* Andrew Shugg (andrew@...) wrote:
> I've been doing a bit of frobbing with LXR.
> While the Enlightenment CVS repository is currently around 95MB I
> decided that only the 'e17/*' stuff would be necessary for the initial
> LXR work. Depending on how easy it is to beat LXR into operating on
> multiple source trees, the eterm/* and misc/* modules may be added at a
> later point (which would increase the load a fair bit). However as each
> of e17, misc and eterm are independent of one another I felt that it
> would not be a problem at this point.
> The e17 module is currently only 25MB of source and as a very pleasant
> surprise my test web server built the LXR index in about 90 seconds and
> the glimpse freetext index in 11 seconds. So (at least initiall) I
> think the system will be able to handle the load. =)
> The main LXR site (indexing about six different Linux kernel versions)
> is a dual Pentium-II, but Enlightenment has a long way to go before it
> reaches the level of cross-referential complexity of the Linux kernel or
> the Mozilla project.
Right.. with suc a small amount of resource usage, do you think you
could simply have it checkout CVS, perform the cross-referencing, and
then upload the new resulting pages? If it becomes a bandwidth problem,
we can discuss moving the computation/uploading elsewhere.
> Term: this test webserver is (unfortunately) not yet available on the
> public 'net, but when it is I shall let you know so you can have a
> fiddle with it.
Sorry it's taken me a few days to get back to you on this, I've been
ordering things by priority the last few days :) (and since there's no
server available, all I can really do is say "cool!" :). But I
appriciate the work. Thanks.