Please check if the PR fulfills these requirements
[ ] The commit message follows our guidelines
[ ] Docs have been added / updated in README.md (for features)
What kind of change does this PR introduce? (Bug fix, feature, docs update, ...)
This fixes a bug observed with the which command that is used to detect whether commands exist. It is used with the battery status command upower, but that doesn't necessarily exist on all machines. In that case, which upower might write an error message into stderr, which is currently not piped into /dev/null, i.e. it writes the error message to the screen, whenever the prompt is udpated.
What is the current behavior? (You can also link to an open issue here)
which upower writes an error message into stderr, which is then printed on the screen, if the upower command is not installed. This was observed on SL6 virtual machines (see attached screenshot).
What is the new behavior (if this is a feature change)?
Pipe both stderr and stdout into /dev/null to completely suppress output of the which command. This fails silently on machines without the upower command and doesn't produce any output.
Does this PR introduce a breaking change? (What changes might users need to make in their configuration due to this PR?)
No breaking change expected.
eendroroy
commented
6 years ago
Please check if the PR fulfills these requirements
[ ] The commit message follows our guidelines
[ ] Docs have been added / updated in README.md (for features)
What kind of change does this PR introduce? (Bug fix, feature, docs update, ...) This fixes a bug observed with the
whichcommand that is used to detect whether commands exist. It is used with the battery status commandupower, but that doesn't necessarily exist on all machines. In that case,which upowermight write an error message into stderr, which is currently not piped into /dev/null, i.e. it writes the error message to the screen, whenever the prompt is udpated.What is the current behavior? (You can also link to an open issue here)
which upowerwrites an error message into stderr, which is then printed on the screen, if the upower command is not installed. This was observed on SL6 virtual machines (see attached screenshot).What is the new behavior (if this is a feature change)? Pipe both stderr and stdout into /dev/null to completely suppress output of the
whichcommand. This fails silently on machines without the upower command and doesn't produce any output.Does this PR introduce a breaking change? (What changes might users need to make in their configuration due to this PR?) No breaking change expected.
Other information: