Mercurial > dovecot > original-hg > dovecot-1.2
comparison NEWS @ 965:6f005d5d9931 HEAD
0.99.6 out
author | Timo Sirainen <tss@iki.fi> |
---|---|
date | Mon, 13 Jan 2003 23:49:04 +0200 |
parents | 4f619b97ee65 |
children | 7bd8508ed0fa |
comparison
equal
deleted
inserted
replaced
964:951c022b9f8f | 965:6f005d5d9931 |
---|---|
1 v0.99.6 2003-01-13 Timo Sirainen <tss@iki.fi> | |
2 | |
3 + THREAD=REFERENCES extension support. ORDEREDSUBJECT would be easy to | |
4 add, but I think it's pretty useless. | |
5 + SORT is much faster now. | |
6 + mbox: If ~/mail directory isn't found, create it. | |
7 + Log login usernames | |
8 * Some coding style changes (less typedefs) | |
9 - Mails with nested MIME parts might have caused incorrect BODY and | |
10 BODYSTRUCTURE fetches and sometimes might have crashed dovecot | |
11 (assert at imap-bodystructure.c). If client had already successfully | |
12 done the BODY fetching a couple of times, the bug couldn't happen | |
13 anymore since Dovecot then began caching the BODY data. So, this | |
14 mostly happened with new users. | |
15 - non-UID SEARCH might gave wrong replies in certain conditions. | |
16 - SORT replied always with UIDs instead of sequences. | |
17 - If authentication was aborted by client ("*" reply to AUTHENTICATE), | |
18 the login process crashed later. | |
19 - STATUS command gave invalid reply for mailboxes with spaces in name | |
20 - Timezones were parsed wrong with message dates | |
21 - Digest-MD5: We used "qop-options" instead of "qop", which was | |
22 incompatible with at least Cyrus SASL. | |
23 - Realms in passwd-file were buggy | |
24 - Literals didn't work when logging in | |
25 - Crashed if it had to wait for mbox lock | |
26 - With invalid configuration auth and login processes were just dying | |
27 and master filling log files infinitely. | |
28 - We didn't work with some 64bit systems | |
29 | |
1 v0.99.5 2003-01-02 Timo Sirainen <tss@iki.fi> | 30 v0.99.5 2003-01-02 Timo Sirainen <tss@iki.fi> |
2 | 31 |
3 * This release includes a lot of code cleanups, especially related to | 32 * This release includes a lot of code cleanups, especially related to |
4 security. Direct buffer modifying was replaced in several places | 33 security. Direct buffer modifying was replaced in several places |
5 with a new buffer API, which provides a safe way to do it. Code that | 34 with a new buffer API, which provides a safe way to do it. Code that |