NAME Parse::RPN - Is a minimalist RPN parser/processor (a little like FORTH) SYNOPSIS use Parse::RPN; $result=rpn(string ...); @results=rpn(string ...); string... is a list of RPN operator and value separated by a coma in scalar mode RPN return the result of the calculation (If the stack contain more then one element, you receive a warning and the top value on the stack) in array mode, you receive the content of the stack after evaluation DESCRIPTION RPN receive in entry a scalar of one or more elements coma separated and evaluate as an RPN (Reverse Polish Notation) command. The function split all elements and put in the stack. The operator are case insensitive. The operator are detect as is, if they are alone in the element of the stack. Extra space before or after are allowed (e.g "3,4,ADD" here ADD is an opeartor but it is not the case in "3,4,ADD 1") If element is not part of the predefined operator (dictionary), the element is push as a litteral. If you would like to put a string which is part of the dictionary, put it between quote or double-quote (e.g "3,4,'ADD'" here ADD is a literal and the evaluation reurn ADD and a warning because the stack is not empty) If the string contain a coma, you need also to quote or double-quote the string. (be care to close your quoted or double-quoted string) The evaluation follow the rule of RPN or FORTH or POSTCRIPT or pockect calcutor HP. Look on web for documentation about the use of RPN notation. I use this module in a application where the final user need to create an maintain a configuration file with the possibility to do calculation on variable returned from application. The idea of this module is comming from Math::RPN of Owen DeLong, owen@delong.com that I used for more then a year before some of my customer would like more ... I correct a bug (interversion of > and >=), add the STRING function, pattern search and some STACK functions. OPERATORS The operators get value from the stack and push the result on top In the following explanation, the stack is represented as a pair of brackets () and each elements by a pair of square barcket [] The left part is the state before evalutation and the right part is the state of the stack after evaluation Arithmetic operators --------------------- + | ADD ([a][b]) ([a+b]) ++ | INCR ([a]) ([a+1]) - | SUB ([a][b]) ([a-b]) -- | DECR ([a]) ([a-1]) * | MUL ([a][b]) ([a*b]) / | DIV ([a][b]) ([a/b]) % | MOD ([a][b]) ([a%b]) POW ([a][b]) ([a*a]) SQRT ([a][b]) ([SQRT a]) ABS ([a][b]) ([ABS a]) INT ([a][b]) ([INT a]) & | AND ([a][b]) ([a&b]) | | OR ([a][b]) ([a|b]) XOR ([a][b]) ([a^b]) NOT ([a][b]) ([NOT a]) Logically negate of [a] ~ ([a][b]) ([~ a]) Bitwise complement of [a] Rationnal operators ------------------- SIN ([a]) ([SIN a]) COS ([a]) ([COS a]) TAN ([a]) ([TAN a]) LOG ([a]) ([LOG a]) EXP ([a]) ([EXP a]) Logical operator ---------------- < ([a][b]) ([1]) if [a]<[b] else ([0]) <= ([a][b]) ([1]) if [a]<=[b] else ([0]) = | == ([a][b]) ([1]) if [a]==[b] else ([0]) >= ([a][b]) ([1]) if [a]>=[b] else ([0]) > ([a][b]) ([1]) if [a]>[b] else ([0]) <=> ([a][b]) ([-1]) if [a]>[b],([1]) if [a]<[b], ([0])if [a]==[b] IF ([a][b][c]) ([c]) if [a]==0 else ([b]) Other operator ---------------- MIN ([a][b]) ([a]) if [a]<[b] else ([b]) MAX ([a][b]) ([a]) if [a]>[b] else ([b]) TIME () ([time]) time in ticks RAND () ([rand]) a random numder between 0 and 1 LRAND ([a]) ([rand]) a random numder between 0 and [a] String operators ---------------- EQ ([a][b]) ([1]) if [a] eq [b] else ([0]) NE ([a][b]) ([1]) if [a] ne [b] else ([0]) LT ([a][b]) ([1]) if [a] lt [b] else ([0]) GT ([a][b]) ([1]) if [a] gt [b] else ([0]) LE ([a][b]) ([1]) if [a] le [b] else ([0]) GE ([a][b]) ([1]) if [a] ge [b] else ([0]) CMP ([a][b]) ([-1]) if [a] gt [b],([1]) if [a] lt [b], ([0])if [a] eq [b] LEN | LENGTH ([a]) ([LENGTH a]) CAT ([a][b]) ([ab]) String concatenation REP ([a][b]) ([a x b]) repeat [b] time the motif [a] SUBSTR ([a][b][c]) ([SUBSTR [a], [b], [c]) UC ([a]) ([UC a]) LC ([a]) ([LC a]) UCFIRST ([a]) ([UCFIRST a]) LCFIRST ([a]) ([LCFIRST a]) PAT ([a][b]) ([r1]...) use the pattern [b] on the string [a] and return result if more then one result like $1, $2 ... return all the results PAT ([a][b]) ([r]) use the pattern [b] on the string [a] and return 1 if pattern macth otherwise return 0 Stack operators --------------- DEPTH ([r1]...) ([re1]...[nbr]) Return the number of elements in the statck DUP ([a]) ([a][a]) SWAP ([a][b]) ([b][a]) POP ([a][b]) ([a]) ROT ([a][b][c]) ([b][c][a]) ROLL ([a][b][c][d][e][n]) ([a][c][d][e][b]) rotate the [n] element of the stack (here [n]=4) if [n] =3 it is equivalent to ROT PICK ([a][b][c][d][e][n]) ([a][b][c][d][e][b]) copy element [n] on top PUT ([a][b][c][d][v][n]) ([a][v][b][c][d]) put element [v] at level [n] (here [n]=3) DU |DUMP ([a][b][c][d]) () dump the stack in a string, each elements separated by a blank This avoid the warning if the stack is not empty and prevent use of array DS |DUMPS ([a][b][c][d][sep]) () dump the stack in a string, each elements separated by [sep] (could be \n) This avoid the warning if the stack is not empty and prevent use of array EXAMPLES use Parse::RPN; $test ="3,5,+"; $ret = rpn($test); # $ret = 8 $test = "Hello World,len,3,+"; $ret = rpn($test); # $ret = 14 $test = "'Hello,World',len,3,+"; $ret = rpn($test); # $ret = 14 $test = "'Hello,World,len,3,+"; ---------^-----------^- $ret = rpn($test); # $ret = 8 with a warning because the stack is not empty ([Hello] [8]) # be care to close your quoted string $test = "'Hello,world',',',pat,',',eq,'Contain a coma','Without a coma',if" $ret = rpn($test); # $ret = "Contain a coma" $test = "'Hello world',',',pat,',',eq,'Contain a coma','Without a coma',if" $ret = rpn($test); # $ret = "Without a coma" AUTHOR Fabrice Dulaunoy SEE ALSO perl(1).