Mon, 16 Sep 2002 08:44:59 +0300 |
Timo Sirainen |
bugfix
HEAD
|
Mon, 16 Sep 2002 08:40:16 +0300 |
Timo Sirainen |
don't set flags dirty when they were updated by external change
HEAD
|
Mon, 16 Sep 2002 08:34:44 +0300 |
Timo Sirainen |
data file wasn't initialized properly
HEAD
|
Mon, 16 Sep 2002 08:28:17 +0300 |
Timo Sirainen |
don't set mmap_used_length to 0 before msync()ing with it..
HEAD
|
Mon, 16 Sep 2002 08:20:54 +0300 |
Timo Sirainen |
report --without-gnutls better..
HEAD
|
Mon, 16 Sep 2002 08:19:40 +0300 |
Timo Sirainen |
another typo .. maybe --without-gnutls works now :)
HEAD
|
Mon, 16 Sep 2002 08:18:00 +0300 |
Timo Sirainen |
typo..
HEAD
|
Mon, 16 Sep 2002 08:16:17 +0300 |
Timo Sirainen |
added --without-gnutls option
HEAD
|
Mon, 16 Sep 2002 08:09:58 +0300 |
Timo Sirainen |
..and don't #undef the linux mremap() support..
HEAD
|
Mon, 16 Sep 2002 08:09:34 +0300 |
Timo Sirainen |
portability fixes
HEAD
|
Mon, 16 Sep 2002 08:00:37 +0300 |
Timo Sirainen |
mbox fixes
HEAD
|
Mon, 16 Sep 2002 07:28:13 +0300 |
Timo Sirainen |
fixes
HEAD
|
Mon, 16 Sep 2002 07:22:52 +0300 |
Timo Sirainen |
hash creation was buggy
HEAD
|
Mon, 16 Sep 2002 07:16:16 +0300 |
Timo Sirainen |
bugfixes
HEAD
|
Mon, 16 Sep 2002 07:01:35 +0300 |
Timo Sirainen |
Fixes for no diskspace handling. seems to work now.
HEAD
|
Mon, 16 Sep 2002 06:33:06 +0300 |
Timo Sirainen |
previous t_push()/t_pop() changes break things. fixed :)
HEAD
|
Mon, 16 Sep 2002 06:13:10 +0300 |
Timo Sirainen |
updated
HEAD
|
Mon, 16 Sep 2002 06:11:40 +0300 |
Timo Sirainen |
Modify log is now able to stay in memory. Some other fixes/cleanups.
HEAD
|
Mon, 16 Sep 2002 05:07:17 +0300 |
Timo Sirainen |
t_push()/t_pop() around various *sprintf() functions, so "%m" format won't
HEAD
|
Sun, 15 Sep 2002 14:09:08 +0300 |
Timo Sirainen |
s/io_buffer_read_data/io_buffer_read_data_blocking/ and fixed the various
HEAD
|
Sun, 15 Sep 2002 11:36:14 +0300 |
Timo Sirainen |
we shouldn't crash anymore when modify log gets full
HEAD
|
Sun, 15 Sep 2002 10:20:35 +0300 |
Timo Sirainen |
minor fixes
HEAD
|
Sun, 15 Sep 2002 10:15:43 +0300 |
Timo Sirainen |
messages flags were added twice to header
HEAD
|
Sun, 15 Sep 2002 10:12:31 +0300 |
Timo Sirainen |
always verify the seen/deleted message counts in index header
HEAD
|
Sun, 15 Sep 2002 10:00:39 +0300 |
Timo Sirainen |
removed extra whitespace
HEAD
|
Sun, 15 Sep 2002 09:48:54 +0300 |
Timo Sirainen |
creating data file didn't work
HEAD
|
Sun, 15 Sep 2002 09:31:47 +0300 |
Timo Sirainen |
mremap check again..
HEAD
|
Sun, 15 Sep 2002 09:31:25 +0300 |
Timo Sirainen |
mremap() check
HEAD
|
Sun, 15 Sep 2002 09:30:29 +0300 |
Timo Sirainen |
More cleanups. lib-storage should handle "out of disk space" conditions
HEAD
|
Sat, 14 Sep 2002 14:09:42 +0300 |
Timo Sirainen |
lots of cleanups. also index/datafile is now capable of staying in memory,
HEAD
|
Fri, 13 Sep 2002 03:01:23 +0300 |
Timo Sirainen |
Added extra functions for easier printing of error messages. Moved
HEAD
|
Fri, 13 Sep 2002 02:09:45 +0300 |
Timo Sirainen |
make sure failure handlers don't modify errno.
