Date: 08/20/98
- Next message: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Previous message: rasmus: "[PHP-DEV] CVS update: php3/doc/functions"
- In reply to: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Next in thread: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Reply: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Thu, 20 Aug 1998, Rasmus Lerdorf wrote:
>
> > o Keep the current test regression scheme which does *NOT*
> > require httpd running.
> >
> > o Being able to quickly do a `make test' with just the cgi version
> > of PHP3 is very valuable -- after things looks good, one can then move
> > on to making Apache / SSL / etc third party packages/modules.
APIS+> The cipher dev here needs to check if `core' is sound as first
step in Q/A -- thence their need for the current core tests in
php3/test/. Any changes to core made locally are often spotted
with just the first 'make test' *stand-a-lone* php.
Please, either keep the current test/ tree or as first part
of new regression scheme, do a simple non-httpd core verification
and then run the Perl/Regression to test the more complex tests in
a server/browser environment.
I have no complaint/criticism of new work on more realistic
regression test by Rocky or others. Yes, imap/ldap/gd/ssl etc etc
all will benefit from more rigorous and appropriate tests in a
server/browser regression suite. If I was not clear about understanding
that need in the note, I do. I applaud improved regression
suite(s).
I just wanted to assure that a 'core' test standalone *before*
the server/browser section(s) either with the current test/ left
intact in the distribution or with some inclusion of the simplest
test interface for just core issues. The cipher dev like the
current awk-sh-diff etc scheme as it isolates the test driver to
as simple a form as possible. That means a simple, stable test
driver -- so one is *not* also debugging the test driver :)>>
Rocky -- dev here will watch/test/comment on your notions
and implementation with enthusiasm. Thanks for your help on Q/A
certification. Just work from the simplest first test and proceed
in an order to be sure each superior test runs only after the
inferior dependent requirement(s) have passed first. This is the
tricky bit to sort out in any Q/A suite. Thanks
APIS->
>
-NNNN
-- PHP Development Mailing List http://www.php.net/ To unsubscribe send an empty message to php-dev-unsubscribe <email protected> For help: php-dev-help <email protected>
- Next message: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Previous message: rasmus: "[PHP-DEV] CVS update: php3/doc/functions"
- In reply to: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Next in thread: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Reply: Rasmus Lerdorf: "Re: [PHP-DEV] First look at revamping regression tests for php3"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

