Commit Graph

105 Commits

Author SHA1 Message Date
Kurt Zeilenga
3c598e89fb Happy new year 2004-01-01 19:15:16 +00:00
Kurt Zeilenga
85013a9d98 acknowledgements and notices 2003-11-26 22:51:15 +00:00
Kurt Zeilenga
327880b984 Notice updates 2003-11-25 23:17:08 +00:00
Kurt Zeilenga
db097047af code cleanup 2003-07-08 03:46:20 +00:00
Hallvard Furuseth
1fa40181d9 Cast unsigned char *p to char *. 2003-06-03 17:51:38 +00:00
Howard Chu
13aa48e47a Fix reading multibyte lengths 2003-05-22 06:24:23 +00:00
Kurt Zeilenga
6035ecfe7b Reverse unintended commit 2003-05-22 05:33:04 +00:00
Kurt Zeilenga
df1381fa38 cleanup 2003-05-22 05:32:06 +00:00
Kurt Zeilenga
6d36ba9d9f Add trickle code to test ber_get_next.
Use etest {...} | dtest {...} to test.
2003-05-22 04:37:44 +00:00
Kurt Zeilenga
bfa6808c48 Fix dtest ber_get_next() usage 2003-05-22 03:35:36 +00:00
Howard Chu
3d9fd17678 ber_get_next - set errno to 0 before calling ber_int_sb_read(), otherwise
an EOF will return with the previously set errno value.
2003-05-20 04:08:35 +00:00
Jong Hyuk Choi
4f58efc4d1 - change in message to modlist conversion 2003-05-20 01:45:12 +00:00
Howard Chu
3fc96a24ee More ITS#2490 - fix retries for multi-byte lengths 2003-05-19 12:14:15 +00:00
Howard Chu
5eb0ec070c ITS#2490 - more retry checks 2003-05-19 10:37:09 +00:00
Howard Chu
9757665aed ITS#2490 - allow retry for incomplete reads of BER tag & len 2003-05-19 10:33:18 +00:00
Howard Chu
a413bab19a Clean up prev commit 2003-04-24 04:20:17 +00:00
Howard Chu
b6111d793e Fix multi-byte tag parsing 2003-04-24 03:59:45 +00:00
Howard Chu
bcf7ab26e4 ITS#2465 fix? ber_get_next must read at least sizeof(tag)+sizeof(len)
which should be at most 8 bytes. However if we read more than the minimum
message length, we have a problem because we steal bytes from any following
message, and there is no buffer mechanism to push back excess data.
The shortest legitimate message is Unbind at 7 bytes, but there shouldn't
be anything following it. Abandon at 8 bytes is next, so always requesting
at least 8 bytes should be safe. Always requesting 9 was a problem.

Please double-check these assumptions...
2003-04-24 02:10:18 +00:00
Howard Chu
42e7845d81 More malloc context tweaks 2003-04-10 06:14:16 +00:00
Howard Chu
ca12bb75c7 Extended API for ber_mem functions, add context argument.
Must always use ber_mem funcs, even if CSRIMALLOC is present.
2003-04-09 13:25:11 +00:00
Howard Chu
e240135305 ITS#2434 ber accessors 2003-04-09 09:41:02 +00:00
Kurt Zeilenga
e41ca42de1 mark a few translatable strings 2003-04-06 06:19:13 +00:00
Kurt Zeilenga
1b37568b00 Use ifdef not comment to comment out code 2003-03-22 22:28:57 +00:00
Howard Chu
50ecec94eb Fix ITS#2344, prev commit changed too much. 2003-03-03 23:44:22 +00:00
Howard Chu
ec497cf529 Use fixed sizes for ber_get_next to avoid 64-bit hangs. 2003-03-02 04:02:39 +00:00
Howard Chu
5354dbf269 ber_flush shouldn't be logging with _ANY, changed to _TRACE 2003-01-31 22:45:19 +00:00
Kurt Zeilenga
bd9bf709aa ITS#2275: PROTOS ber_get_next() fix
Needs review!!!!
2003-01-22 22:29:11 +00:00
Kurt Zeilenga
6939c53170 Happy new year 2003-01-03 20:20:47 +00:00
Howard Chu
8b3b044dac Fix ITS#2246, fix in rev 1.73 was in the wrong place 2002-12-20 00:33:10 +00:00
Howard Chu
6a37f1ae80 Added ber_flatten2 - uses passed in struct berval, allocates copy or
directly re-uses existing ber buf.
2002-12-18 19:15:27 +00:00
Kurt Zeilenga
2a8d73e25a Bump EXBUFSIZ up a bit 2002-10-08 20:56:03 +00:00
Julius Enarusai
6107ba67d2 Coverted LDAP_LOG macro to use subsystem ID int values instead of string values 2002-07-11 20:33:24 +00:00
Howard Chu
2adfd47e88 Fix SEGV on single-byte connection input 2002-07-11 12:26:35 +00:00
Howard Chu
5e6c17bb2b ITS#1891 in ber_flush, no-op if towrite == 0 2002-06-18 01:00:27 +00:00
Howard Chu
bb17493d31 ITS#1732 signed/unsigned fixes 2002-04-11 08:47:37 +00:00
Howard Chu
933f6a5fdd For ITS#1601, add ber_init2() using given buffer in-place instead of
allocating a copy.
2002-02-14 13:32:40 +00:00
Howard Chu
b1c39eff65 Fix previous commit, need to detect very short packets. 2002-02-13 11:29:05 +00:00
Howard Chu
ab099caed5 ber_get_next: return error if decoded ber_len is smaller than actual count 2002-02-13 09:59:10 +00:00
Kurt Zeilenga
ce2d8ebc7e Misc cleanup, lint removal, and minor optimizations 2002-01-13 05:00:59 +00:00
Howard Chu
4298f5d991 Rewrote ber_get_next, fixed byte-order dependency on ber_tag and optimized 2002-01-13 03:00:45 +00:00
Howard Chu
b92e0011dd Prep for in-place parsing. Cache a ber_tag byte.
No visible impact from these changes.
2002-01-06 02:23:36 +00:00
Pierangelo Masarati
1cf91cd6db meaningful message in case max_incomingis hit 2002-01-05 17:16:18 +00:00
Kurt Zeilenga
0e2af54a3f Update copyright statements 2002-01-04 21:17:25 +00:00
Howard Chu
fbc54d47d3 Added ber_free_buf() 2001-12-31 06:01:59 +00:00
Kurt Zeilenga
6cc6a187da Fix an off by one bug which could cause a realloced buffer to be
too small.  Changed extra size algorithm to use a much simplier
approach and detailed non-optimized version.
2001-09-22 03:18:02 +00:00
Kurt Zeilenga
978e417699 Make some additional UTF8 public
Remove lint
2001-07-21 23:13:04 +00:00
Kurt Zeilenga
9046576228 Misc cleanup 2001-05-07 03:43:23 +00:00
Kurt Zeilenga
9275b09954 Add a bit of hardening 2001-05-06 19:07:24 +00:00
Kurt Zeilenga
61d874af10 reimplementation of sb_max_incoming 2001-05-04 21:15:09 +00:00
Kurt Zeilenga
00c645b9a1 ITS#133 2001-05-03 05:51:39 +00:00