cpdo  Timeline

Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.

50 most recent check-ins

2016-02-01
15:20
signed-and-unsigned compare fix in the mysql driver. Leaf check-in: [1596ee5be0] user: stephan tags: trunk
15:15
added a kludge to define __func__, as gcc -pedantic got more pedantic. check-in: [2a71ce0090] user: stephan tags: trunk
2013-07-16
17:54
More fixes/cleanups for printf string precision support. check-in: [11c08ac44e] user: stephan tags: trunk
17:13
Ported in string precision formatting support (e.g. %.*s) from whprintf tree. check-in: [0cc983fbb4] user: stephan tags: trunk
2013-06-25
12:52
Updated several sqlite4 bits for upstream hanges. Cannot test them because the sqlite4 trunk is currently broken. check-in: [ca04f65fc1] user: stephan tags: trunk
12:51
removed an out-of-date comment. check-in: [a0b18b6209] user: stephan tags: trunk
2013-06-23
10:27
Updated sqlite4 driver for recent v4 API changes. Leaf check-in: [7f46210784] user: stephan tags: mysql5-param-name-multi
2013-05-13
16:52
Fixes to broken configuration-related macro usages. check-in: [19f9e8d8c6] user: stephan tags: trunk
2013-05-12
17:46
doc updates regarding the "faked param count" fiasco. check-in: [bd29b7e337] user: stephan tags: mysql5-param-name-multi
17:25
mysql driver bound-param-count op now returns its 'older' value (the number bound placeholders) instead of the faked count because we get into trouble when we mix '?' and ':named' parameters this way. check-in: [9d34008624] user: stephan tags: mysql5-param-name-multi
17:06
Trying to improve the the mysql-level error reporting in the step() operation. check-in: [0a73d34096] user: stephan tags: mysql5-param-name-multi
15:40
a few minor formatting changes. fixed an sqlite4-only test code problem. check-in: [7ccbfa9567] user: stephan tags: mysql5-param-name-multi
15:17
Enabled mysql5 blob-as-string feature by default. Renamed CPDO_CAPABILITY_MULTIPLE_STATEMENTS to MULTIPLE_FETCHES and added test code confirming that mysql cannot do it. check-in: [4b3eb7dfbd] user: stephan tags: mysql5-param-name-multi
13:46
Some refactoring of cpdo_bind_val, mainly to fix a potential illegal union access. Side effect is that the get-blob-column op now (for numeric columns) returns pointers to the underlying raw numeric values (instead of a TypeError like before). check-in: [2ab8f52503] user: stephan tags: mysql5-param-name-multi
12:04
fixed: get-blob now works on string fields in mysql driver. Found/marked a potential invalid union member access violation (depending on app-side usage). check-in: [e4145d02dd] user: stephan tags: mysql5-param-name-multi
2013-05-09
08:27
Simplified cpdo_my_calc_real_param_count() a small bit. It's still ugly. check-in: [8a067f5fd0] user: stephan tags: mysql5-param-name-multi
2013-05-07
20:23
++driver version number. Doc additions. check-in: [38c60b3dc3] user: stephan tags: mysql5-param-name-multi
20:08
Cleanups in the multi-param-use code. Fixed a segfault. check-in: [00e3b5989b] user: stephan tags: mysql5-param-name-multi
19:54
bind::param_count() for the mysql driver how returns the same as sqlite3: each unique non-? parameter is counted only once. check-in: [48d5a40672] user: stephan tags: mysql5-param-name-multi
19:48
bind::param_count() for the mysql driver how returns the same as sqlite3: each unique non-? parameter is counted only once. check-in: [d20b4f5c10] user: stephan tags: mysql5-param-name-multi
19:14
Added some experimental code for binding multiple instances of a given named parameter at a time in the mysql5 driver. Due to how that driver manages the named parameters, the param_count() of such statements is currently different between sqlite3 and mysql5. check-in: [516c6a550f] user: stephan tags: mysql5-param-name-multi
