Jeremy Huntwork wrote:
> Hey Guys,
>
> As I mentioned in a reply to one of Richard's posts, I've been working
> on a way to get rid of the xsltproc dependency with jhalfs. This may or
> may not be useful - time will tell. One of the main reasons for my
> attempting this was to beef up my C skills. I've learned a lot in the
> past week or so. :)
>
> Anyway, I'm working on a C parser designed specifically to parse the
> LFS book for it's commands. I'm not quite ready to share the code yet,
> but I'll share the results so far.
>
> Right now it just prints out it's progress and what it has found:
>
> $ time { ./parser ../lfs-trunk/index.xml > parser.log; }
>
> real 0m0.782s
> user 0m0.010s
> sys 0m0.020s
>
> And here's the parser.log:
>
> http://linuxfromscratch.org/~jhuntwork/parser.log
Of course, its your project and your time, but I don't see the point. I
did a quick check by stripping out everything except the xsltproc
command and got:
$ ./jhalfs.test
Extracting commands...
real 0m13.999s
user 0m13.612s
sys 0m0.353s
Writing a program in C to save 13 seconds out of approximately 4 hours
build time seems a bit out of place to me.
Besides that, the .xsl is about 200 lines. That will be a lot easier to
maintain than a C program that will be considerably longer. The xsl
also does not have to be built with a separate compile.
The purpose of xslt is to convert documents from one form to another;
in this case from xml to a Makefile. Using a general purpose tool to do
what a specialized tool already does seems like a less than optimal use
of resources.
-- Bruce
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
