From Andrea, on a separate thread:
ps (for Frank): BTW a “bibliography-nosort” mode in the test suite
(see sort_BibliographyNosortOption.txt) is confusing. I spent half an
hour in trying to understand why my code was failing just to discover
about this new fancy new mode…
Sorry about that, this is a Zotero-ism. Shall I move this to the
local tests for citeproc-js?
Frank
At the present time citeproc-hs passes 385 tests out of 527. I code
using the test-suite and would like to eventually pass all tests.
Nonetheless I’m not going to implement stuff whose only purpose is to
pass a test that is meaningless for a batch processing engine like
citeproc-hs.
So, even though there is no hurry (I’m going to make a 0.3 release as
soon as possible, so I do not need to pass every test by then), I
think the goal should be to remove from the standard test-suite
everything which is not supposed to be supported by every processor.
Hopefully in some time the list of the tests failed by citeproc-hs
should be approaching the list of tests to be removed, so I shall be
able to come back with a list of specific tests to take a decision
over.
Andrea
From Andrea, on a separate thread:
ps (for Frank): BTW a “bibliography-nosort” mode in the test suite
(see sort_BibliographyNosortOption.txt) is confusing. I spent half an
hour in trying to understand why my code was failing just to discover
about this new fancy new mode…Sorry about that, this is a Zotero-ism. Shall I move this to the
local tests for citeproc-js?At the present time citeproc-hs passes 385 tests out of 527. I code
using the test-suite and would like to eventually pass all tests.
Nonetheless I’m not going to implement stuff whose only purpose is to
pass a test that is meaningless for a batch processing engine like
citeproc-hs.So, even though there is no hurry (I’m going to make a 0.3 release as
soon as possible, so I do not need to pass every test by then), I
think the goal should be to remove from the standard test-suite
everything which is not supposed to be supported by every processor.Hopefully in some time the list of the tests failed by citeproc-hs
should be approaching the list of tests to be removed, so I shall be
able to come back with a list of specific tests to take a decision
over.
The tests should be partitioned into several groups. In an earlier
discussion, I think the idea of three groups was proposed: tests aimed
at batch-processed bibliographies; those aimed at batch-processed
documents with references; and those aimed at interactive systems.
I’m happy to start working on the split anytime, just say the word.
Frank
+1
Not only would it be good to be able to easily run those different
test groups, but it’d be good if the reporting mechanisms can easily
summarize them as such.
Bruce
I’ve been free riding such a huge amount of work you did that I do not
feel quite comfortable to say the word… still I’d be tempted to…
andrea
