DBD::mSQL for DBI - Written by Alligator Descartes <descarte@hermetica.com>
18/07/95: Started.
Version 0.0.0 ( Totally pre-alpha! )
19/07/95:22:00 Code now basically seems to work. I've got connection to
a database, and clean disconnection.
23:45 I'm now working on the statement cycle stuff, which I've
mapped out. It's just a case of coding, which shouldn't
take too long, hopefully.
Posted notice of approaching doom to perldb-interest.
20/07/95:01:25 Fetching appears to work OK now. I need to read the API on
msqlFieldSeek, since I can't work out why fetch is returning
the same row over and over!
21/07/95:09:22 Added a field into imp_sth (currow) which keeps track of the
current row that's been fetched from the result struct. If I
can fix the return codes coming from fetchrow, it'll work!
21/07/95:10:30 Pondered bind variables in cursors. Since there isn't a bind
function in the API, I may have to frig the way that prepare/
execute works, ie, move the call to msqlQuery *after* it's
done some parsing for bind values......Hm.
21/07/95:10:35 Twiddled some bits in the fetchrow cycle.
23/07/95:15:50 Fetch is fetching a row, but it's not returning it to the
calling cycle, so I need to fart about with the sv stuff, which
is not amusing me greatly!
26/07/95:23:42 Decided in agreement with Andreas that the first release is
to be 0.61!
27/07/95:14:14 Finally! It fucking works! After splendid quantities of
hacking around with the prepare/execute/fetchrow/finish cycle
it's actually returning rows correctly!
NOTE: The SV handling within dbd_describe is shot to buggery,
but I've kludged it in mSQL.xs::fetchrow to do a
sv_2mortal instead of a sv_mortalcopy.
27/07/95:14:22 Tidied up mSQL.xs::finish to do an msqlFreeResult. Annoyingly
enough, there's no return code to msqlFR, so I can't test
whether or not it's worked! Bah!
27/07/95:15:15 Altered test to do several bits and bobs. I'm testing INSERT
statements now. These only execute an msqlQuery, but the
stuff needs to continue to pass through the execute and finish
stages cleanly.......*sigh*
27/07/95:15:22 My dubious 'INSERT' check hack works, sort of. Pity it converts
the entire statement to lower case.....!
27/07/95:17:33 Twiddled some more stuff so it seems to do CREATE TABLE,
INSERT statements too. Of course, there's no error checking
yet, but it seems to be OK........Hm.
27/07/95:17:38 Mailed out a statement saying the 0.61 code would be up for
grabs as of Monday 31st July! Still waiting on Alpha reports.
27/07/95:12:11 Fixed the lower-case all the INSERT/CREATE &c. statement bug.
It now checks to see if you're in quotes.
28/07/95:xx:xx Got a report back from Andreas to say compilation barfs on
his Indy. Fixed a load of do_error bugs in dbdimp.c.
13/08/95:18:25 Finally got back to this after a rather long break. Fixed some
Makefile.PL bugs that were kicking about. Finally fixed
( or appeared to fix ) the strlen signedness bug in dbdimp.c
that may halt compilation on the Indy.
Emailed Karsten the debug info to see what's causing problems
on the NeXTStep platform.
14/08/95:13:48 Got email back from Andreas. signedness broke mSQL.xs as well!
Fixed that and emailed him a quick patch.
14/08/95:14:45 Andreas now has a successful compile. The tests are crap, but
appear to ( sort of ) work.
29/08/95:23:18 Converted driver across to new DBI format. It now appears
to compile, install and run fairly reasonably. There are
some serious messes that need cleared up in it, but it's
fundamentally OK, I hope. Announced for pl0 testing.
04/09/95:15:04 Started back on getting the 0.60pl0 out. Tidied up the parsing
of statements for INSERT, CREATE &c statements. It just takes
the first word of the statement now.
04/09/95:15:19 Looks OK. Tested against DBI 0.63 and looks reasonable.
Announced release of 0.60pl1 and put up for ftp.
20/09/95:15:20 Patched a load of stuff including NULLs, and local device
handling ( SV ). Released pl4, which was bogus. Released
pl5 which should fix those issues.
20/09/95:22:17 Fixed overhang of old DBI bug with DBIc_ENDING remaining in
DESTROY in mSQL.xs. Spotted by Marti Rijken and fixed.
( Hopefully! )
18/10/95:15:13 Added in the missing API chunks for ListTables, ListDBs
and ListFields.
ListFields isnae working, but I'll fix that sometime soon....
05/11/95:11:32 Fixed $sth->readblob to return undef. Fixed Makefile.PL,
dbdimp.h, mSQL.h, dbdimp.c for release of 0.60pl7. Also tested
'make test' which now works!
23/11/95:19:22 Fixed ListFields from the pointers given to me by Sven V.
24/11/95:03:13 Fixed error handling in all the func methods
24/11/95:13:01 Added 'NumRows' method to statement handles to return the
number of rows returned ( or affected ) by a statement.o
pre-release distributed to ANDK
25/11/95:09:20 Real release released! No major problems reported.
30/12/95:18:10 Altered mSQL.pm to allow for hostname:port style connections
to known remote port machines. Sets ENV var since the
msqlConnect API call doesn't support port arguments.
30/12/95:18:15 Added 'length' key into the hash returned by ListFields, so
we can get the scale of the field.
24/03/96:22:34 Fixed bugs and upgraded versioning for perl5.002 to work
properly. Released as pl10
-> Lamentable lack of ChangeLog information here!
27/03/97: Tidied up things with regard to 0.61 release.
28/03/97: Patched NULL as undef bug
28/03/97: Patched broken test
-> More lack of ChangeLog information......*sigh*
04/06/97: Did all the outstanding fixes from 0.63 to 0.64. That's me. I'm
done.
08/06/97: Oh no I'm not! Added support for the optional attribute hash in
DBI->connect().
Patched a monster bug that returned the wrong error values upon
DBI->connect() failing, in that, it didn't think it was an error.
Yike.
Added support for $sth->{NAME} as a temporary kludge.
Released 0.65
12/06/97: Added POD documentation for DBD::mSQL's quirks. Applied Gary Shea's
patches for zero-length strings and some test.pl patches.
Released 0.66