Move away from arrays to strings to fix several shellcheck-reported
issues. It isn't useful to expand the found files into arrays, because
only things that are checked are if the array is empty or the contents
of the first array item.
Fix also a shellcheck issue about using a literal string as regexp
match. It appears that the original reason for using a regexp was to
avoid specifying the directory in which the script is run. However, due
to the need of calling 'make generated_files', the directory is fixed
anyway, and the regexp can be left out.
Testing the change can be done with the following script which emulates
the different cases which the script can see. In the output the variable
'X' is the array and 'Z' is the string.
#!/bin/bash
set -o errexit
set -o nounset
set -o pipefail
function find_genfiles() {
find . \
\( \
-not \( \
\( \
-path ./_\* -o \
-path ./.\* \
\) -prune \
\) \
\) -name "$1"
}
# $1 = filename pattern as in "zz_generated.$1.go"
# $2 timestamp file
function newer() {
find_genfiles "$1" | while read -r F; do
if [[ "${F}" -nt "$2" ]]; then
echo "${F}"
fi
done
}
STAMP=stamp
mkdir -p xxx
touch xxx/foobar
touch "${STAMP}"
mkdir -p foo
touch foo/foobar
mkdir -p bar
touch bar/foobar
# two newer files
X=($(newer foobar "${STAMP}"))
if [[ "${#X[*]}" != 0 ]]; then
echo "X1:"
echo " ${X[*]:-(none)}"
fi
Z="$(newer foobar "${STAMP}")"
if [[ -n "$Z" ]]; then
echo "Z1:"
echo " ${Z}" | tr '\n' ' '
echo ""
fi
# no newer files
touch "${STAMP}"
X=($(newer foobar "${STAMP}"))
if [[ "${#X[*]}" != 0 ]]; then
echo "X2:"
echo " ${X[*]:-(none)}"
fi
Z="$(newer foobar "${STAMP}")"
if [[ -n "$Z" ]]; then
echo "Z2:"
echo " ${Z}" | tr '\n' ' '
echo ""
fi
# one newer file, name matches
touch "${STAMP}"
touch bar/foobar
X=($(newer foobar "${STAMP}"))
if [[ "${#X[@]}" != 1 || ! ( "${X[0]}" =~ "bar/foobar" ) ]]; then
echo "X3:"
echo " ${X[*]:-(none)}"
fi
Z="$(newer foobar "${STAMP}")"
if [[ -z "${Z}" || ${Z} != "./bar/foobar" ]]; then
echo "Z3:"
echo " ${Z:-(none)}" | tr '\n' ' '
echo ""
fi
# one newer file, name doesn't match
touch "${STAMP}"
touch foo/foobar
X=($(newer foobar "${STAMP}"))
if [[ "${#X[@]}" != 1 || ! ( "${X[0]}" =~ "bar/foobar" ) ]]; then
echo "X4:"
echo " ${X[*]:-(none)}"
fi
Z="$(newer foobar "${STAMP}")"
if [[ -z "${Z}" || ${Z} != "./bar/foobar" ]]; then
echo "Z4:"
echo " ${Z:-(none)}" | tr '\n' ' '
echo ""
fi
The expected output from running this script:
X1:
./bar/foobar ./foo/foobar
Z1:
./bar/foobar ./foo/foobar
X4:
./foo/foobar
Z4:
./foo/foobar
In 99cdc069e2, we added a check for
whether the directory the user was attempting to test existed or not.
However, adding this check prevented us from running `make test
WHAT=./path/to/pkg/...`, because we believed that `./path/to/pkg/...`
wasn't a valid path to a directory.
Fix this by stripping `/...` before checking whether the directory
exists.
We already do this in hack/verify-generated-files.sh so we should do it
in verify-generated-files-remake.sh as well.
The idea is that any local changes made because of code generation
should not persist beyond the current run of the script.
Change-Id: I7af176773ae16c393dc2b46c006595243c9fa05b
'read' will not handle backslashes properly. 'read -r' is safer to use.
The find_genfiles() will not insert backslashes, so if there are any,
they will be from directory names.
The pattern used in the file is this:
echo " ${X[@]:-(none)}"
What happens is that the array is expanded to separate strings, and it
is checked if that's set (for the default value assignment). However,
the correct way is to check if the concatenated array string is set to
avoid a type mismatch:
echo " ${X[*]:-(none)}"
Tests show that at least bash 4.4.23 behaves the same:
X=(foo bar)
echo " ${X[@]:-(none)}"
echo " ${X[*]:-(none)}"
X=()
echo " ${X[@]:-(none)}"
echo " ${X[*]:-(none)}"
produces:
foo bar
foo bar
(none)
(none)
Both verify-golint.sh and verify-shellcheck.sh have the same logic
which checks failure_file in alphabetical order.
In addition, we'd like to add another script which requires the
same logic. So this add a common function for cleanup.
The placeholder documentation introduces a couple of problems:
- it complicates the contributor-experience (forces the CI to run
N times before the contributor finds out that they need to call an .sh
script and include certain files from docs/)
- it forces CLI related pull requests for tools like kubeadm and kubectl
to require top level approval from docs/OWNERS as such PRs still need
to touch the .generated_docs file
Stop tracking the placeholder documentation by applying the
following actions:
- remove the utility set-placeholder-gen-docs()
- make verify-generated-docs.sh only generate in a temporary folder
and not match .generated_docs
- mark generate-docs.sh as an alias for update-generated-docs.sh
- remove all current placeholder files in docs folders admin, man,
user-guide, yaml
- ignore the above folders and .generated_docs in a .gitignore file
recent changes to kubekins broke local-e2e with the following error
```
E: Unable to locate package sudo
```
Change-Id: I9ad324a2a070bc068ed1f0f88a912eafb191ad90
Kubernetes Prow Robot