I see another problem with this as well: VS wouldn't know what all the
packets were, so adding/deleting would require a manual solution file
change.

On 10/28/06, John Hurliman <[EMAIL PROTECTED]> wrote:
Ryan Gahl wrote:
> Hello all. I understand busy is busy... but I have a huge request,
> something that would vastly enable people to more quickly grok working
> with the various packet classes...
>
> Please, let's consider chaning the autogeneration process to break out
> the huge __Packets__.cs file into 1 file per class in a folder called
> "Packets"... This is theoretically not a very hard change to make, and
> I'm sure I'll end up doing it on my own and maintaining my local copy
> that way if the team doesn't agree this is a good idea... but overall
> the community here would benefit because it's not currently possible
> to leisurely scroll down that file to browse all the available classes
> and their members, it's just too big.
>
> John, please give this some thought. I think it would really speed
> things along. Thanks.
>

There is no useful information to be gleaned from the _Packets_.cs file,
it's just a bunch of constructors and serialization/deserialization
functions. If you are looking for protocol documentation, the only
resource we have at the moment is the message_template.msg file (which
still has some Linden comments as well), and eventually we may look at a
way of writing xml documentation for each packet that can be fed in to
mapgenerator and eventually output with the rest of the ndoc generated
html. I'm not ruling it out as a possibility, just a heads up that the
_Packets_ file is the wrong place to be looking for documentation. An
upside of breaking it out in to 500+ separate files is it would be
easier to see what packets have changed by looking through svn logs, a
downside would be the overhead on the compiler of handling all those
separate files. If you do modify mapgenerator to output separate files,
submit the patch to the gna.org page and we can look at some benchmarks
with VS2005, NAnt, and Mono to see what the impact is.

John


_______________________________________________
libsecondlife-dev mailing list
libsecondlife-dev@gna.org
https://mail.gna.org/listinfo/libsecondlife-dev
http://www.libsecondlife.org/



--
--Jesse

_______________________________________________
libsecondlife-dev mailing list
libsecondlife-dev@gna.org
https://mail.gna.org/listinfo/libsecondlife-dev
http://www.libsecondlife.org/

Reply via email to