The reversed equate expression language is a postfix (reverse polish) notation (read cryptic but efficient in terms of space and speed) language. It is rather FORTH like in its appearance. The normal equate expressions are tokenised and converted into this language for processing. However an equate can be defined directly in this langauge if preferred (it is retained for compatibility since it used to be all there was). An understanding of this language can be useful for debugging purposes as this is what any trace output will currently show. However for most users there should be no need to know anything about this (except perhaps if defining equate based filter expressions or patterns which currently require reversed equates) so you can skip to the next part (section Hard Limits).
All operators are single or double character combinations. Five interchangeable data types are supported. The language is stack based but in addition global variables, local variables, and system variables are available. Most operators accept an immediately following variable operand instead of one stack operand. Looping and conditional structures can also be used - as well as a function call mechanism. All in all there are a total of about 34 unique operators, some of which have multiple functions dependent on the data types of their operands.
Due to the large number of operators some have become ambiguous and it is
reccommended that separators are used almost exclusively between them, the
most obvious separator being whitespace. Note that some combinations need
special attention, for example
~~= which has a different meaning from
=~~.
Go to the first, previous, next, last section, table of contents.