Re: [PATCH 08/10] t: forbid piping into 'test_i18ngrep'

On Fri, Jan 26, 2018 at 7:24 PM, Junio C Hamano [off-list ref] wrote:

SZEDER Gábor [off-list ref] writes:

quoted

See two of the previous patches for the only such cases we had in our
test suite.  However, reliably preventing this antipattern is arguably
more important than supporting these cases, which can be worked around
by only minor inconveniences.

I am not sure if that inconveniences will be minor.  Is this too
contrived an example, for example?

  check () {
        pattern=$1 file=$2 script=./runme

        test_i18ngrep "$pattern" "$file" &&
        write_script "$script" &&
        test_expect_success "check $pattern" '
                "$script"
        '
  }

  check foo file <<-EOF
  ... test script comes here ...
  EOF

With 'test_i18ngrep' outside the 'test_expect_success' block!?
Definitely too contrived. :)

OTOH, what about flipping the order of 'test_i18ngrep' and
'write_script'?  If we can't do that, then I wonder what the reason
might be that is not too contrived.

quoted

Signed-off-by: SZEDER Gábor <redacted>
---
 t/test-lib-functions.sh | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/t/test-lib-functions.sh b/t/test-lib-functions.sh
index 92ed02937..e381d50d0 100644
--- a/t/test-lib-functions.sh
+++ b/t/test-lib-functions.sh

@@ -719,6 +719,10 @@ test_i18ncmp () {
 # under GETTEXT_POISON this pretends that the command produced expected
 # results.
 test_i18ngrep () {
+     ( read line ) &&
+     error "bug in the test script: data on test_i18ngrep's stdin;" \
+           "perhaps a git command's output is piped into it?"
+
      if test -n "$GETTEXT_POISON"
      then
          : # pretend success