ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Conor MacNeill" <co...@ebinteractive.com.au>
Subject RE: How to participate?
Date Tue, 18 Apr 2000 03:57:51 GMT
> -----Original Message-----
> From: Robin Green [mailto:greenrd@hotmail.com]
> >I found the similar library on:
> >http://www.alphaworks.ibm.com/tech/jikesbt
> >May be it will be interesting for you.
>
> Alphaworks software is normally time-limited or evaluation only AFAIK.
>

I wrote something which sounds very similar. I am happy to donate the source
to Apache if it would be useful (there is a lot of it). one thing I used it
for was to 'repackage' a 1.2 class file to work with 1.1 Collection package
names.

> >Yes, but parsing is too slow (1),
>
> I disagree. I think the process will be disc-bound as I said above, so the
> speed of the filesystem interaction will be the key factor, not the speed
of
> the parsing.

Well, I don't think performance is a great issue here anyway. In general,
however, a parser is much harder to develop than a class file analyser,
IMHO.

> Example: A.java contains
>
> if (Constants.DEBUG) logger.println ("done it");
>
> Now if this line is optimised away by the compiler, and then only
> the file
> Constants.java is changed, you will not necessarily be able to tell from
> looking at A.class that A.java depends on Constants, because
> there might be
> no reference to Constants at all in A.class.
>

I take your point about these sort of dependencies which might me optimised
out. Are we sure that such optimisations are allowed? As I read it they are
discouraged in the Language spec

   "In general we suggest that Java implementations use
    late-bound (run-time) code generation and optimization."

Other than these case, however, I think the class file may be a better
indication of true dependencies than the source code.

Cheers
Conor


Mime
View raw message