diff options
| author | Jeff Quast <contact@jeffquast.com> | 2015-09-21 11:43:54 -0700 |
|---|---|---|
| committer | Jeff Quast <contact@jeffquast.com> | 2015-09-21 11:43:54 -0700 |
| commit | 8c97bc616f24da4b006147d3cb5d4b9a5431c964 (patch) | |
| tree | 6067335ef79dedadc1180ed7d43c7f0d99ea8b7d | |
| parent | f5993888e092bd8cecc98ac9558700d4fe8624cf (diff) | |
| download | pexpect-timing-issue-remove.tar.gz | |
Remove 'add sleep before isalive()' recommendationtiming-issue-remove
This is no longer a problem, we've added p.isalive()
assertions to a great majority of our test cases
across a great many platforms of varying speeds that
contradicts this recommendation.
| -rw-r--r-- | doc/commonissues.rst | 14 |
1 files changed, 0 insertions, 14 deletions
diff --git a/doc/commonissues.rst b/doc/commonissues.rst index 26d0e2b..d3aa9d0 100644 --- a/doc/commonissues.rst +++ b/doc/commonissues.rst @@ -49,20 +49,6 @@ off:: child = pexpect.spawn ("ssh user@example.com") child.delaybeforesend = 0 -Timing issue with isalive() ---------------------------- - -Reading the state of :meth:`~pexpect.spawn.isalive` immediately after a child -exits may sometimes return 1. This is a race condition. The child has closed its -file descriptor, but has not yet fully exited before Pexpect's -:meth:`~pexpect.spawn.isalive` executes. Addings a slight delay before the -:meth:`~pexpect.spawn.isalive` call will help. For example:: - - child = pexpect.spawn('ls') - child.expect(pexpect.EOF) - time.sleep(0.1) - print child.isalive() - Truncated output just before child exits ---------------------------------------- |