2013-05-04
13:26
tiny internal efficiency hack in the sqlite3 bind_xxx() functions. check-in: [90720063ba] user: stephan tags: trunk
13:25
minor doc fix. check-in: [816a125f06] user: stephan tags: trunk
2013-02-12
11:40
Ported in generic buffer class from cson tree to simplify some blob handling. Patched a few C99isms in test code to compile in C89 mode. check-in: [e52a52340b] user: stephan tags: trunk
2013-02-10
13:32
Minor build tweaks. check-in: [fed8435446] user: stephan tags: trunk
2012-07-07
09:33
Okay, so it wasn't a false positive, but a misunderstanding. scan-build is clean. check-in: [24f4128545] user: stephan tags: trunk
09:28
Now compiles cleanly with scan-build except for one false-positive in certify.cpp. check-in: [03dd796995] user: stephan tags: trunk
08:19
minor doc additions. check-in: [1ea3555f00] user: stephan tags: trunk
08:19
Minor error handling cleanup. check-in: [43df6f984f] user: stephan tags: trunk
08:09
Added test file to help track down a memleak in sqlite4. check-in: [b831bb6dd0] user: stephan tags: trunk
08:07
Added beginnings of a capabilities-querying API. About a week ago, just forgot to check it in. check-in: [4b734aaa01] user: stephan tags: trunk
2012-06-30
17:09
merged sqlite4 branch into trunk. check-in: [e3a646f1a3] user: stephan tags: trunk
17:04
Another experimental change for sqlite4_initialize()/shutdown(). Closed-Leaf check-in: [3a901c0135] user: stephan tags: sqlite4
09:29
corrected an API doc change made yesterday. check-in: [5eef39a3e5] user: stephan tags: sqlite4
03:06
pulled in more recent impls of some sqlite3 driver code into the v4 driver (outdated due to accidental forking from older trunk). check-in: [d0f9cbd815] user: stephan tags: sqlite4
02:54
sqlite4 driver now certifies with the caveat that autoincrement/last_row_id do not work. check-in: [253550a21e] user: stephan tags: sqlite4
02:20
minor doc touchups. check-in: [f073e43ebe] user: stephan tags: trunk
01:54
cherrypicked [95469c6f82]. check-in: [7f2a6c8d8b] user: stephan tags: trunk
01:53
updated for SQLITE4_ macro renaming. check-in: [c9efbb4389] user: stephan tags: sqlite4
01:53
removed unused code. check-in: [95469c6f82] user: stephan tags: sqlite4
2012-06-29
15:30
improved auto-registration of drivers a bit. check-in: [19a41dcd41] user: stephan tags: sqlite4
14:54
build fixes for the amalgamation. It now disables sqlite3 if sqlite4 is present, due to symbol collisions between the two (nothing i can fix - that change has to happen at sqlite.org, and DRH is considering the problem). check-in: [43e5fd5c64] user: stephan tags: sqlite4
13:55
added some test data to sql script. check-in: [e4edc049b1] user: stephan tags: trunk
2012-06-28
20:14
removed an inaccurate comment and fixed a typo which could cause multiple definitions of a temp internal variable. check-in: [115475849c] user: stephan tags: sqlite4
19:53
ooops - merged with the proper trunk. check-in: [ae0c5445c3] user: stephan tags: sqlite4
19:48
initial sqlite4 driver. Fails certification due to differences in autoincrement handling, but otherwise seems to work so far. check-in: [6b631153fa] user: stephan tags: sqlite4
19:46
Create new branch named "sqlite4" check-in: [49ec4721d2] user: stephan tags: sqlite4
2012-05-04
14:19
s|enablegetstringblob|enableblobtostring|g check-in: [3983e03490] user: stephan tags: trunk
14:10
added a missing get-string-from-blob block to the mysql5 driver. check-in: [6ca7e013f8] user: stephan tags: trunk
2012-04-21
02:37
Fixed col_get std::string specialization. Gotta stop typing :(. check-in: [b87a95e07c] user: stephan tags: trunk