[plt-scheme] compiling flickr.plt via PLaneT doesn't create a "compiled" directory

When I try this, I get the message "read-xml: parse-error: expected
root element - received #<eof>" at compile-time (on both OSX and
Linux). I haven't tried to debug it carefully, but it looks like the
flickr.plt package tries do do a bunch of networking/communications
type stuff at compile time and that something in there is failing; I
would strongly suspect that to have something to do with what you're
seeing.

Are you not getting the read-xml error when you run your program?
Also, are you running the program in mzscheme or drscheme?

-jacob

On 10/15/07, Eric Hanchrow <offby1 at blarg.net> wrote:
> I have a trivial program that uses flickr.plt (attached
>
> ); whenever I run it on a particular box, it takes a _long_ time to
> run.  I suspect it's compiling itself every time I run it (instead of
> what I would expect, which is: it compiles itself just the _first_
> time I run it, and thenceforth it's fast, since it can load the
> already-compiled code).
>
> Now, the problem only happens on my *nix box, which is running a
> not-too-old build of mzscheme (371.2, which I built myself) -- but on
> Windows, using a pre-release labeled 371.3, it works fine: it was slow
> the first time, but took only a second or two the second and
> subsequent times.
>
> I also note that on the troublesome *nix box, I do not have a
> directory named
> ~/.plt-scheme/planet/300/371.3/cache/dvanhorn/flickr.plt/1/0/compiled
> whereas on Windows, I do have the equivalent directory.
>
> So my guess is that somehow the compilation process is failing
> silently.  How can I debug this?
> --
> When a cryptographer and a symbologist get together, it
> usually ends in tears.
>         A. O. Scott, reviewing 'The Da Vinci Code' in the New York Times
>
> _________________________________________________
>   For list-related administrative tasks:
>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme
>
>
>