HEAD
|
Wed, 11 Sep 2002 20:45:03 +0300 |
Timo Sirainen |
Index data file wasn't re-mmap()ed always when file size was changed,
HEAD
|
Tue, 10 Sep 2002 06:40:58 +0300 |
Timo Sirainen |
expunge() shouldn't update mmap, since we could be just going through the
HEAD
|
Tue, 10 Sep 2002 06:35:50 +0300 |
Timo Sirainen |
Don't fsck() at mbox_index_rewrite() if there's no dirty flags.
HEAD
|
Tue, 10 Sep 2002 06:32:42 +0300 |
Timo Sirainen |
SEARCH fixes
HEAD
|
Tue, 10 Sep 2002 06:26:05 +0300 |
Timo Sirainen |
io_buffer_read_mmaped() now also returns correct amount of bytes
HEAD
|
Tue, 10 Sep 2002 06:24:02 +0300 |
Timo Sirainen |
fixes for io_buffer_read_max()
HEAD
|
Tue, 10 Sep 2002 06:18:34 +0300 |
Timo Sirainen |
cleanup
HEAD
|
Tue, 10 Sep 2002 06:06:46 +0300 |
Timo Sirainen |
message/rfc822 mime parts weren't parsed correctly
HEAD
|
Tue, 10 Sep 2002 05:58:48 +0300 |
Timo Sirainen |
crashfix for searching from message body
HEAD
|
Tue, 10 Sep 2002 05:52:09 +0300 |
Timo Sirainen |
Actually, don't bother fsyncing after each written mail, it slows down too
HEAD
|
Tue, 10 Sep 2002 05:45:24 +0300 |
Timo Sirainen |
Don't rewrite mbox if no flags were changed.
HEAD
|
Tue, 10 Sep 2002 04:52:08 +0300 |
Timo Sirainen |
Date-field wasn't saved when ENVELOPE was cached into index.
HEAD
|
Tue, 10 Sep 2002 04:17:43 +0300 |
Timo Sirainen |
Broke saving new mail
HEAD
|
Tue, 10 Sep 2002 01:49:34 +0300 |
Timo Sirainen |
cleanups / reliability fixes
HEAD
|
Tue, 10 Sep 2002 00:45:39 +0300 |
Timo Sirainen |
error message update
HEAD
|
Mon, 09 Sep 2002 16:32:00 +0300 |
Timo Sirainen |
cast i_toupper() and i_tolower() return as char, so comparing highascii
HEAD
|
Mon, 09 Sep 2002 16:30:48 +0300 |
Timo Sirainen |
typofix for parsing 'S' timeszone
HEAD
|
Mon, 09 Sep 2002 16:22:27 +0300 |
Timo Sirainen |
Highascii + space at end of some ENVELOPE header got into infinite loop.
HEAD
|
Mon, 09 Sep 2002 15:03:44 +0300 |
Timo Sirainen |
crashfix if client closed before SSL handshake was done
HEAD
|
Mon, 09 Sep 2002 15:02:47 +0300 |
Timo Sirainen |
crashfix
HEAD
|
Mon, 09 Sep 2002 14:55:27 +0300 |
Timo Sirainen |
custom flag & copy crash fixes.
HEAD
|
Mon, 09 Sep 2002 10:18:50 +0300 |
Timo Sirainen |
EXPUNGE works with mbox. finally.
HEAD
|
Mon, 09 Sep 2002 10:18:37 +0300 |
Timo Sirainen |
offset wasn't updated properly to output buffers with sendfile().
HEAD
|
Mon, 09 Sep 2002 05:11:01 +0300 |
Timo Sirainen |
X-IMAPbase is now parsed correctly.
HEAD
|
Sun, 08 Sep 2002 18:34:27 +0300 |
Timo Sirainen |
Replaced INT_TO_POINTER and POINTER_TO_INT macros with POINTER_CAST and
HEAD
|
Sun, 08 Sep 2002 18:22:45 +0300 |
Timo Sirainen |
more size_t fixes
HEAD
|
Sun, 08 Sep 2002 18:19:39 +0300 |
Timo Sirainen |
more size_t fixes.
HEAD
|
Sun, 08 Sep 2002 18:09:05 +0300 |
Timo Sirainen |
more size_t fixes
HEAD
|
Sun, 08 Sep 2002 18:08:37 +0300 |
Timo Sirainen |
ssize_t should be checked against long before long long.
HEAD
|
Sun, 08 Sep 2002 18:02:26 +0300 |
Timo Sirainen |
size_t breakages
HEAD
|
Sun, 08 Sep 2002 18:00:28 +0300 |
Timo Sirainen |
forgot to change some ints into ssize_t
HEAD
|
Sun, 08 Sep 2002 17:59:13 +0300 |
Timo Sirainen |
..and now it uses size_t properly..
