diff options
Diffstat (limited to 'lib/kernel/test')
| -rw-r--r-- | lib/kernel/test/os_SUITE.erl | 17 |
1 files changed, 16 insertions, 1 deletions
diff --git a/lib/kernel/test/os_SUITE.erl b/lib/kernel/test/os_SUITE.erl index 382fd6f6a9..e302bf7378 100644 --- a/lib/kernel/test/os_SUITE.erl +++ b/lib/kernel/test/os_SUITE.erl @@ -29,7 +29,8 @@ suite() -> [{ct_hooks,[ts_install_cth]}]. all() -> [space_in_cwd, quoting, space_in_name, bad_command, - find_executable, unix_comment_in_command, evil]. + find_executable, unix_comment_in_command, deep_list_command, + evil]. groups() -> []. @@ -237,6 +238,20 @@ unix_comment_in_command(Config) when is_list(Config) -> ?line test_server:timetrap_cancel(Dog), ok. +deep_list_command(doc) -> + "Check that a deep list in command works equally on unix and on windows."; +deep_list_command(suite) -> []; +deep_list_command(Config) when is_list(Config) -> + %% As a 'io_lib' module description says: "There is no guarantee that the + %% character lists returned from some of the functions are flat, they can + %% be deep lists." + %% That's why os:cmd/1 can have arguments that are deep lists. + %% It is not a problem for unix, but for windows it is (in R15B02 for ex.). + ?line os:cmd([$e, $c, "ho"]), + %% FYI: [$e, $c, "ho"] =:= io_lib:format("ec~s", ["ho"]) + ok. + + -define(EVIL_PROCS, 100). -define(EVIL_LOOPS, 100). -define(PORT_CREATOR, os_cmd_port_creator). |