On Apr 18, 2004, at 2:04 PM, Edi Weitz wrote:
On Sun, 18 Apr 2004 12:27:22 -0400, Randall Randall randall@randallsquared.com wrote:
I'm new to lisp and new to cl-ppcre.
Welcome! I hope you'll have fun.
Thank you!
After building cl-ppcre, every time I start lisp, even if I don't explicitly load cl-ppcre.ppcf,
What's cl-ppcre.ppcf?
The compiled cl-ppcre library, if I understand correctly. I'm on Mac; otherwise it might be cl-ppcre.x86f .
Of course, in production it won't matter, since I'll only very occasionally be restarting, but it's surprising and I wonder if it might be indicative of an actual problem with the build or install or something?
Might be. But before I can say anything I need more details. Which version of CL-PPCRE? Which Lisp? (From the messages above I'd guess it's CMUCL or SBCL but who knows...) Which version of the compiler? Which operating system? How did you obtain and build CL-PPCRE? Manually? Using ASDF-INSTALL? Or did you use tools provided by your OS distribution like apt-get on Debian or the Ports system on FreeBSD?
Okay, I'm on OSX Panther (10.3), on a Powerbook. I'm using CMUCL for OSX, and in spite of various warnings about how it might cook and eat my cat, I haven't had that problem yet.
I downloaded cl-ppcre-0.7.4 and followed the manual installation instructions in the doc/index.html included with it, since at that time I didn't have asdf or the other defsystem thing installed.
After compilation, that file says one can simply concatenate the compiled files together (if one is using the Python compiler (and I'm using CMUCL, remember)), into a single file, cl-ppcre.x86f. Of course, I'm using ppc, not x86, so I assumed I knew what I was doing and changed it.
In any case, I put in my .cmucl-init.lisp this line: (load "/Users/randall/Development/lisp/libraries/cl-ppcre-0.7.4/cl- ppcre.ppcf") which includes the path to that file.
When starting lisp, I then got another compilation which resulted in the 150 or so lines mentioned earlier. This happens every time I start lisp. The library, in a cursory check, appears to work correctly using snippets copied from the documentation. The weird thing, though, is that I now find I don't need the load statement I just mentioned in my .cmucl-init.lisp file to get the laborious compile on startup. That makes me wonder if something's wrong.
I also have an x86 Gentoo machine on which I used the system tools (emerge) to install cl-ppcre, and that went flawlessly, and doesn't do this. It also works correctly so far as I can tell.
Thanks for your help!
-- Randall Randall randall@randallsquared.com 'I say we put up a huge sign next to the Sun that says "You must be at least this big (insert huge red line) to ride this ride".' -- tghdrdeath@hotmail.com