Andrey Mazo
e99dedd6c6
Print failed statement in backtrace
...
Update tests accordingly.
2014-05-21 16:31:39 -04:00
Andrey Mazo
bd812090eb
Make load handle absolute paths too
...
Add 2 simple tests on this.
2014-05-21 16:31:22 -04:00
Andrey Mazo
72c49b88e3
Add run variant with stdout and stderr separated
...
The code for storing stdout and stderr separately is taken
from answer [1] by TheConstructor [3] to question [2].
[1] http://stackoverflow.com/a/18086548
[2] http://stackoverflow.com/questions/11027679/bash-store-stdout-and-stderr-in-different-variables
[3] http://stackoverflow.com/users/1266906/theconstructor
2014-05-21 00:16:33 -04:00
Sam Stephenson
ddd03ab852
Remove unused variable
2013-11-04 14:00:58 -06:00
Sam Stephenson
c8d63dd7e0
Correctly log errors in setup and teardown functions
...
Fixes #30
2013-11-04 12:20:55 -06:00
Sam Stephenson
bfa4ebcd0f
Prefer let x+=1 for incrementing counters
...
The `((x++))` syntax is shorthand for `let x++`. According to `help let`:
If the last ARG evaluates to 0, let returns 1; 0 is returned
otherwise.
Thus the exit status of the expression `x=0; let x++` is 1, since the post-increment `++` operator evaluates to the value of the variable before incrementing.
In Bash 4, this non-zero exit status properly triggers `set -e`'s error trap, but in Bash 3 it does not. That's why the tests were passing on OS X (Bash 3) but not Linux (Bash 4).
We can work around the problem by choosing an incrementation expression that never evaluates to 0, such as `+=` or the pre-increment `++` operator. For consistency and clarity, I've changed to `x+=1` everywhere.
Ref. #25 , #27
2013-10-28 21:01:51 -05:00
Sam Stephenson
417acfff66
Revert "Revert "Merge pull request #25 from sstephenson/stack-trace""
...
This reverts commit cb658ba91f
2013-10-28 20:13:45 -05:00
Sam Stephenson
cb658ba91f
Revert "Merge pull request #25 from sstephenson/stack-trace"
...
This reverts commit 07bdee33a1c36ad10d8c
2013-10-28 20:10:42 -05:00
Sam Stephenson
de1970fb8f
Log the full stack trace when a test fails
2013-10-28 15:10:48 -05:00
Sam Stephenson
8930e4e0d1
Fix for test files without trailing newlines ( closes #12 )
2013-10-21 12:45:49 -05:00
Sam Stephenson
8873aab79f
Extended syntax: "begin" line before each test is run
2013-10-21 11:32:09 -05:00
Sam Stephenson
7849374964
bats_test_info -> bats_test_begin; begin calls setup
2013-10-21 11:21:17 -05:00
Sam Stephenson
a773171777
Don't indent the output of failed tests
2013-10-21 10:58:21 -05:00
Sam Stephenson
c8c56a987c
More descriptive message for failing line and file
2013-10-21 10:57:53 -05:00
Ross Duggan
f78324dfec
Support for TAP compliant skip directive inside test blocks.
2013-10-18 22:05:20 +01:00
Sam Stephenson
bc72b85871
Print test file path and line number after a failing test
2013-05-23 12:47:17 -05:00
Adam Jacob
af9794634f
Using shell redirection rather than cat
2013-05-23 00:52:08 -04:00
Adam Jacob
9c8425d005
Supporting DOS line endings in BATS tests
2013-05-22 21:20:39 -04:00
Sam Stephenson
04be7dc717
Abort when load is invoked with a nonexistent script
2012-11-16 14:35:48 -06:00
Sam Stephenson
f8f78b5cd3
bats-exec -> bats-exec-test
2012-04-09 10:59:36 -05:00
