BREAKING CHANGE: Make Literal more standards compliant #2460
Conversation
aucampia
force-pushed
the
aucampia/20230609T2355-literal_datatype
branch
from
d051727 to
078a117
Compare
Raise an error when nested graphs occur in TriG. With this change, the <http://www.w3.org/2013/TriGTests/#trig-graph-bad-07> test passes. --------- Co-authored-by: WhiteGobo <richardfechner@posteo.net> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Iwan Aucamp <aucampia@gmail.com>
Skip the following tests that rely on SPARQL 1.0 grammar and are incompatible with the SPARQL 1.1 grammar: - <http://www.w3.org/2001/sw/DataAccess/tests/data-r2/basic/manifest#term-6> - <http://www.w3.org/2001/sw/DataAccess/tests/data-r2/basic/manifest#term-7> --------- Co-authored-by: WhiteGobo <richardfechner@posteo.net> Co-authored-by: Iwan Aucamp <aucampia@gmail.com>
TBD checkpoint aucampia/20230609T2355-literal_datatype: checkpoint 20230817T000443
aucampia
force-pushed
the
aucampia/20230609T2355-literal_datatype
branch
from
078a117 to
376c034
Compare
| if lang is not None or datatype == _RDF_LANGSTRING: | ||
| if lang is None: | ||
| raise ValueError(f"language cannot be None if datatype is {datatype}") | ||
| if not lang: | ||
| raise ValueError("language cannot be empty string") | ||
| if not _lang_tag_regex.match(lang): | ||
| raise ValueError( | ||
| "language tag must match the pattern " | ||
| "^[a-zA-Z]+(?:-[a-zA-Z0-9]+)*$ " | ||
| f"but was {lang}" | ||
| ) | ||
|
|
||
| if datatype is None: | ||
| datatype = _RDF_LANGSTRING | ||
|
|
There was a problem hiding this comment.
Let q=Literal("ex", datatype=something). This will also overwrite any datatype given to a new Literal Literal(q, lang="en").
#prevous:
Literal(q, lang="en")
# >> Literal("ex", lang="en", datatype=something)
# now:
Literal(q, lang="en")
# >> Literal("ex", lang="en", datatype=RDF.langString)
Before the datatype was inherited. But i havent seen this behaviour documented, so there was no expected behaviour before. But we could mention this change in behaviour nonetheless.
Also this will handle issue #2603
| _XSD_PFX = "http://www.w3.org/2001/XMLSchema#" | ||
| _RDF_PFX = "http://www.w3.org/1999/02/22-rdf-syntax-ns#" | ||
|
|
||
| _RDF_LANGSTRING = URIRef(_RDF_PFX + "langString") |
There was a problem hiding this comment.
It might be more useful to make _RDF_LANGSTRING a class attribute for Literal. Then someone can change the default datatype for langstring by using a childclass instead of changing it programwide.
But im not sure
|
@WhiteGobo @aucampia are either of you likely to be able to update / improve this PR? |
|
I'd like to see this merged after the release of v7.1.0, along with this related change #2866 as part of the "lots of breaking changes" release that RDFLib-8 will be. |
be
None.to
rdf:langString.Related issues:
NoneandXSD.string#2123