Hi fellow perl developers,
I'm just working on my first module with probable public interest and would
like to register as a developer (other than being perl porter for the BS2000
operating system ;-).
My name is
Thomas Dorner
My (company) email address is
[EMAIL PROTECTED]
My (private) homepage http://www.th-dorner.de doesn't have anything to with
computers at all and is in german only, so maybe we better skip it.
My preferred user ID on CPAN would be DORNER, followed by THDORNER or
TDORNER.
The module I'm working on is a fairly simple one:
It will cache filenames and directories for multiple file searches in an
object, so that a script only has to browse the filesystem once. To achive
this, it mainly uses a hash containing filenames and their corresponding
directories.
I need this module for a script analyzing a big tree of source modules in a
proprietary programing language. The include files are scattered within
many directories, but the include statement in the modules doesn't contain
these directory informations - I have to search the whole tree for each
module. I haven't found anything like it on CPAN, so I started my own. It
should notably decrease the time spend looking for the include files.
I would suggest
File::Locate
as the name for this module.
An example usage would be (using a known directory tree for example
reasons):
$includes = File::Locate->new(Path => ['/usr/include',
'/usr/local/include'],
Filter => '\.h$');
@time_h = $includes->findInPath('time.h');
$time_h = $includes->findBestInPath('time.h', \&nifty_evaluation_function);
@std_h = $includes->findMatch('^(?i:std)');
findInPath in above example would return a list at least containing
('/usr/include/time.h', '/usr/include/sys/time.h').
Any comments / objections?
Yours sincerely, Thomas Dorner
________________________________________
Thomas Dorner
Start Amadeus, Technology-Strategy (I-T)
E-Mail: [EMAIL PROTECTED]
