1 Note that there is a separate todo.html with different content than this.
4 - audit if there are further untrusted data report_*() calls.
5 - Debian Bug #475239, MIME decoder may break up words (need to quote results)
6 - put bare IP addresses in brackets for SMTP (check if there are RFC
7 1123/5321/5322 differences)
8 - Fix further occurrences of SMTP reply code handling:
9 - for proper smtp_reponse caching of multiline codes (there are some)
10 - for stomping over control characters.
11 - check if smtpname and smtpaddress in particular work as advertised,
12 thread "Fetchmail with Postfix virtual users" around 2009-09-23 on
13 fetchmail-users@, by Joost Roeleveld and Gerard Seibert.
14 - virtual domain DOCUMENTATION (rewriting @example.com to
15 @virtual.example.com possible? Joost Roeleveld, thread "Fetchmail with
16 Postfix virtual users" around 2009-09-23 on fetchmail-users@).
19 - support NIL and strings where they are alternatives to literals
20 - Debian Bug #531589: fetchmail ignores SIGUSR1 in idle mode.
21 seems non-trivial to fix: in imap_idle(), we wait for untagged
22 responses, and may be deep in SSL_peek -- and that restarts the
23 underlying blocking read() from the socket, so we never break out of
24 the SSL_peek() with SIGUSR1.
25 - add repoll for all kinds of auth failures
26 (requires framework to track which auth failed in auto mode)
27 - SockOpen sometimes exits with errno == 0, confusing users (found with
28 Google RealTime on Twitter)
29 - make sure the man page completely lists all options (f. i. sslcertpath) in
31 - allow \Deleted without \Seen, rf.
32 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=466299
33 - document IMAP4 ACL requirements
34 - CRYPTO: log configured server name on certificate mismatch (perhaps pay
35 attention to via entries and stuff like that)
36 - CRYPTO: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=432618
37 * write a table of combinations of TLS/SSL options
38 - add To: header to warning mails (authfail for instance)
39 - Fix TOCTOU race around prc_filecheck*
40 - Read CAPABILITY from greeting if present, saves one round trip.
41 - Check if LAST argument is properly validated against message count.
42 - add Message-ID: header and other SHOULD headers to warning mails?
43 - report actual SMTP error with "SMTP listener refused delivery", sugg.
44 Richard Brooksby, fetchmail-users 2010-04-06.
47 - find a better replacement for sscanf parsing - we don't usually
48 detect errors in format strings such as "* %d FETCH " because we don't
49 check if the FETCH is (a) present, (b) consumed.
50 - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=471176
51 => fetchmail: support utf-8 encoding in log file
52 Revisit http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=400719
53 => syslog messages are localised
54 http://lists.debian.org/debian-devel/2006/11/msg01068.html
55 - when logging to syslog, disable locale?
56 - Get rid of peeking in socket.c? MSG_PEEK seems non-portable.
57 - revise ticker, and add something like .........20%.... ... 100%
61 - Convert POP3 UIDs to X-UIDL?
62 - fetch IMAP message in one go (fetchmail-devel by Adam Simpkins
63 <simpkins@cisco.com> around Nov 2nd)?
66 - Properly free host/user entries (through C++ class instantiation and destructors...)
67 - Remove stupid options, such as spambounce, or deferred bounces for anything
69 - Do not ever guess envelope from content headers such as To:/Cc:/Resent-To: or
71 - Replace sscanf/fscanf by functions that do range checking
72 (strtol/strtoul), and make arguments unsigned long.
74 - use PS_PROTOCOL for pre-/post-connect command failures - 6.3 uses
75 PS_SYNTAX, and that's not necessarily the case. At least if the
76 command terminates with a signal, we should report PS_PROTOCOL.
77 - revisit maximum allowed rcfile permissions, fix inconsistency
78 (silently allowing g+x).
79 - make UID code more efficient, parsing is O(n^2), should be no worse
80 than O(n log n), lookup is O(n), should be O(log n).
81 * Idea for C: use <search.h> tfind/tsearch. Need to split idlist up
82 so it only keeps the ids, and use an array to track status.
83 - help systematic debugging
84 - by making logging more strict (Postfix's msg_* as example??)
85 - by adding a --loggingtest or something that emits
86 a set of test log messages at various severity levels, in order to
87 make sure people get complete logs.
88 - by adding messages where fetchmail stuffed its output.
