forrest-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adam R. B. Jack" <>
Subject Re: Fw: Hermes Run (was Re: Missing Documentation)
Date Thu, 06 May 2004 03:08:39 GMT
> Sure we are. It is great that you do and we want to help
> to keep it that way. Gump is an excellent use-case.

Thanks, that is appreciated.

> > This link is on one site, and references others:
> >
> >
> >
> > The Python code generates the XML, then launches forrest
> > (pretty recent versions) against that.
> I gather that you mean that you are using a recent version
> of the SVN head, i.e. experimental.

In some ways yes, in some ways no. Meaning I think a few skin/SVN things
forced us to go with recent, but 99.9% of the content is simple/basic stuff.
Non of the skin stuff has changes in months, and daily nothing really
changes but content values.

> Have you checked the logs? There are memory usage messages.
> You may need to tweak logkit.xconf to get more messages.

Hmm, when  I said those values. Let me see...

I do take build output files, escape and/or encode the data so it fits into
<source> fields .. and, oh yeah, splitting into multiple source files @
approximately 20000 byte boundaries (so no string, when escaped, is > than
the 36K limit). I do that for a lot of file, typically a few hundred bytes
here or there, maybe a K or few ... and this one (when debug was
automatically turned on) a 36Mb one!!!!!! Oh my God, what was I thinking!!!

I now check if the data is significant, and just copy into content/reference
the file instead.

Finger's crossed I've found my 'tweak', sorry for the noise. ;-)

BTW: What log files in particular, what message? Just curious for futures,



View raw message