Re: [xml] XML regression test cases...



Hi,

On Mon, 2005-10-10 at 23:07 -0700, Jain, Nilesh wrote:
When I run external regression test (./testsuite) I see some errors, is
it common errors or something is not setup correctly on my m/c.

The errors are partly due to incorrect tests, which have not
yet been fixed by the WC3 XML Schema test collection initiative,
partly due to not yet supported features and partly to
incorrect behaviour of Libxml2's schema processor.
The implementation of the schema processor is still work
in progress, so the results will still change quite often
in the near future.

## XML Schemas datatypes test suite from James Clark
Ran 1035 tests, 11 errors, 0 leaks

This tests xmlschemastypes.c via relaxng.c (which uses
a tiny bit of xmlschemas.c)
  --> so the XML Schema data type module
The errors are expected, since XML Schema data types
disagrees here with some of Clark's tests.

## Relax NG test suite from James Clark
Ran 253 tests, no errors
## Relax NG test suite for libxml2
Ran 183 tests, no errors

this tests relaxng.c (xmlschemastypes.c as well?)
--> so the RelaxNG module

## NIST test suite for Schemas version NIST2004-01-14
Ran 23170 tests (3953 schemata), no errors

this tests xmlschemas.c with a strong focus on
xmlschemastypes.c and XML Schema facets
--> so the XML Schema data type module

## Sun test suite for Schemas version Sun2002-01-16
Ran 193 tests (40 schemata), 27 errors (0 internals), 0 leaks
## Microsoft test suite for Schemas version MS2002-01-16
Ran 7230 tests (4686 schemata), 546 errors (1 internals), 0 leaks
Total 32064 tests, 584 errors, 0 leaks

this tests xmlschemas.c and xmlschemastypes.c
--> so the XML Schema structure and data type modules
Here we get the errors I mentioned at the top.

Another question related to internal regression test suite (runtest),
Are they used to confirm the behavior of public interface or What part
of library do they validate? Do you think they are important to be part
of LSB runtime test suite which is used to confirm distribution against
the specification.


Regards,

Kasimier



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]