Zebra_Database

This release adds support for caching query results using memcache, fixes a bug which caused the script to crash if the object was instantiated more than once and the language method was called for each of the instances, completely rewrites the dlookup method, which was not working correctly if anything other than a comma-separated list of column names was used (like an expression, for example), and lets the "connect" method take an additional argument instructing it to connect to the database right away rather than using a "lazy" connection.

The name of the "get_columns" method was changed to "get_table_columns", as it returned the number of columns in a given table. A new "get_columns" method was added, which takes as argument a resource and returns the number of columns in the given resource. Some clarifications in the documentation were also made.

This released added a "get_link" method, which returns the MySQL link identifier associated with the current connection to the MySQL server. This is necessary because the library uses "lazy connection", so there is no link identifier available when calling the connect method. The insert and insert_bulk methods have a new argument for creating INSERT IGNORE queries which will skip records that would cause a duplicate entry for a primary key. The default value of the "debug" property was set to FALSE.

A bug was fixed in the insert_bulk method. A new method was added, called table_exists, which checks to see if a table with the name given as argument exists in the database. The select method now also accepts limit and order arguments; due to this change, this method is not compatible with previous versions. Some documentation refinements were made.

A bug was fixed where the script would generate a warning if the update method was called with invalid arguments, and this release changes how the insert_bulk method needs to receive arguments, making it easier to use.