HEAD
|
Sun, 08 Sep 2002 17:58:30 +0300 |
Timo Sirainen |
nearest_power() uses now size_t.
HEAD
|
Sun, 08 Sep 2002 17:54:17 +0300 |
Timo Sirainen |
result of ssize_t check wasn't reported
HEAD
|
Sun, 08 Sep 2002 17:53:18 +0300 |
Timo Sirainen |
fixed ssize_t and off_t checks.
HEAD
|
Sun, 08 Sep 2002 17:39:05 +0300 |
Timo Sirainen |
move size_t fixes
HEAD
|
Sun, 08 Sep 2002 16:20:28 +0300 |
Timo Sirainen |
size_t fixes for lib/. Changed OFF_T_FORMAT to PRIuOFF_T which is more
HEAD
|
Sun, 08 Sep 2002 16:08:26 +0300 |
Timo Sirainen |
flags-file isn't needed anymore
HEAD
|
Sun, 08 Sep 2002 16:06:21 +0300 |
Timo Sirainen |
mbox compilation fix
HEAD
|
Sun, 08 Sep 2002 15:12:41 +0300 |
Timo Sirainen |
partial changes to add X-IMAPbase parser update custom flags
HEAD
|
Sun, 08 Sep 2002 14:36:03 +0300 |
Timo Sirainen |
added dec2str() function and largest_t type, removed itoa() and ltoa()
HEAD
|
Sun, 08 Sep 2002 13:25:02 +0300 |
Timo Sirainen |
Better fix for POINTER_TO_UINT
HEAD
|
Sun, 08 Sep 2002 13:17:01 +0300 |
Timo Sirainen |
POINTER_TO_UINT() fixed to not give warnings with 64bit pointers.
HEAD
|
Sun, 08 Sep 2002 12:52:19 +0300 |
Timo Sirainen |
missed header file with some conditions.
HEAD
|
Sun, 08 Sep 2002 02:31:56 +0300 |
Timo Sirainen |
Moved custom flags handling into lib-index.
HEAD
|
Sat, 07 Sep 2002 22:01:14 +0300 |
Timo Sirainen |
nonblocking handshake was a bit buggy
HEAD
|
Sat, 07 Sep 2002 05:09:53 +0300 |
Timo Sirainen |
Don't log PUSH_ERRORs.
HEAD
|
Sat, 07 Sep 2002 05:06:36 +0300 |
Timo Sirainen |
updated error messages for FETCH/STORE when expunged messages were
HEAD
|
Sat, 07 Sep 2002 04:52:21 +0300 |
Timo Sirainen |
Moved lib-index/mail-messageset.* to lib-storage/index/index-messageset.*.
HEAD
|
Sat, 07 Sep 2002 04:46:07 +0300 |
Timo Sirainen |
Put back the "not all messages were found" error to
HEAD
|
Sat, 07 Sep 2002 04:32:49 +0300 |
Timo Sirainen |
Don't strip extra spaces after "field: ".
HEAD
|
Sat, 07 Sep 2002 00:51:21 +0300 |
Timo Sirainen |
Convert 8bit text in headers to quoted-printable in ENVELOPE.
HEAD
|
Fri, 06 Sep 2002 22:27:32 +0300 |
Timo Sirainen |
typofix
HEAD
|
Fri, 06 Sep 2002 20:29:16 +0300 |
Timo Sirainen |
minor cleanup
HEAD
|
Fri, 06 Sep 2002 20:28:15 +0300 |
Timo Sirainen |
comment update
HEAD
|
Fri, 06 Sep 2002 20:27:11 +0300 |
Timo Sirainen |
Added io_buffer_read_blocking() which can be used to read data blockingly,
HEAD
|
Fri, 06 Sep 2002 18:11:31 +0300 |
Timo Sirainen |
comment cleanup
HEAD
|
Fri, 06 Sep 2002 16:45:50 +0300 |
Timo Sirainen |
continuing mbox updates
HEAD
|
Fri, 06 Sep 2002 16:45:15 +0300 |
Timo Sirainen |
Whops, didn't compile :) Removed beginning of wrong way to fix custom flags.
HEAD
|
Fri, 06 Sep 2002 16:43:58 +0300 |
Timo Sirainen |
Bigger changes.. full_virtual_size was removed from index record and
HEAD
|
Fri, 06 Sep 2002 16:29:51 +0300 |
Timo Sirainen |
net_accept() set listening socket nonblocking, not the new socket. thanks to
HEAD
|
Thu, 05 Sep 2002 15:52:49 +0300 |
Timo Sirainen |
Flush output buffer to client at exit, and send a nice "BYE Server shutting
HEAD
|
Thu, 05 Sep 2002 15:47:54 +0300 |
Timo Sirainen |
Allow UID messageset traversal to go past the synced messages.
