On Mon, 18 Jun 2001, Doug MacEachern wrote:

> On Mon, 18 Jun 2001, Stas Bekman wrote:
>
> > [Mon Jun 18 01:23:25 2001] [error] [client 127.0.0.1] input filters not
> > yet supported at /home/stas/perl-bleed/lib/5.7.1/Test.pm line 128,
> > <fh00003b> line 1.
>
> i haven't seen this first-hand, but one of my co-workers (sax) has, i've
> been meaning to look at it with his setup.

I've done some more investigation: if I specify all the tests on the
command line via "--run-tests protocol modules apr hooks...", the tests
pass. However if I specify none (like 'make test' does)  half of the tests
fail with the above error. That's the only difference I can see.

Now the reason you couldn't reproduce the failure is because the error
occurs when the tests are executed in a particular order. Since your
files are organized differently on your file system, the order they are
being read by Apache::TestHarness::run (when tests aren't specified at the
command line) is different from mine. And again, just like with the
repeating of the same tests that causes problems, changing the order of
the tests causes problems too.

Now here is at least one combination of tests that fails:
./t/TEST --run-tests=1 `cat badlist`

The same list of tests in a different order passes the tests:
./t/TEST --run-tests=1 `cat goodlist`

both files are attached.

that was a tricky one to catch especially since originally I've tried to
use Data::Dumper to dump the list in Apache::TestHarness and was getting a
different result on every re-run. Now I got it right.

This non-reproducible scenario makes me think about a need to add yet
another option that will randomize the order of tests as they are run.
This shouldn't be the default for users since it'll confuse them if they
see a different order in tests if they rerun the 'make test' stage, but we
definitely need this for developers to run white-box testing (if I
remember correctly the term from quality assurance terminology).

_____________________________________________________________________
Stas Bekman              JAm_pH     --   Just Another mod_perl Hacker
http://stason.org/       mod_perl Guide  http://perl.apache.org/guide
mailto:[EMAIL PROTECTED]   http://apachetoday.com http://eXtropia.com/
http://singlesheaven.com http://perl.apache.org http://perlmonth.com/

protocol/echo.t \
modules/cgi.t \
modules/cgiupload.t \
hooks/access.t \
hooks/authen.t \
hooks/authz.t \
hooks/fixup.t \
hooks/trans.t \
filter/api.t \
filter/buckets.t \
filter/input_body.t \
filter/input_msg.t \
filter/lc.t \
filter/reverse.t \
apr/base64.t \
apr/constants.t \
apr/lib.t \
apr/netlib.t \
apr/pool.t \
apr/uuid.t \
api/access.t \
api/aplog.t \
api/conn_rec.t \
api/lookup_uri.t \
api/module.t \
api/r_subclass.t \
api/request_rec.t \
api/response.t \
api/rutil.t \
api/send_fd.t \
api/sendfile.t \
api/server_rec.t \
api/uri.t \
apache/cgihandler.t \
apache/compat.t \
apache/conftree.t \
apache/constants.t \
apache/post.t \
apache/read.t \
apache/scanhdrs.t \
apache/write.t
protocol/echo.t \
modules/cgiupload.t \
modules/cgi.t \
hooks/access.t \
hooks/authen.t \
hooks/authz.t \
hooks/trans.t \
hooks/fixup.t \
filter/input_body.t \
filter/input_msg.t \
filter/lc.t \
filter/buckets.t \
filter/api.t \
filter/reverse.t \
apr/base64.t \
apr/constants.t \
apr/lib.t \
apr/netlib.t \
apr/pool.t \
apr/uuid.t \
api/access.t \
api/send_fd.t \
api/sendfile.t \
api/aplog.t \
api/conn_rec.t \
api/lookup_uri.t \
api/module.t \
api/r_subclass.t \
api/request_rec.t \
api/response.t \
api/rutil.t \
api/server_rec.t \
api/uri.t \
apache/conftree.t \
apache/cgihandler.t \
apache/compat.t \
apache/constants.t \
apache/post.t \
apache/read.t \
apache/scanhdrs.t \
apache/write.t
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to