Hide Forgot
Description of problem: Version-Release number of selected component (if applicable): file-5.04-6.el6 How reproducible: always Steps to Reproduce: # rpm -ql perl | grep -F .ph | xargs file /usr/lib64/perl5/_h2ph_pre.ph: ASCII text /usr/lib64/perl5/asm-generic/bitsperlong.ph: ASCII text /usr/lib64/perl5/asm-generic/int-l64.ph: ASCII text /usr/lib64/perl5/asm-generic/int-ll64.ph: ASCII text /usr/lib64/perl5/asm-generic/ioctl.ph: ASCII text /usr/lib64/perl5/asm/auxvec.ph: ASCII text /usr/lib64/perl5/asm/bitsperlong.ph: ASCII text /usr/lib64/perl5/asm/cputable.ph: ASCII text /usr/lib64/perl5/asm/elf.ph: ASCII text, with very long lines /usr/lib64/perl5/asm/ioctl.ph: ASCII text /usr/lib64/perl5/asm/ioctls.ph: ASCII text ... Actual results: Expected results:
I'm not sure it will be possible to improve Perl recognition significantly, because Perl doesn't have so unique syntax in comparison with other languages. The only thing I can probably do is to check if the file ends with "1;", because that's what usually perl headers do, but I don't believe it would be accepted by upstream. I'll keep it open for a while and be thinking more about it.
After some discussions with people who know Perl better than me I think there's no way how to significantly improve File Perl patterns and don't break detection of other scripting languages. Closing this as WONTFIX.