89 - Debian Bug #454291 fetchmail --quit: should check, that pid file
90 really contains pid of fetchmail process (Dmitry Nezhevenko)
92 - feature to skip first N or all messages upon first download, or fetch
94 - download only messages whose headers match a filter (by Toby, Usenet
95 Nov 2007 de.comm.software.mailserver
96 Message-ID: <1195033726.123704.296060@k79g2000hse.googlegroups.com>)
97 - feature request by Daniel Goering <g_daniel@gmx.net> on
98 fetchmail-devel 2007-11-15: populate .fetchids from current messages.
99 (it'll probably be useful to limit this to "all but the 10 latest" or
100 "all before date this-and-that")
101 - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=374514
102 "Deleting fetchids file" vs. POP3 - probably unneeded if we use IMAP UIDs.
103 - CRYPTO: change SSL/TLS UI incompatibly, making it easier to use
104 - CRYPTO: allow selection of SSL/TLS Ciphers
105 - CRYPTO: perhaps switch to libgsasl and libgnutls (which doesn't talk SSLv2,
106 but we should abandon that anyways).
107 - CRYPTO: perhaps port to NSS? Check license and features and required procedure
108 changes. - Redhat Bugs #333741 (crypto consolidation), #346891 (port fetchmail to NSS)
109 - CRYPTO: make the SSL default v3 (rather than v23).
110 - CRYPTO: remove sslfingerprint? too easily abused (see NEWS)
111 - CRYPTO: force sslcertck
112 - CRYPTO: by default forbid cleartext or other compromising password
113 schemes over insecure connections?
114 - put more hints to the FAQ (should we call it FGA?) as first support place
115 - make sure we print socket error messages such as connection reset by
116 peer to hint users the problem is not in fetchmail
117 - require envelope option for multidrop ; if not found in header, defer message?
118 - > b) When the envelope header is specified, but it isn't found, a
119 > warning is logged and the mail is forwarded to the postmaster
120 > ("WARNING: Envelope header $HEADER found, cannot identify the
121 > recipient. Forwarding to the postmaster")
122 Rob MacGregor 2007-03-16 Patch to the man page, fetchmail-devel@
123 - integrate Isaac Wilcox's test rig
124 - look at Tony Earnshaw's .spec file
125 - look at Gentoo patches
126 - remove dead replacement functions: strdup (Andreas Krennmair), ...
127 - more SMTP/LMTP error detail on message rejections even outside verbose mode.
128 Ian D. Allen, fetchmail-users. Two messages with examples.
129 What goes here? fetchmailrc location, server, port, user, folder, anything else?
130 - see if "AUTH SSH" can be made to work for POP3
131 - revisit BerliOS patches
132 - check recent list mail
133 - check Debian BTS and other bug trackers
134 - better logging (log all headers, log forward destination + method)
135 - check strict envelope N Received parsing,
136 see mail from Admin Att on fetchmail-users
137 - 6.3.4-pending-deletes.patch
138 - fetchmail -s with running daemon complains rather than silently
140 - send warning message when connection fails?
141 - check FAQ O5 - do we really prefer From: over envelope?!?
142 - add code to allow safe authentication schemes if TLS fails
143 - make APOP an authenticator, integrate with regular auto authentication
144 but stuff it at the end
145 - allow forcing RETR (RETR vs. TOP, fetchmail-users, drbob 2008-01-11)
147 - make logfile more useful (redirect not only in daemon mode)
148 - close/reopen logfile on certain signals (for newsyslog/logrotate
150 - for virtual mapping, we don't currently support local user aliases or
151 regexp matching with replacement. This would be useful for hosting
152 several virtual domains in one multidrop mailbox, as in
159 some@one.example maps to bob-one-some@serverdomain
160 another@two.example maps to bob-two-example@serverdomain
162 we can strip the bob-*, but but we don't support domain catchalls
163 here, and we cannot currently delegate or rewrite the remaining
164 one-* or two-* users in a useful way. All local parts of the upstream
165 domains must be known and mapped in the configuration file.
166 (See fetchmail-users, Michelle Konzack, 2009-05-03, "4 multidrop but one
167 destination server" and followup)
169 This probably entails some form of extended user mapping inside
170 fetchmail, for map_name() in transact.c, and possibly
171 find_server_names() ibidem.
172 - more verbose diagnostics, what, why, how, ... (what does fetchmail do, what
173 does it expect, what does it get instead, what does that mean, how can the
174 user fix it; references to the manual)
178 - Add info whether Keywords are global, server or user keywords
179 - consolidate multidrop documentation
180 - HOWTO (on configuration, and on SSL in particular)