Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | - avoid using pblocks for more than one internal operation - the server | Nalin Dahyabhai | 2012-01-11 | 4 | -42/+31 |
| | | | | | | allocates internal state each time but doesn't clean up any that's aready there if you reuse the block - correctly free values we use when constructing compat entries | ||||
* | - don't use the deprecated slapi_dn_parent() function, the result of which ↵ | Nalin Dahyabhai | 2012-01-11 | 1 | -3/+7 |
| | | | | we used to leak | ||||
* | - don't ever need to call slapi_free_search_results_internal() after ↵ | Nalin Dahyabhai | 2012-01-11 | 1 | -2/+0 |
| | | | | slapi_search_internal_callback_pb() | ||||
* | - adapt to slapi_dn_normalize() becoming a no-op because it's deprecated | Nalin Dahyabhai | 2012-01-11 | 1 | -8/+11 |
| | |||||
* | - have slapi_search_internal_get_entry() always set its result to NULL on ↵ | Nalin Dahyabhai | 2012-01-11 | 4 | -6/+5 |
| | | | | failure | ||||
* | - be more careful when logging the names of callback entries | Nalin Dahyabhai | 2012-01-10 | 1 | -6/+12 |
| | |||||
* | - carry the parent TXN ID down | Nalin Dahyabhai | 2012-01-10 | 1 | -1/+13 |
| | |||||
* | - start with a generic sub-pblock function | Nalin Dahyabhai | 2012-01-10 | 2 | -0/+7 |
| | |||||
* | - get the disabling logic right | Nalin Dahyabhai | 2012-01-10 | 2 | -6/+6 |
| | |||||
* | - use an internal sequence number to disable the skip-boring-updates logic, ↵ | Nalin Dahyabhai | 2012-01-10 | 1 | -1/+48 |
| | | | | and test that while our results don't change, we did actually do processing | ||||
* | - add a testing option for toggling the skip-uninteresting-updates logic | Nalin Dahyabhai | 2012-01-10 | 4 | -2/+23 |
| | |||||
* | - add and test a "unique" operator | Nalin Dahyabhai | 2012-01-10 | 1 | -0/+108 |
| | |||||
* | - make refint and memberof tests the default | Nalin Dahyabhai | 2012-01-10 | 1 | -2/+3 |
| | | | | - add an option to add -Wl,-z,nodelete to the link flags | ||||
* | - change how we skip the be-txn-post hook so that it doesn't produce a ↵ | Nalin Dahyabhai | 2012-01-09 | 3 | -3/+21 |
| | | | | | | misleading debug message - set IPV6_V6ONLY to avoid logging an expected EADDRINUSE error | ||||
* | - most of what's needed to hook back-end txn postoperations, except for the ↵ | Nalin Dahyabhai | 2012-01-09 | 7 | -0/+116 |
| | | | | passing the TXN ID around, which means we deadlock if we actually do it | ||||
* | - Make a note of which attributes we read from any entry when evaluating ↵ | Nalin Dahyabhai | 2012-01-06 | 6 | -53/+300 |
| | | | | data, and when we're later called for a modify request which doesn't modify any of those attributes, skip recalculating the entry contents (should make a dent in #771493). | ||||
* | - remove use of 'memberOf' for computing members of the group.byname and | Nalin Dahyabhai | 2012-01-06 | 1 | -3/+3 |
| | | | | | | group.bygid maps; in IPA, the memberOf plugin ensures that it's only ever going to have redundant data, so there's nothing to be gained by examining the relationships from that end | ||||
* | - use the NDN instead of the target DN as the map key | Nalin Dahyabhai | 2012-01-04 | 1 | -3/+3 |
| | |||||
* | - it helps if i don't typo in a way that breaks compilation | Nalin Dahyabhai | 2012-01-04 | 1 | -4/+4 |
| | |||||
* | - use regsubi() rather than regsub() to match {crypt} hashes | Nalin Dahyabhai | 2012-01-04 | 1 | -4/+4 |
| | |||||
* | - add regsubi/regmatchi, which are like the originals but case-insensitive | Nalin Dahyabhai | 2012-01-04 | 1 | -26/+86 |
| | |||||
