[Maya] Regular Expression

sama.van

I am being lost reading several Google and Wiki tutorial because of the difference of each programming language.

Maya is using common parser but I do not success to do what I would like...

You can give a quick look at the Maya`s documentation here :

- http://download.autodesk.com/us/maya/2009help/index.html


They explain some tips in the "gmatch" and "substitute" definition.


I would like to find an easy way extracting index or object name in the following string :

"|Render|Props_Crate_001.f[5]"



I already have my way to do that but it looks weird in my code and I really would like to have better use of the Regular Expression.



The weird methods are :


- Object name :


--- Method A :

string $temp[] = stringToStringArray ("|Render|Props_Crate_001.f[5]", ".");
string $obj = $temp[0];

--- Method B :

listRelatives -p `listRelatives -p |Render|Props_Crate_001.f[5]`




- Component index :


--- Method A :

string $tmp[];
int $numTokens = `tokenize "|Render|Props_Crate_001.f[5]" "[]" $tmp`;
int $index = $tmp[ $numTokens - 1 ];

--- Method B :

string $temp = `substitute "^.*\\[" "|Render|Props_Crate_001.f[5]" ""`;
$temp = `substitute "\\].*" $temp ""`;




Is there any challenger? ^_^

MoP

I always use tokenize for extracting stuff like the path name, namespace, or component index.

I only use regular expressions for matching things in string comparisons.

For the object name in your example you would tokenize by "|", take the last value from the buffer, then tokenize by "." and take the first value from the buffer.

For the index you just tokenize by "[]" and take the last value from the buffer as you said.

You could also write a little helper procedure for that so you could return the value:
int $vertIndex = getComponentIndex( "|group1|pCube1.f[4]" );

sama.van

Oh yeah the tokenize is in my "weird method".

And the "getComponentIndex" could be a solution...

My workflow currently works with a "Common" library.
For example it means I have a Polygon.mel file and I insert all very basic function I create. I already have several files like that...

then for the "getComponentIndex" I shouldn't not create common files but a sub directory called "Polygon" (for the example) and then each function in a separate *.mel.
And then if the files name match with the function name, the call is direct and I do not need to source...


Damn. I have for one year to convert my current system to the new one
But very interesting I will do some test.


But about the Regular Expression it was more for understanding advance use...
And to try using in a more complex situtation!

Look the following one :


string $temp = `substitute "^.*\\[" "|Render|Props_Crate_001.f[5]" ""`;
$temp = `substitute "\\].*" $temp ""`;


Don't you know a way to insert the second expression in the first one? is there now flag for "AND" ???

sama.van

shit, the call of the procedure is not automatic if I drop it in a sub directory ...
It only work to the root of the "maya\2009-x64\scripts\" directory...

claydough

Sorry code tags are hanging poly count again :poly127:

below are some of my matching notes... ( Because I forget what works for me, all the time :whyme: )

basically,
for selection: pCube1.f[27]

//method 1 using ".[]"

".[]"
are all illegal characters..
Therefore, they only need to be accounted for once in any match.
Tokenize then seperates the returns in the order:

[0] object
[1] component type
[2] index address

pCube1
f
27


// method 2 using "\\[.*\\]"

Match and tokenize produce different returns using the same search expression.
Returning the component address with brackets can be just as important/useful.
In which case I use match.

using the expression "\\[.*\\]"

tokenize [2]:
27

vs

match:
[27]

refer to the print results below:





// begin expression notes
{
string $headSelection[] =`ls -sl -head 1`;
string $tokenized[];

// method 1
tokenize $headSelection[0] ".[]" $tokenized;

print ( "\n"+ $tokenized[0] + " <---- ( the 1rst array member ) the object\n" );
print ( $tokenized[1] + " <---- ( the 2nd array member ) the component type\n" );
print ( $tokenized[2] + " <---- ( the 3rd array member ) the component index\n" );
print " wow, illegal character presumptions makes string seperation easy!\n";

// The result:
/*
pCube1 <---- ( the 1rst array member ) the object
f <---- ( the 2nd array member ) the component type
27 <---- ( the 3rd array member ) the component index
wow, illegal character presumptions makes string seperation easy!
*/


// method 2
tokenize $headSelection[0] "[URL="file://\\"]\\[/URL][.*\\]" $tokenizedIsDifferent; // "
$matchInstead = ` match "[URL="file://\\"]\\[/URL][.*\\]" $headSelection[0]`;

print " \nThe following Tokenized [2] search expression resulted in: \n\n";
print $tokenizedIsDifferent[2];
print " \nWhich will not be equal to the results gained from Match using same exact search expression...\n";
print "Match's result: \n\n";
print $matchInstead;
print "\n\nHowever, returning the index in brackets can be just as important/useful when replacing strings.\n";

// The result:
/*
The following Tokenized [2] search expression resulted in:

27

Which will not be equal to the results gained from Match using same exact search expression...
Match's result:

[27]

However, returning the index in brackets can be just as important/useful when replacing strings.
*/
}
// end notes

Shawner

Have you considered using Python. It's really is excellent for this type of thing. I forced myself to start using python about 2 years ago and have not regretted it.

mainStr = "|Render|Props_Crate_001.f[5]"

#split the string into a array/list using '.' as the argument
splitMe = mainStr.split('.')
print splitMe

#splice out the 'f['  and ']'
index = splitMe[1][2:-1]
print index

#now to do this all in one line
finalVtx = mainStr.split('.')[1][2:-1]
print finalVtx

#To get the name
name = mainStr.split('.')[0]
print 'name: %snface: %s' %(name, finalVtx)