HEAD
|
Thu, 05 Sep 2002 14:49:13 +0300 |
Timo Sirainen |
also, it shouldn't crash :)
HEAD
|
Thu, 05 Sep 2002 14:47:26 +0300 |
Timo Sirainen |
EXPUNGE shouldn't do "internal error" when there's nothing to expunge..
HEAD
|
Thu, 05 Sep 2002 04:55:06 +0300 |
Timo Sirainen |
io_buffer_send_iobuffer() was buggy with non-sendfile() method.
HEAD
|
Wed, 04 Sep 2002 19:17:11 +0300 |
Timo Sirainen |
message_parse_header() was buggy with big headers, and io_buffer_read_data()
HEAD
|
Wed, 04 Sep 2002 19:15:44 +0300 |
Timo Sirainen |
FETCH BODY[HEADER.FIELDS ...] wasn't working
HEAD
|
Wed, 04 Sep 2002 18:24:05 +0300 |
Timo Sirainen |
s/updateid/sync_id/ like it was with modify log.
HEAD
|
Wed, 04 Sep 2002 18:19:42 +0300 |
Timo Sirainen |
index wasn't re-mmap()ed when it's size was changed. this is now done by
HEAD
|
Wed, 04 Sep 2002 17:45:59 +0300 |
Timo Sirainen |
make sure the hash gets rebuilt if it wasn't opened yet when message was
HEAD
|
Wed, 04 Sep 2002 14:53:15 +0300 |
Timo Sirainen |
mail_index_uidset_foreach() bugfix/cleanup for handling deleted messages
HEAD
|
Wed, 04 Sep 2002 13:58:06 +0300 |
Timo Sirainen |
Hash wasn't updated when compressing the index
HEAD
|
Wed, 04 Sep 2002 02:08:20 +0300 |
Timo Sirainen |
more verbose error messages for hash corruption
HEAD
|
Wed, 04 Sep 2002 02:04:38 +0300 |
Timo Sirainen |
mail_index_mark_flag_changes() didn't update deleted-flag changes if there
HEAD
|
Wed, 04 Sep 2002 02:01:42 +0300 |
Timo Sirainen |
mail_index_uidset_foreach() still sometimes returned "Some messages weren't
HEAD
|
Wed, 04 Sep 2002 01:56:48 +0300 |
Timo Sirainen |
always use nonblocking I/O
HEAD
|
Wed, 04 Sep 2002 01:51:20 +0300 |
Timo Sirainen |
index_expunge_seek_first() now separates between "not found" and "failure".
HEAD
|
Wed, 04 Sep 2002 00:21:40 +0300 |
Timo Sirainen |
don't warn about UNEXPECTED_PACKET_LENGTH errors in SSL, they just mean the
HEAD
|
Wed, 04 Sep 2002 00:17:57 +0300 |
Timo Sirainen |
mail_index_uidset_foreach() now never replies with "some messages weren't
HEAD
|
Wed, 04 Sep 2002 00:12:21 +0300 |
Timo Sirainen |
verify that mail_hash_lookup_uid() returns offset to correct record.
HEAD
|
Wed, 04 Sep 2002 00:02:32 +0300 |
Timo Sirainen |
more INBOX-sensitivity fixes
HEAD
|
Tue, 03 Sep 2002 23:52:56 +0300 |
Timo Sirainen |
"CREATE Inbox.sub" created it using that case, instead of "INBOX.sub" which
HEAD
|
Tue, 03 Sep 2002 23:44:35 +0300 |
Timo Sirainen |
"Mailbox doesn't exist" error now gives the name of the mailbox that was
HEAD
|
Tue, 03 Sep 2002 23:35:31 +0300 |
Timo Sirainen |
"revert back to shared lock" actually tried to set exclusive lock again
HEAD
|
Tue, 03 Sep 2002 23:34:05 +0300 |
Timo Sirainen |
messageset traversal: if user function requested stop, we leaked t_pop().
HEAD
|
Tue, 03 Sep 2002 23:28:09 +0300 |
Timo Sirainen |
s/io_buffer_send_buf/io_buffer_send_iobuffer/, also we set buf_errno when
HEAD
|
Tue, 03 Sep 2002 23:26:07 +0300 |
Timo Sirainen |
mail_hash_rebuild() now tries to lock itself, fixing crash if trying to
HEAD
|