Commit graph

17 commits

Author SHA1 Message Date
Domas Mituzas
2b94d61317 add reportQueryError, halting on any error 2004-08-19 13:00:34 +00:00
Domas Mituzas
8aa6ebc2da Add Database*::timestamp(), using wfTimestamp() 2004-08-10 11:12:18 +00:00
Brion Vibber
62e98ad47c Suppress some more warnings on separate execution of files. Shouldn't
be anything exploitable, it's just annoying.
2004-08-06 22:30:47 +00:00
Brion Vibber
1c7177a65c Fix parse error (missing semicolon). Don't know if this runs currently though. 2004-08-06 21:33:51 +00:00
Tim Starling
1773cd1e76 * Introducing bit field for database parameters
** Database constructor calling sequence, and Database::newFromParams()
** Including flags in the server initialisation structs
** Support for setting appropriate flags from legacy globals in Setup.php
** Moved some defines to Define.php so that they can be used in LocalSettings.php, most importantly the bit field constants

* Changes related to post-parse link colouring
** Turn the link cache back on when using it for updating the links table
** No longer need to call preFill() on page view

* Better synchronisation of slave servers
** Rearranged getConnection()
** System for beginning and committing transactions when multiple connections are open
** wfAbruptExit() commits transactions, wfErrorExit() does not. Various functions changed to use wfErrorAbort()

* Allowed reporting of database errors during deferred updates by moving them above output()
2004-07-24 07:24:04 +00:00
Tim Starling
ac549401d4 * Support for table name prefixes throughout the code. No support yet for converting static SQL, which also means no installation. But it has been tested by creating the tables in the ordinary way and then renaming them
* DB_WRITE now called DB_MASTER, DB_READ now called DB_SLAVE
* Converted to use SQL wrapper functions instead of direct SQL in various places
* Experimental method for preserving the chronological order of events when slave servers are used. Untested.
* Fixes to the new post-parse existence test feature
* Some.. other stuff
2004-07-18 08:48:43 +00:00
Domas Mituzas
d9e1fc0a6d Unbreak the class (some unfinished code inside replace())
Alias DatabasePgsql as DatabasePostgreSQL (according to filename)
2004-07-15 15:09:32 +00:00
Tim Starling
b96b707efa OOP calling convention for database functions. DBMS abstraction implemented by means of functions instead of global variables, PostgreSQL object converted to a subclass instead of a drop-in replacement. Also the beginnings of a flexible table name feature. 2004-07-10 03:09:26 +00:00
Tim Starling
3c4d38faee Removed arbitrary change in timestamp formats between MySQL and PostgreSQL, moved global functions to GlobalFunctions 2004-06-29 12:23:59 +00:00
Domas Mituzas
56589ff499 fix isOpen (mOpened was not defined on connect) 2004-06-21 12:07:41 +00:00
Tim Starling
49749f3091 wfSetRef doesn't work, making isWindows() function which I was going to use somewhere else but decided against it 2004-06-12 06:05:02 +00:00
Domas Mituzas
5f2b3f41e0 * fix affected_rows, uses resultset instead of connection as an argument
* allow wfGetSQL to fetch non-table variables
* InvertTimestamp creates old-style inverse timestamp
* insertArray needed quote-fixing
* added wfLimitResult function, in order to solve SQL LIMIT incompatibilities
2004-06-11 14:32:05 +00:00
Domas Mituzas
83a3aa9f8f added wfFetchRow() besides wfFetchObject. Used by objectcache code 2004-06-10 13:02:27 +00:00
Domas Mituzas
654052ad23 * PostgreSQL does not have buffered/unbuffered modes.
There's no need for that due to row-level locking.

* pg_result_error is for asynchronous operation.
  Maybe in future it could be nice, but now we
  perform only one request at a time.
2004-06-10 06:37:12 +00:00
Domas Mituzas
81138304bd * fixed indexExists
* wfStrencode now uses pg_escape_string instead of addslashes
* timestamp functions designed to use internal PG timestamp
2004-06-09 16:13:46 +00:00
Antoine Musso
99afe8250e typo, looks like objectcache need rewrite cause postgresql doesnt have a FROM_UNIXTIME() function 2004-06-07 11:19:03 +00:00
Antoine Musso
d7eb2f8c1d DO NOT USE ! Simple database layer for postgresql based on existing mysql one. I dont think it even works but I want to share the ongoing work on this. 2004-06-07 02:59:58 +00:00