Why can I access the contents of the msp folder using file glob in Puredata, but not in Plugdata. No matter how I click, there is no response. I tried versions 0.9.1 and 0.9.2 but all encountered the same problem.

@dreason said:

Why can I access the contents of the msp folder using file glob in Puredata, but not in Plugdata.

The way to test something like this is to fall back to the simplest possible case and look for differences in behavior.

If I create a patch:

[symbol *(
|
[file glob]
|
[print]

-- note here *, not */msp -- since */msp isn't giving any useful data in plugdata, it's not useful for troubleshooting. Making it the most permissive possible search should at least give more information.

Now, if I run this patch before saving, plugdata prints the contents of my user home directory.

Then, I save the patch into a different location and run it again. Now it prints the contents of the location where the patch was saved. oops! It's not! It's the first location that showed up in the plugdata file browser when I hit save. Yeesh. That's weird.

OK, so I open that patch in Pd 0.55.2 with the expectation that it will also print the contents of the patch's location.

Instead, it prints the contents of the home directory.

Now, which one of those is correct? Both of those seem off to me.

In general, if a file path is not written out in full (e.g. /home/xyz/msp/* or c:/Users/xyz/etc/etc), Pd is supposed to resolve it relative to the patch's location, and, failing that, check other locations defined in Paths preferences. In this case, plugdata is following the precedent of other Pd vanilla objects, while Pd is breaking its own precedent by reverting to ~/*. -- [file] help explicitly says symbol * should match "all files/directories in this directory" but it doesn't do that on my machine = bug.

But plugdata's file location isn't consistent (seems to depend on some undefined user actions in the file browser).

So they're both question marks for me.

hjh