in reply to Re: Regexp and Inline::C
in thread Regexp and Inline::C

you can use something like this to perform your task (largely based on code from Inline::C-Cookbook)

It's a bad example in the Cookbook, and thank you for reminding me that it needs to be fixed. Your code (and the Cookbook example) are fine if run just once - but if you call the Inline::C sub more than once, then things can go screwy.

For example, with the code you posted, if I call it 10 times:
c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB'); c_func('AAA:BBB');
I get this output:
AAA - BBB CCC - DDD CCC - DDD
If I call the sub 10 times in a loop:
for(1 .. 10) {c_func('AAA:BBB')}
I get:
Use of uninitialized value $1 in concatenation (.) or string at try.pl + line 10. Use of uninitialized value $2 in concatenation (.) or string at try.pl + line 10. - Use of uninitialized value $1 in concatenation (.) or string at try.pl + line 10. Use of uninitialized value $2 in concatenation (.) or string at try.pl + line 10. - AAA - BBB CCC - DDD Use of uninitialized value $1 in concatenation (.) or string at try.pl + line 10. Use of uninitialized value $2 in concatenation (.) or string at try.pl + line 10. - AAA - BBB CCC - DDD AAA - BBB CCC - DDD CCC - DDD
The problem is that the Inline Stack macros are deficient for performing callbacks - and one should instead base one's callbacks on the perlcall documentation (from where my example was drawn).

Cheers,
Rob