FW: HURD source code at hurd.gnu.org

[Jim Franklin]

-----Original Message-----
From: address@hidden [mailto:address@hidden Behalf Of
Ian Duggan
Sent: Tuesday, May 08, 2001 12:30 PM
To: web-hurd
Subject: Re: HURD source code at hurd.gnu.org



>  There is not really a capacity for dynamic links at gnu.org at this time.
> We are able to run scripts so are going for a batch type operation. The
more

Right. I'm proposing a possible way of using lxr in batch mode.

1) Checkout from CVS somewhere.
2) Setup a webserver with lxr locally, pointing at your checkout (or
CVS, however lxr does it)
3) "wget -r" your local copy. It will store the files with all of the ?
& + and junk on your file system.
4) rsync the output of "wget -r" to gnu.org.

I just tested it quickly against "wget -e robots=off -r
http://lxr.mozilla.org"; and I was able to pull down the site and surf it
statically using Apache on my local machine. A few quirks:

        DefaultType text/html
        ForceType text/html

were needed for some things to come through right (.cpp files, etc.).

Also, it would be necessary to urldecode the filenames, turning ? into
%3F and = into %3, etc... The same would have to be done for the links
in the pages as well, but the lxr source is available, right?

It would probably be trivial to modify it this way and to adjust the
filenames that "wget -r" spits out likewise.

Are we able to adjust the apache config via .htaccess files on gnu.org?
This would be predicated on that ability, so that we could tune the mime
types appropriately.

Does this seem like too much to pursue, or would people like at least a
staticly browsable lxr archive (no search)?

I'm not sure how big it would be. That would have to be tested.

--Ian

_______________________________________________
Web-hurd mailing list
address@hidden
http://mail.gnu.org/mailman/listinfo/web-hurd