summaryrefslogtreecommitdiff
path: root/tests
Commit message (Collapse)AuthorAgeFilesLines
* - Added check for CXX compiler to configure.in, to make the tests compileR_2_1_0_Betakwaclaw2012-03-031-1/+2
| | | | - Fixed invalid pointer cast (on 64bit systems) in runtests.c
* Applied patch #3120699 with modifications.kwaclaw2012-02-261-3/+2
|
* Added Windows project files (same VC++ 6.0 format as the others) andkwaclaw2012-02-213-4/+154
| | | | modified minicheck.h to build under Windows.
* Modified Windows project files to import properly into Visual Studio 2010.kwaclaw2012-02-211-4/+4
|
* - Move diff to unifiedhartwork2008-08-031-3/+4
| | | | | | - Hint about variable modification inside script (not through environment) - Integrate forgotten case ibm/not-wf/p28a (only case with lower p) - Tell we expected not well-formed when we do instead of the opposite
* Removed need for amiga_main().ssolie2007-06-032-12/+5
|
* SF bug #1515266: as noted, suspending the parser has the same problem thatfdrake2006-07-011-33/+59
| | | | the abort code has; added a check and regression test or that
* SF bug #1515266: missing check of stopped parser in doContext() for loopfdrake2006-07-011-0/+54
| | | | (added regression test, closing bug report)
* if the test reports a failure message, always display it (it wasfdrake2006-07-011-0/+6
| | | | | ignored); it makes no sense to ever hide this, so we don't worry about the verbosity setting for this
* - suppress compiler warning when the tests are built with C++fdrake2006-07-013-20/+1
| | | | | - remove non-implemented forking API from minicheck; it is not used, and we always use minicheck instead of check
* Modified code to use XML_ParserReset() instead of creating a newkwaclaw2006-04-071-5/+7
| | | | parser instance on each iteration of the main loop.
* Applied patch #1437840. Fix for bugs #1414066 and #1033965 .kwaclaw2006-03-062-1/+8
|
* Added support for XML_LARGE_SIZE.kwaclaw2006-01-011-1/+8
|
* Small corrections.kwaclaw2006-01-011-2/+2
|
* Adjusted to the new return type of XML_GetCurrentLineNumberkwaclaw2005-12-281-11/+24
| | | | and XML_GetCurrentColumnNumber when XML_LARGE_SIZE is defined.
* Updated to reflect that we use our own Check implementation now.kwaclaw2005-12-281-3/+3
|
* *** empty log message ***ssolie2005-12-242-0/+15
|
* make errors reported display more like GCC error messages, so editorsfdrake2005-01-281-2/+2
| | | | like Emacs can provide functions like "go to next error", etc.
* implement verbose mode in minicheck to make it easier to determine whichfdrake2005-01-282-5/+35
| | | | tests failed
* make all C strings ASCII (7-bits) to avoid editors from trying tofdrake2005-01-281-3/+3
| | | | | re-encode the source text incorrectly; we really want the input data to be exactly what it should be
* test using Expat from C++fdrake2005-01-282-0/+7
| | | | (closes SF bug #1006708)
* remove use of the check library; always use our own testing packagefdrake2005-01-271-6/+1
| | | | (this reduces variables and simplifies the build process)
* make the headers for support code more C++ friendlyfdrake2004-09-302-0/+16
|
* convert MSVC project and workspace files to binary files that must alwaysfdrake2004-09-302-132/+132
| | | | | | use the DOS/Windows CR-LF line end convention (closes SF bug #1007088)
* - avoided assumptions that glob patterns before a trailing slash willfdrake2004-09-301-44/+48
| | | | | | | be expanded; this seems to be a problem in some non-bash shells (see SF bug #1033923) - move counter updates to a shared function - modified one of the output messages to be (hopefully) less confusing
* update tests to reflect that a more specific error message is being generatedfdrake2004-07-151-3/+3
| | | | for some cases
* - avoid using the "check" library's suite_free() function; it is notfdrake2004-07-153-10/+4
| | | | | | | available in all versions of the library, and srunner_free() is sufficient (closes SF bug #985192) - avoid some compiler warnings
* If the check library is not available, use an abbreviated implementation offdrake2004-04-304-3/+235
| | | | | the check API. This allows the unit and regression tests to be run on any system without requiring an external package.
