and have boinc_real_escape_string
call it rather than mysql_real_escape_string(),
to ensure that a DB connection exists
(you need one to call mysql_real_escape_string()).
- some additional conversion to BoincDb-based classes
svn path=/trunk/boinc/; revision=13977
Here's how things should eventually be:
- inc/db_conn.inc defines DbConn, a PHP class representing
a connection to a database with
various generic operations (update, insert etc.).
DbConn prepends the DB name to table names,
so you don't have to select a DB;
this means you can access multiple databases
(e.g. BOINC and Bossa) and use persistent connections.
ALL MYSQL-SPECIFIC CODE WILL EVENTUALLY BE HERE.
- inc/boinc_db.inc defines classes BoincUser, BoincTeam etc.,
representing the various BOINC DB tables.
All access to the DB should use these classes.
When you use this framework, no initialization calls are needed.
db.inc is deprecated and should eventually be removed.
I changed get_logged_in_user()
and all team-related code to use this framework.
- user web: show team founder link to "management functions"
even if they're not currently a member;
allow team founder to perform all operations even if not a member.
- turned on E_ALL error checking,
and fixed a bunch of warnings
and a couple of bugs ($team, not team)
- database: update schema/constraints for team fulltext search
- user web: forgot to change Serbia and Montenegro to Serbia
svn path=/trunk/boinc/; revision=13968
representing a connection to a database.
This is independent of Bossa,
and we should rewrite the BOINC PHP code based on this.
svn path=/trunk/boinc/; revision=13967
<hosts_per_page> elements to config.xml;
determine the number of entries to show per page
in top user/team/host lists
svn path=/trunk/boinc/; revision=13964
live in a different database (and/or server) than the BOINC tables.
Also eliminated the need to call a DB init function.
This involved:
1) adding <bossa_db_host> etc. entries to config.xml
2) prefixing DB name to table name in all Bossa queries
(this is needed if Bossa DB is on same server as BOINC DB,
since we're using persistent connections)
- user web: added Montenegro, and changed Serbia and Montenegro to Serbia
(requires DB update to make latter change in existing records)
svn path=/trunk/boinc/; revision=13962
The problem: the various checks were surrounded by things like
#ifdef HAVE__DEV_MOUSE
If the host on which the client is built doesn't have a mouse
(as is the case with our build machine)
then the resulting executable doesn't check /dev/mouse,
even if it exists on the target host.
- removed configure checks for /dev/mouse, /dev/kbd, and /dev/tty1
svn path=/trunk/boinc/; revision=13947
1) a WU is marked as ready for assimilation and has no errors;
2) it has no canonical result
In this case, the assimilate handler gets called anyway,
typically with the canonical result of the previous WU as arg.
Note: this situation doesn't arise normally;
it might happen if some results are deleted accidentally.
The fix:
- identify this situation, and set the WU.error_mask to a new code
(WU_ERROR_NO_CANONICAL_RESULT)
- zero out the "canonical_result" variable passed to the handler,
so even if the handler fails to check wu.error_mask,
at least it won't assimilate the same result twice.
Thanks to Hendrik Verhoek for finding this bug.
- DB schema: team table type is MyISAM, not InnoDB
svn path=/trunk/boinc/; revision=13938
containing '-l', such as '-funroll-loops'. These are not really supposed to be contained in
"curl-config --libs", but apparently this happens at least in Gentoo, as reported here:
http://bugs.gentoo.org/show_bug.cgi?id=139621.
I have applied the patch supplied by Darragh Bailey (of 2006-07-07).
This should fix Gentoo Bugzilla Bug 139621.
svn path=/trunk/boinc/; revision=13919
Rom Walton