Pike has three indexing operators:

• [ ]
• __-> __
• __.__

  For values other than objects the operators perform the same operation.

  The '[]' operator can index with any pike value, and uses the `[]() lfun in

The '->' operator is for indexing with identifiers, and uses the `->() lfun in objects.

The '.' operator is also for indexing with identifiers, and uses the `[]() lfun in objects. In Pike 7.6 and earlier this indexing operator was only supported for constant values, and the index operation was performed at compile-time. In Pike 7.7 and later the operator is also supported for non constant expressions, in which case the indexing operation will be performed at runtime.

The main place where the difference between '->' and '.' shows up is for directory modules, where '->' indexes the object that implements the module, and '[]' and '.' index the module itself. Huh?

See Also

Finding Functions by Name

Powered by PikeWiki2