From eb8bc7a4ec1b5ca19a4b50040d96eb8b1024fd34 Mon Sep 17 00:00:00 2001
From: Nikolaus Schulz <microschulz@web.de>
Date: Sun, 26 Sep 2010 20:17:53 +0200
Subject: [PATCH] IMAP: correctly handle IMAP `literal' and unquoted `astring'
 in LIST reply

The LIST reply handling in imap_find_mailboxes() was buggy in several
ways.  It expected a non-NIL hierarchy delimiter, and it assumed that the
mailbox name in a LIST reply is always quoted; but it can be an astring, a
quoted string, or a literal (see RFC 3501 for the definitions of these
tokens).  These variants should now all be interpreted correctly.

In addition, quoted mailbox names are now handled more strictly to the
letter of the RFC; this only affects mailbox names containing a " or a
backslash, though.
---
 archivemail      | 41 +++++++++++++++++++++++++++++++++++------
 test_archivemail | 20 ++++++++++++++++++++
 2 files changed, 55 insertions(+), 6 deletions(-)

diff --git a/archivemail b/archivemail
index 9e8d82f..5931992 100755
--- a/archivemail
+++ b/archivemail
@@ -1410,6 +1410,21 @@ def _archive_imap(mailbox_name):
 
 ###############  IMAP  functions  ###############
 
+def imap_quote(astring):
+    """Quote an IMAP `astring' string (see RFC 3501, section "Formal Syntax")."""
+    if astring.startswith('"') and astring.endswith('"'):
+        quoted = astring
+    else:
+        quoted = '"' + astring.replace('\\', '\\\\').replace('"', '\\"') + '"'
+    return quoted
+
+def imap_unquote(quoted):
+    """Un-quote a `quoted' IMAP string (see RFC 3501, section "Formal Syntax")."""
+    if not (quoted.startswith('"') and quoted.endswith('"')):
+        unquoted = quoted
+    else:
+        unquoted = re.sub(r'\\(\\|")', r'\1', quoted[1:-1])
+    return unquoted
 
 def parse_imap_url(url): 
     """Parse IMAP URL and return username, password (if appliciable), servername
@@ -1462,7 +1477,7 @@ def imap_getdelim(imap_server):
         "server says '%s'" % response[0])
 
     # Response should be a list of strings like 
-    # '(\\Noselect \\HasChildren) "." "boxname"'
+    # '(\\Noselect \\HasChildren) "." boxname'
     # We parse only the first list item and just grab the delimiter. 
     m = re.match(r'\([^\)]*\) (?P<delim>"."|NIL)', response[0])
     if not m: 
@@ -1505,7 +1520,7 @@ def imap_smart_select(srv, mailbox):
         vprint("examining imap folder '%s' read-only" % mailbox)
     else:
         vprint("selecting imap folder '%s'" % mailbox)
-    result, response = srv.select(mailbox, roflag)
+    result, response = srv.select(imap_quote(mailbox), roflag)
     if result != 'OK':
         unexpected_error("selecting '%s' failed; server says: '%s'." \
                 % (mailbox, response[0]))
@@ -1533,12 +1548,12 @@ def imap_find_mailboxes(srv, mailbox):
             vprint("Looking for mailboxes matching '%s'..." % curbox)
         else:
             vprint("Looking for mailbox '%s'..." % curbox)
-        result, response = srv.list(pattern=curbox)
+        result, response = srv.list(pattern=imap_quote(curbox))
         if result != 'OK': 
             unexpected_error("LIST command failed; " \
                 "server says: '%s'" % response[0])
         # Say we queried for the mailbox "foo". 
-        # Upon success, response is e.g. ['(\\HasChildren) "." "foo"'].
+        # Upon success, response is e.g. ['(\\HasChildren) "." foo'].
         # Upon failure, response is [None].  Funky imaplib!
         if response[0] != None: 
             break
@@ -1546,8 +1561,22 @@ def imap_find_mailboxes(srv, mailbox):
         user_error("Cannot find mailbox '%s' on server." % mailbox)
     mailboxes = []
     for mailbox_data in response:
-        m = re.match(r'\((.*?)\) "." "(.*?)"', mailbox_data)
-        attrs, name = m.groups()
+        if not mailbox_data:    # imaplib sometimes returns an empty string
+            continue
+        try:
+            m = re.match(r'\((.*?)\) (?:"."|NIL) (.+)', mailbox_data)
+        except TypeError:
+            # May be a literal.  For literals, imaplib returns a tuple like
+            # ('(\\HasNoChildren) "." {12}', 'with "quote"').
+            m = re.match(r'\((.*?)\) (?:"."|NIL) \{\d+\}$', mailbox_data[0])
+            if m is None:
+                unexpected_error("cannot parse LIST reply %s" %
+                        (mailbox_data,))
+            attrs = m.group(1)
+            name = mailbox_data[1]
+        else:
+            attrs, name = m.groups()
+            name = imap_unquote(name)
         if '\\noselect' in attrs.lower().split():
             vprint("skipping not selectable mailbox '%s'" % name)
             continue
diff --git a/test_archivemail b/test_archivemail
index c49508f..6fff170 100755
--- a/test_archivemail
+++ b/test_archivemail
@@ -617,6 +617,26 @@ class TestParseIMAPUrl(unittest.TestCase):
         archivemail.options.verbose = False
         archivemail.options.pwfile = None
 
+########## quoting and un-quoting of IMAP strings ##########
+
+class TestIMAPQuoting(unittest.TestCase):
+    stringlist = (
+            ('{braces} and space', '"{braces} and space"'),
+            ('\\backslash', '"\\\\backslash"'),
+            ('with "quotes" inbetween', '"with \\"quotes\\" inbetween"'),
+            ('ending with "quotes"', '"ending with \\"quotes\\""'),
+            ('\\"backslash before quote', '"\\\\\\"backslash before quote"')
+    )
+
+    def testQuote(self):
+        for unquoted, quoted in self.stringlist:
+            self.assertEqual(archivemail.imap_quote(unquoted), quoted)
+
+    def testUnquote(self):
+        for unquoted, quoted in self.stringlist:
+            self.assertEqual(unquoted, archivemail.imap_unquote(quoted))
+
+
 ########## acceptance testing ###########
 
 class TestArchive(TestCaseInTempdir):