Hello Lauri,

Many thanks for your help.

At 23:01 08/10/30, Lauri Aalto wrote:
>On Thu, Oct 30, 2008 at 6:43 AM, Martin Duerst <[EMAIL PROTECTED]> wrote:

>> When using "make check", all 15 raptor tests pass with PASS,
>> but next all Turtle tests (30+3 in total) FAIL, at which time
>> "make check" gives up.
>
>First check that you have unzipped the test files without any
>end-of-line conversion. The test suite does a binary diff of the
>output. rapper/raptor-produced output has LF line endings and binary
>diffing them fails if the reference output files (*.out) are converted
>to CRLF line endings.
>
>Default cygwin installations come with the "dos2unix" utility which
>you can use to convert CRLF line endings to LF.
>
>If this doesn't help, please provide some "make check" output from 
>tests/turtle.

This helped quite a bit. However, it was the other way round.
The files used by the tests for checking whether the output is
correct or not (.out files) used LF, but the programs produced
CRLF. So I used
$ unix2dos */*/*.out */*/*/*.out
to change the line endings in all .out files to CRLF.

I found out about this by using notepad2 (which has very clear
display of line endings when switched on) and commenting out
some "rm" lines in the Makefile, so that the test output produced
was still available.

Now I get errors for raptor/tests/turtle, test-17 and test-18,
for the rdf diffs (but not for the straight file diffs earlier
in the tests).

Looking at test-17.err and test-18.err, CRLFs in literals seem
to have morfed into CRCRLF :-(. As the straight file diffs
worked okay, this seems to be an issue with rdfdiff rather than
raptor, but that's just a guess. I'm attaching test-17.err and
test-18.err.

Regards,    Martin.



#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp       mailto:[EMAIL PROTECTED]     

Attachment: test-17.err
Description: Binary data

Attachment: test-18.err
Description: Binary data

_______________________________________________
redland-dev mailing list
[email protected]
http://lists.librdf.org/mailman/listinfo/redland-dev

Reply via email to