After successfully installing hvcc and updating the board I tried to compile a hello world-like patch. I got the error message
./build_pd_hvcc.sh: line 42: hvcc: command not found
Do you know what could be the problem and how to work around it? (on macOS Big Sur)
Thanks a lot!
/Leo

Did you install hvcc on bela or on your computer? Which of the two are you running that command on?

  • Noia replied to this.

    giuliomoro They are installed on the computer. I ran the script from the Bela script folder on the computer.

      So I assume that if you run hvcc in a terminal on your computer you'll get the error "command not found" ?
      That means you didn't actually install hvcc in a standard path

        I’m not very familiar with command line and pip, but I ran:
        pip3 install -e .
        As per the instructions and it returned (after a lot else):
        Successfully installed Jinja2-3.1.4 MarkupSafe-2.1.5 hvcc-0.6.3 importlib_resources-6.4.0 json2daisy-0.4.2 zipp-3.19.0

        Is there anything more I need to do or check?

        • Noia replied to this.

          Noia or should I actually run the scripts from the bela itself?

            pip3 list gives me among others:
            Package Version Editable project location

            hvcc 0.6.3 /Users/noia/Downloads/hvcc

            Trying to run build_pd_hvcc.sh with the full path edited into line 42 gives me a Permission denied error, though I've given full permissions for that file.

            giuliomoro So I assume that if you run hvcc in a terminal on your computer you'll get the error "command not found" ?

            did you try this?

            When I do pipx install -e . (using pipx from brew), at the end I get:

            ⚠️  Note: '/Users/giulio/.local/bin' is not on your PATH environment variable. These apps will not be globally accessible until your PATH is updated. Run `pipx ensurepath` to automatically add it, or manually
    modify your PATH in your shell's config file (i.e. ~/.bashrc).

            I don't know if you got something similar when installing it. Try re-running it with --force to force displaying this warning again. It's not like I recommend using pipx, because it doesn't really work for me as it's missing daisy2json at runtime (same for python -m pip install btw), though a workaround for it is:

            diff --git a/hvcc/__init__.py b/hvcc/__init__.py
index ab1d8c7..0035b52 100644
--- a/hvcc/__init__.py
+++ b/hvcc/__init__.py
@@ -29,7 +29,10 @@ from hvcc.generators.ir2c import ir2c
 from hvcc.generators.ir2c import ir2c_perf
 from hvcc.generators.c2fabric import c2fabric
 from hvcc.generators.c2js import c2js
-from hvcc.generators.c2daisy import c2daisy
+try:
+    from hvcc.generators.c2daisy import c2daisy
+except ImportError:
+    print("Daisy disabled")
 from hvcc.generators.c2dpf import c2dpf
 from hvcc.generators.c2owl import c2owl
 from hvcc.generators.c2pdext import c2pdext

              Thanks a lot!
              This did the trick!
              In pip3 I didn't get that warning, running pipx it was not necessary too --force to get it. I then needed to run
              pipx ensurepath to get it to work in bash.
              (making init.py executable and fiddling around manually in .bashrc , which I tried before getting your answer was not enough)

              In this project I can live comfortably without the Daisy support.

              I think this problem might be unrelated, but since I'm unsure I'll go on here: hvcc and the build_pd_hvcc.sh script now work fine in most cases, but when I try the script on the custom render example it seems to ignore the custom render.cpp file and render the patch with the default render, resulting in a steady sinetone.

              You are right, that was not being picked up. Try today's latest commit from Bela's dev branch, it should work there.

                giuliomoro Thanks a lot for your effort!
                It now confirms using the custom render but exits with:

                ...

                Linking...
bash: line 0: printf: -l: invalid option
printf: usage: printf [-v var] format [arguments]
 ...done
Running /root/Bela/projects/custom-render-heavy-test/custom-render-heavy-test
Error: TODO: a different number of channels for inputs and outputs is not yet supported
Error: unable to initialise audio
Makefile:615: recipe for target 'run' failed
make: *** [run] Error 1
Connection to 192.168.7.2 closed.

                  Could that be related to the multichannel expander?

                  Make sure your custom heavy/render.cpp file is based on the latest scripts/hvresources/render.cpp . The example PureData/custom-render has been updated as well to reflect that

                  • Noia replied to this.

                    giuliomoro Good catch!
                    That was me trying to compile the patch from the master branch with the script from the dev branch. Hopefully I’ll be on track now!