Lair of The Phoenix
Home Main list of site contents

Blog The all–new, dull as f#ck blog!

News Personal news items and pictures


IMPORTANT UPDATE — 28th. June 2004

Now that my RiscPC is essentially just gathering dust underneath my desk, I can unfortunately no longer properly support the regex module. Stefan Bellon has helpfully agreed to step in and maintain the module, so I would strongly recommend you visit his site now to download the latest & greatest version, which now includes full 32-bit support.

My thanks go to Stefan for taking this on. Feel free to contact me if you have any questions about regex, however. Here follows the original rest of this page, which refers to the last version I produced (3.02):

Here lies my Acorn/RISC OS RegEx module, currently at vn. 1.02. The source is supplied within the archive, in accordance with the the GNU license of the original regex code.

Here are the PGP signatures of the Acorn RegEx Module and original regex tarball.

There’re no functional changes between 1.01 & 1.02; I have merely updated the contact and download details.

1.02 also doesn't have the promised StrongHelp manual; but since I lost it (I called both the module & stronghelp file ‘RegEx’ and wiped the stronghelp file when I copied in the module) I doubt I'll have the time nor inclination to redo it :-(

Steve Drain has stepped into the breach and created StrongHelp manuals for the module, and regex in general; you can find them on his site — thanks, Steve.

Please see the file regexmod/h in the archive for full instructions.

Please take note of the SWIs GetDefaultSyntax and SetDefaultSyntax. These affect the syntax used when a new expression is compiled with a syntax of ‘-1’ ("use default" — remember that if you’re reusing a handle, then a syntax of ‘-1’ means "use previous syntax", not the module’s default.

Also note the RE_CASE_INSENSITIVE syntax flag: simply get the default syntax at start–up, then use that ‘or’d with the new flag as your compilation syntax instead of ‘-1’.

I recommend, for most applications, that the default is used, since it allows the user to set their favourite RegEx syntax at boot–up, and use that syntax in any & all client applications. But you’re free to choose!

Check out the GNU licence in the .tar.gz archive for details; basically, you can use, edit and generally hack–about the thing as much as you want, as long as you always distribute full-source.

As I understand it, you can’t include GNU code in commercial software. But, since this is a module, you can distribute this for free, and then access RegEx SWIs in your commercial software. Well, that’s my story, anyway, and the reason I did all of this!

In the archive is an example BASIC proggy, ‘RegExEg’. It now works, and also gives an example of a (very simplistic!) Extended pattern — just move the REM from one line to the other.