I just lost a couple of hours trying to debug an issue with a dependency that uses napa. I wasn't having an issue until suddenly today I started noticing the files that the folder this dependency was copying didn't contain all the files it was meant to have.
I ended up tracking the problem to an invalid napa cache (perhaps from an interrupted installation?). Since there doesn't seem to be a convenient command line flag to clear the cache I ended up setting up a separate project and using a package.json script to run the napa command on the same github repo of the broken cache and using "napa-config" to set "cache" to false.
This seems to have solved the problem, but perhaps there should be a better way to avoid this problem for future users? Either using a checksum to validate caches or only creating the cache if the process finished running?
I just lost a couple of hours trying to debug an issue with a dependency that uses napa. I wasn't having an issue until suddenly today I started noticing the files that the folder this dependency was copying didn't contain all the files it was meant to have.
I ended up tracking the problem to an invalid napa cache (perhaps from an interrupted installation?). Since there doesn't seem to be a convenient command line flag to clear the cache I ended up setting up a separate project and using a package.json script to run the napa command on the same github repo of the broken cache and using "napa-config" to set "cache" to false.
This seems to have solved the problem, but perhaps there should be a better way to avoid this problem for future users? Either using a checksum to validate caches or only creating the cache if the process finished running?