Update tests to use triple term <<(·)>> syntax instead of reified triple syntax <<·>>. #165
Conversation
…plel syntax <<·>>.
|
Yes, they are old CG syntax. They shouldn't have triple terms as subjects for RDF 1.2. I think NQ is lagging NT due to "too many things to do". |
|
Yeah, that's what I figured. Going through them and making some fixes as I go. |
|
I think more updates were made to N-Triples Triple Terms. All N-Triples tests should be duplicated as N-Quads tests, and it probably needs N-Quads specific tests (although, that may be covered here). |
| @@ -1 +1 @@ | |||
| << <http://example/s> <http://example/p> <http://example/o> >> <http://example/q> <http://example/z> <http://example/g> . | |||
| <<( <http://example/s> <http://example/p> <http://example/o> )>> <http://example/q> <http://example/z> <http://example/g> . | |||
There was a problem hiding this comment.
The Manifest needs to be updated for this to be a Negative Syntax Test.
| @@ -1 +1 @@ | |||
| << <http://example/s1> <http://example/p1> <http://example/o1> >> <http://example/q> << <http://example/s2> <http://example/p2> <http://example/o2> >> <http://example/g> . | |||
| <<( <http://example/s1> <http://example/p1> <http://example/o1> )>> <http://example/q> <<( <http://example/s2> <http://example/p2> <http://example/o2> )>> <http://example/g> . | |||
There was a problem hiding this comment.
The Manifest needs to be updated for this to be a Negative Syntax Test.
| @@ -1 +1 @@ | |||
| <<<http://example/s1><http://example/p1><http://example/o1>>><http://example/q><<<http://example/s2><http://example/p2><http://example/o2>>><http://example/g>. | |||
| <<(<http://example/s1><http://example/p1><http://example/o1>)>><http://example/q><<(<http://example/s2><http://example/p2><http://example/o2>)>><http://example/g>. | |||
There was a problem hiding this comment.
The Manifest needs to be updated for this to be a Negative Syntax Test.
There was a problem hiding this comment.
Yeah, good catch. I just did a search and replace, and hadn't pulled back far enough to recognize the issue that introduced in these tests.
| <<(<<(<http://example/s1><http://example/p1><http://example/o1>)>><http://example/q1><<(<http://example/s2><http://example/p2><http://example/o2>)>>)>><http://example/q2><<(<<(<http://example/s3><http://example/p3><http://example/o3>)>><http://example/q3><<(<http://example/s4><http://example/p4><http://example/o4>)>>)>><http://example/g>. |
There was a problem hiding this comment.
The Manifest needs to be updated for this to be a Negative Syntax Test.
gkellogg
changed the title
|
The TriG test suite will require similar changes (#161). |
Some tests seemed to be using the old CG syntax, but I believe the current state of specs means that the reified triple syntax tokens
<<and>>should never appear in n-quads data. This PR replaces them with the triple term tokens<<(and)>>.