Doesn't Mac need WAV or wav?

[marybeckman]

I don't understand this procedure at lines 86-88:

procedure audio_extension
.extension$ = if (macintosh or unix) then ".WAV" else ".wav" endif
endproc

Since Mac and Unix are case sensitive, shouldn't the procedure be one that checks to see which case the extension has and then use it instead?

[tjmahr]

Sure, do that. I don't remember the rationale for that choice; I was refactoring code from the very first version of the NWR script when I made that one-liner.

[marybeckman]

Tristan, I think the rationale for that choice was that my original code that you were refactoring was not only not very well commented but also convoluted to the point of total opacity, so ...

@patrickreidy, can you think of a simpler way to do this than I could?

[tjmahr]

To be fair re rationales, I didn't test on any Unix or Mac systems so I just trusted any platform-specific code in our scripts and hoped it would work.

[patrickreidy]

Mary, I guess that you're suggestion of checking the case of the actual extension and then using that is the simplest solution... But can I ask whether the current L2T-Audio script causing errors due to audio extensions, or is this a modification to make the code clearer?

[marybeckman]

Yes, the current L2T-Audio script causes errors for people running the scripts from a Mac, the way that I do, which I get around by renaming the handful of audio files that have extension .wav to be extension .WAV on the mirror of the tier2:DataAnalysis/.../Recordings directories on my hard drive. Doing that on the UMN server as well could be the cleanest solution?