the main content-type and a dictionary of options.
 
     """
-    plist = [x.strip() for x in line.split(';')]
-    key = plist.pop(0).lower()
+    parts = _parseparam(';' + line)
+    key = parts.next()
     pdict = {}
-    for p in plist:
+    for p in parts:
         i = p.find('=')
         if i >= 0:
             name = p[:i].strip().lower()
diff --git a/Lib/test/test_cgi.py b/Lib/test/test_cgi.py
index fa1d37f..800f629 100644
--- a/Lib/test/test_cgi.py
+++ b/Lib/test/test_cgi.py
@@ -354,6 +354,32 @@ this is the content of the fake file
         self.assertEqual([('a', 'A1'), ('b', 'B2'), ('B', 'B3')],
                          cgi.parse_qsl('a=A1&b=B2&B=B3'))
 
+    def test_parse_header(self):
+        self.assertEqual(
+            cgi.parse_header("text/plain"),
+            ("text/plain", {}))
+        self.assertEqual(
+            cgi.parse_header("text/vnd.just.made.this.up ; "),
+            ("text/vnd.just.made.this.up", {}))
+        self.assertEqual(
+            cgi.parse_header("text/plain;charset=us-ascii"),
+            ("text/plain", {"charset": "us-ascii"}))
+        self.assertEqual(
+            cgi.parse_header('text/plain ; charset="us-ascii"'),
+            ("text/plain", {"charset": "us-ascii"}))
+        self.assertEqual(
+            cgi.parse_header('text/plain ; charset="us-ascii"; another=opt'),
+            ("text/plain", {"charset": "us-ascii", "another": "opt"}))
+        self.assertEqual(
+            cgi.parse_header('attachment; filename="silly.txt"'),
+            ("attachment", {"filename": "silly.txt"}))
+        self.assertEqual(
+            cgi.parse_header('attachment; filename="strange;name"'),
+            ("attachment", {"filename": "strange;name"}))
+        self.assertEqual(
+            cgi.parse_header('attachment; filename="strange;name";size=123;'),
+            ("attachment", {"filename": "strange;name", "size": "123"}))
+
 
 def test_main():
     run_unittest(CgiTests)
diff --git a/Misc/NEWS b/Misc/NEWS
index 78578e3..300c819 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -60,6 +60,10 @@ Core and Builtins
 Library
 -------
 
+- Issue #1055234: cgi.parse_header(): Fixed parsing of header parameters to
+  support unusual filenames (such as those containing semi-colons) in
+  Content-Disposition headers.
+
 - Issue #4384: Added integration with warnings module using captureWarnings().
   This change includes a NullHandler which does nothing; it will be of use to
   library developers who want to avoid the "No handlers could be found for
-- 
cgit v0.12