Kev Jackson wrote:
The problem with checking for win9x is having specialized code just for one platform, but to rewrite so that every OS uses a file to dump out the variables would require changing getProcEnvCommand for all OSs and testing on all OSs (a headache to say the least), although having the same behaviour would probably be easier to maintain in the long run.
Also if you dump out the variables to a file, it'd be nice to remove the file when you've finished with it, both of my suggestions would require some tidying afterwards, so they're both equally bad for that! Is there any evidence that the "dump to file" will actually solve this "hang" before trying to implement it? I can't test this as I don't have any win9x systems available. Does anyone else still have ancient versions of windows for testing purposes?
I dont either, but my linux box does have a copy of vmware, and I do have an MSDN subscription. So in theory, I could build up a copy just for testing. Keep it off the public network, only install the bare minimum of stuff (java, ant, jedit) and I've a lightweight image with minimal security risks.
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