* | fix a typo in the error message | Nalin Dahyabhai | 2011-12-05 | 1 | -1/+1 |
| | |||||
* | - when building for 389-ds, use Slapi_RWLocks if they appear to be available ↵ | Nalin Dahyabhai | 2011-08-16 | 1 | -1/+22 |
| | | | | (the rest of #730394/#730403) | ||||
* | - set RQ_FILE when asking tcp-wrappers about client authorization | Nalin Dahyabhai | 2011-08-16 | 1 | -0/+2 |
| | |||||
* | - separate which thread creation/manipulation/destruction API we use | Nalin Dahyabhai | 2011-08-12 | 1 | -14/+18 |
| | | | | from which rwlock creation/manipulation/destruction API we use | ||||
* | - try to handle two one-allocation-fails cases (static analysis) | Nalin Dahyabhai | 2011-07-27 | 1 | -0/+8 |
| | |||||
* | - remove format_* functions that didn't need to be exported | Nalin Dahyabhai | 2011-07-26 | 3 | -108/+150 |
| | | | | | - make SDN lists/queues also have a sorted companion which we can use for presence testing | ||||
* | - fix counting, there | Nalin Dahyabhai | 2011-07-26 | 1 | -2/+2 |
| | |||||
* | - don't dup a list and then just free the original | Nalin Dahyabhai | 2011-07-26 | 1 | -18/+11 |
| | |||||
* | - handle control characters in disallowed-chars lists | Nalin Dahyabhai | 2011-07-26 | 2 | -6/+44 |
| | |||||
* | - get more verbose with logging | Nalin Dahyabhai | 2011-07-26 | 2 | -1/+18 |
| | |||||
* | - we already did this check earlier | Nalin Dahyabhai | 2011-05-26 | 1 | -12/+0 |
| | |||||
* | - handle some unuseds | Nalin Dahyabhai | 2011-05-13 | 1 | -6/+3 |
| | |||||
* | - finish the port to the new yp.x | Nalin Dahyabhai | 2011-05-13 | 1 | -32/+34 |
| | |||||
* | - the field is called "maps", not "list" now | Nalin Dahyabhai | 2011-05-13 | 1 | -3/+3 |
| | |||||
* | - new yp.x uses "stat" where we used to use "status"... fun. | Nalin Dahyabhai | 2011-05-13 | 1 | -25/+25 |
| | |||||
* | - don't assume that libc will provide yp headers -- build them ourselves | Nalin Dahyabhai | 2011-05-13 | 9 | -17/+17 |
| | |||||
* | - fix a type error | Nalin Dahyabhai | 2011-04-21 | 1 | -1/+1 |
| | |||||
* | - oh right, we don't ever use the sorted version | Nalin Dahyabhai | 2011-04-21 | 1 | -2/+3 |
| | |||||
* | - when we can keep the list sorted we can search it faster | Nalin Dahyabhai | 2011-04-21 | 2 | -12/+97 |
| | |||||
* | - try to add a man page for nisserver-plugin-defs | Nalin Dahyabhai | 2011-04-19 | 2 | -0/+39 |
| | |||||
* | - speed up an add-to-list operation which did things the obvious way before | Nalin Dahyabhai | 2011-04-19 | 1 | -7/+3 |
| | |||||
* | speed up building compat entries with attributes with thousands of literal ↵slapi-nis-0.23 | Nalin Dahyabhai | 2011-03-31 | 1 | -27/+55 |
| | | | | values (#692690) | ||||
* | - clean up the no-entries warning for the NULL set | Nalin Dahyabhai | 2011-02-09 | 1 | -1/+1 |
| | |||||
* | we don't use syslog, so don't bother including the header | Nalin Dahyabhai | 2011-01-28 | 3 | -3/+0 |
| | |||||
* | - don't leak single_lengths on out-of-memory allocating lengths (defect #10104) | Nalin Dahyabhai | 2011-01-06 | 1 | -1/+2 |
| | |||||
* | - check for errors building a no-such-map or no-such-domain response | Nalin Dahyabhai | 2011-01-03 | 1 | -6/+16 |
| | | | | packet before sending it in response to yp_all (defect #10090) | ||||
* | - drop some dead code (defect #10091) | Nalin Dahyabhai | 2011-01-03 | 1 | -17/+12 |
| | |||||
* | - only memcpy() 1 or more elements (defect #10094) | Nalin Dahyabhai | 2011-01-03 | 1 | -1/+3 |
| | |||||
* | - don't crash if we hit out-of-memory on startup (defect #10096) | Nalin Dahyabhai | 2011-01-03 | 1 | -10/+12 |
| |