summaryrefslogtreecommitdiff
path: root/docs/databases
diff options
context:
space:
mode:
authorPierre Schmitz <pierre@archlinux.de>2011-06-22 11:28:20 +0200
committerPierre Schmitz <pierre@archlinux.de>2011-06-22 11:28:20 +0200
commit9db190c7e736ec8d063187d4241b59feaf7dc2d1 (patch)
tree46d1a0dee7febef5c2d57a9f7b972be16a163b3d /docs/databases
parent78677c7bbdcc9739f6c10c75935898a20e1acd9e (diff)
update to MediaWiki 1.17.0
Diffstat (limited to 'docs/databases')
-rw-r--r--docs/databases/ibm_db2.txt3
-rw-r--r--docs/databases/postgres.txt100
-rw-r--r--docs/databases/sqlite.txt12
3 files changed, 115 insertions, 0 deletions
diff --git a/docs/databases/ibm_db2.txt b/docs/databases/ibm_db2.txt
new file mode 100644
index 00000000..3c3f381c
--- /dev/null
+++ b/docs/databases/ibm_db2.txt
@@ -0,0 +1,3 @@
+== See also ==
+*[http://www.mediawiki.org/wiki/Manual:IBM_DB2 Installation instructions]
+*[http://ca.php.net/manual/en/function.db2-connect.php PHP Manual for DB2 functions] \ No newline at end of file
diff --git a/docs/databases/postgres.txt b/docs/databases/postgres.txt
new file mode 100644
index 00000000..cec51861
--- /dev/null
+++ b/docs/databases/postgres.txt
@@ -0,0 +1,100 @@
+This document describes the state of Postgres support in MediaWiki.
+
+
+== Overview ==
+
+Support for PostgreSQL has been available since version 1.7
+of MediaWiki, and is fairly well maintained. The main code
+is very well integrated, while extensions are very hit and miss.
+Still, it is probably the most supported database after MySQL.
+Much of the work in making MediaWiki database-agnostic came
+about through the work of creating Postgres support.
+
+
+== Required versions ==
+
+The current minimum version of PostgreSQL for MediaWiki is 8.1.
+It is expected that this will be raised to 8.3 at some point,
+as 8.1 and 8.2 are nearing end of life.
+
+
+
+== Database schema ==
+
+Postgres has its own schema file at maintenance/postgres/tables.sql.
+
+The goal is to keep this file as close as possible to the canonical
+schema at maintenance/tables.sql, but without copying over
+all the usage comments. General notes on the conversion:
+
+* The use of a true TIMESTAMP rather than the text string that
+MySQL uses is highly encouraged. There are still places in the
+code (especially extensions) which make assumptions about the
+textual nature of timestamp fields, but these can almost always
+be programmed around.
+
+* Although Postgres has a true BOOLEAN type, boolean columns
+are always mapped to SMALLINT, as the code does not always treat
+the column as a boolean (which is limited to accepting true,
+false, 0, 1, t, or f)
+
+* The default data type for all VARCHAR, CHAR, and VARBINARY
+columns should simply be TEXT. The only exception is
+when VARBINARY is used to store true binary data, such as
+the math_inputhash column, in which case BYTEA should be used.
+
+* All integer variants should generally be mapped to INTEGER.
+There is small-to-no advantage in using SMALLINT versus
+INTEGER in Postgres, and the possibility of running out of
+room outweighs such concerns. The columns that are BIGINT
+in other schemas should be INTEGER as well, as none of them
+so far are even remotely likely to reach the 32 billion
+limit of an INTEGER.
+
+* Blobs (blob, tinyblog, mediumblob) should be mapped to TEXT
+whenever possible, and to BYTEA if they are known to contain
+binary data.
+
+* All length modifiers on data types should be removed. If
+they are on an INTEGER, it's probably an error, and if on
+any text-based field, simply using TEXT is preferred.
+
+* Sequences should be explicitly named rather than using
+SERIAL, as the code can depend on having a specific name.
+
+* Foreign keys should be used when possible. This makes things
+both easier and harder in the code, but most of the major
+problems have now been overcome. Always add an explicit ON DELETE
+clause, and consider carefully what choice to use (all things
+considered, prefer CASCADE).
+
+* The use of CIDR should be done very carefully, because the code
+will sometimes want to store things such as an empty string or
+other non-IP value in the column. When in doubt, use TEXT.
+
+* Indexes should be created using the original MySQL tables.sql
+as a guide, but keeping in mind the ability of Postgres to use
+partial indexes, functional indexes, and bitmaps. The index names
+should be logical but are not too important, as they are never
+referenced directly by the code (unlike sequence names). Most of
+the indexes in the file as of this writing are there due to production
+testing of expensive queries on a busy wiki.
+
+
+== Keeping in sync with tables.sql ==
+
+The script maintenance/postgres/compare_schemas.pl should be
+periodically run. It will parse both "tables.sql" files and
+produce any differences found. Such differences should be fixed
+or exceptions specifically carved out by editing the script
+itself. This script has also been very useful in finding problems
+in maintenance/tables.sql itself, as it is very strict in the
+format it expects things to be in. :)
+
+
+== Getting help ==
+
+In addition to the normal venues (MediaWiki mailing lists
+and IRC channels), the #postgresql channel on irc.freenode.net
+is a friendly and expert resource if you should encounter a
+problem with your Postgres-enabled MediaWiki.
diff --git a/docs/databases/sqlite.txt b/docs/databases/sqlite.txt
new file mode 100644
index 00000000..b8a45553
--- /dev/null
+++ b/docs/databases/sqlite.txt
@@ -0,0 +1,12 @@
+SQLite shares the MySQL schema file at maintenance/tables.sql, with a set of
+compatibility regexes to convert MySQL syntax to SQLite syntax:
+
+* BINARY() and VARBINARY() fields are converted to BLOB
+* the UNSIGNED modifier is removed
+* "INT" fields are converted to "INTEGER"
+* ENUM is converted to BLOB
+* the BINARY collation modifier is removed
+* AUTO_INCREMENT is converted to AUTOINCREMENT
+* Any table options are removed
+* Truncated indexes are upgraded to full-width indexes
+* FULLTEXT indexes are converted to ordinary indexes