* added regression test for SF bug #824420:fdrake2003-11-021-0/+19
| | | | | | | in non-namespace mode, Expat reported the use of an unbound namespace when encountering an element that looks like it has a namespace prefix (were namespaces enabled) and there's a definition for that prefix in the DTD
* Make it possible to avoid the import of expat_config.h. This makes itfdrake2003-10-162-0/+5
| | | | | easier to compile Expat without the autoconf support. This closes SF patch #458907.
* Deal with issue discussed in SF patch #820946: Expat doesn't handlefdrake2003-10-151-25/+25
| | | | | | | | | | | | | | | | | | | | | | the use of modified default calling conventions in client code. To deal with this issue and generally clean up the mass of macros being used to support bits of the machinery, two new macros are being added: - XMLCALL, which expands to whatever is needed to nail down the calling convention for all calls across the library boundary. This must match the convention used for the system's malloc() implementation. - XMLIMPORT, defined to be whatever magic is needed to mark an entry point as imported from a dynamically loaded module (.dll, .so, .sl, whatever). These macros are used to define the XMLPARSEAPI macro already being used to define the API entry points. In addition, XMLCALL is used to define the types of callback functions, and all example code uses this explicitly in both the distributed applications and the documentation.
* Put in a little protection against the developer not having the checkgstein2003-10-092-5/+25
| | | | | | | | | | | package available. sfbug #664541. * configure.in: look for check.h. This will set up a HAVE_CHECK_H define. * tests/chardata.c, tests/runtests.c: update banner. add test for HAVE_CHECK_H and bail if not present.
* Corrected typo.kwaclaw2003-09-051-1/+1
|
* Explains usage of benchmark utility.kwaclaw2003-09-041-0/+16
|
* Small benchmark utility to test pure parser speed.kwaclaw2003-09-043-0/+233
| | | | Tested on Windows only. Includes MS VC++ 6.0 workspace.
* elaborated test case to cover SF bug #774028; current Expat alreadyfdrake2003-08-271-4/+4
| | | | handles this correctly
* - be more specific about the errors we expect to see reportedfdrake2003-03-121-8/+11
| | | | | - change the name of make_basic_suite() to make_suite(), since there's only one make_*_suite() function
* Add regression tests for SF bugs:fdrake2003-03-121-0/+33
| | | | | #692964 - Case where expat 1.95.6 doesn't report duplicate attributes #695401 - Unbound prefixes not rejected
* - two more tests based on Karl's comments on SF patch #673791fdrake2003-01-281-33/+79
| | | | - pass XML_TRUE or XML_FALSE as the isFinal parameter for XML_Parse()
* Added regression tests for SF bug #673791.fdrake2003-01-251-0/+30
|
* Thread the proper file name and line number to make determination offdrake2003-01-251-5/+13
| | | | which test is failing easier.
* - _expect_failure(): Use _fail_unless() instead of fail() to allowfdrake2003-01-241-1/+15
| | | | | | reporting of the right line number - add test that demonstrates the default namespace can be reset without a syntax error
* Add a test that makes sure xmlns:prefix="" is allowed when namespacefdrake2003-01-241-0/+19
| | | | processing is not enabled.
* Fix test case regarding reporting of undefined entities andfdrake2003-01-211-3/+24
| | | | non-validating parsers (see section 5.1 of the XML 1.0 specification).
* Minor cleanups and conformance with current recommended usage.fdrake2003-01-201-8/+8
|
* Oops, there were more regression tests for SF bug #653180; add thesefdrake2002-12-301-0/+106
| | | | as well.
* Added regression tests from bug; Karl already fixed the bug itself.fdrake2002-12-301-13/+27
| | | | | | Closes SF bug #653180. Removed test that failed to tickle the bug, but should have.
* Added regression test for SF bug #483514: Default handler reportsfdrake2002-11-121-8/+93
| | | | handled events.
* Added regression test for SF bug #620106: XML_SetEncoding() brokenfdrake2002-11-121-0/+43
| | | | for external entities.
* Added regression test for SF bug #616863.fdrake2002-11-071-3/+63
| | | | | Cleaned up some indentation. Report the version of the Expat library if not in "silent" mode.
View Lorry Controller Status