| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
| |
After adding DESCRIBE support to the SPARQL processor the corresponding
test suite case is now passing, this just updates the test report to
indicate this.
The test report is generated by pytest, so this is really just me
checking in the latest pytest output.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We are still using a somewhat outmoded name for our default branch, this
PR captures the decision to change it as a decision record and prepares
the rest of the repo for this change.
Once merged, the main branch of RDFLib will be renamed using the
procedure documented [here](https://github.com/github/renaming).
Other changes:
- Add area for decision records (this may be merged in another PR before
this one).
- Remove duplicated sections from `docs/upgrade4to5.rst` as they cause
problems when using `sphinx.ext.autosectionlabel` and serve no
purpose.
|
| |
|
|
|
|
|
| |
Honour lax cardinality by only considering the first time a specific
binding occurred in a result when testing result equality.
This was causing a test to fail that required lax cardinality when
comparing.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Enable nitpicky mode for Sphinx and fix all warnings and errors
that occur when running with nitpicky enabled.
Other changes:
- Add a tox environment for building docs (-docs). This is so we can
test building docs on various versions of python as there seems to be
some differences in warnings between different versions. This tox
environment is enabled for linux CI builds.
- Change readthedocs to use python 3.9 as earlier versions do not handle
`@typing.overload` with type aliases.
- Fixes https://github.com/RDFLib/rdflib/issues/1878
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- InfixOWL fixes and cleanup.
[PR #2024](https://github.com/RDFLib/rdflib/pull/2024).
- `rdflib.extras.infixowl.Restriction.__init__` will now raise a `ValueError`
if there is no restriction value instead of an `AssertionError`.
- Fixed numerous issues with
`rdflib.extras.infixowl.Restriction.restrictionKind` which was essentially
not working at all.
- Fixed how `rdflib.extras.infixowl.Property.__repr__` uses
`rdflib.namespace.OWL`.
- Removed `rdflib.extras.infixowl.Infix.__ror__` and
`rdflib.extras.infixowl.Infix.__or__` as they were broken.
- Removed unused `rdflib.extras.infixowl.termDeletionDecorator`.
- Added `rdflib.extras.infixowl.MalformedClassError` which will replace
`rdflib.extras.infixowl.MalformedClass` (which is an exception) in the next
major version.
Final stage of update to InfixOWL, removing unused code, tidying, adding some docu-comments, minor improvements in robustness, updating previously-failing tests, adding more tests. Expanded documentation to follow.
Co-authored-by: Iwan Aucamp <aucampia@gmail.com>
|
| |
|
|
|
|
|
|
|
|
| |
This removes the test suite specific commands as output is now written
by default to the right files.
Other changes:
- Write EARL report to file suffixed with `-HEAD` without timestamps by
default. This is to make it easier to keep the checked in test reports
accurate.
- Remove XFAILs for tests that are now passing.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
This patch reworks the SPARQL test suite to use the new DAWG manifest
processor. The actual checker is also more accurate, correctly reporting
negative tests and more correctly checking the resutls of positive
tests.
There are still some tests from the SPARQL test suite that are not executed,
but checkers for these will be implemented later.
Other changes:
- Fixed `test/data/suites/rdflib/sparql/unicode.srx`. This file did not
have the correct content for the query. I verified this with Apache Jena.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch reworks the RDF/XML test suite to use the new DAWG manifest
processor. It also sync the RDF/XML test data with the upstream source
so we report on the official test files.
However, some tests are commented out
in upstream (i.e. <https://www.w3.org/2013/RDFXMLTests/manifest.ttl>), and
without these tests our coverage decreases. The commented out tests are
for [non-normative behavior relating to XML Literals](https://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-XML-literals).
To keep the coverage the same these non-normative tests were moved to a separate test suite `test/data/suites/w3c/rdf-xml-non-normative/` and are reported on in a separate report file, `test_reports/rdflib_w3c_rdfxml_non_normative-HEAD.ttl`.
Other changes:
- Add RDF/XML EARL reports generated with new test suites.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch reworks the turtle-like test suites (N-Quads, N-Triples,
Turtle, TriG). The changes includes:
- A new DAWG Manifest processor:
- Incorporates pytest expected failure support in test parameter generation.
- Supports URI mapping so that local filesystem URIs can be mapped to the
correct remote base URIs and still work correctly with local files.
- Supports custom entry types so that suite specific handling can be
implemented separately from the core manifest processor. This will
be used for SPARQL test suite and JSON test suite processing.
- Updated EARL reporter:
- Support for the new DAWG Manifest processor.
- Writing of EARL reports by default if a report prefix is
specified in the DAWG Manifest processor.
- Reports output is sorted so that it can be compared with text based
diff tools.
- Fixed reporting of xfail tests. These were reported as skipped before,
now they are reported as failed.
The testing is also more accurate, and now tests which were reported as
passing before are now correctly reported as failing. With one exception
all these failures are related to leniency in the parser, as our parser
accepts input that should raise an error when parsing.
Other changes:
- Moved IRI related test utilities to a separate module.
- Added generated reports for turtle-like formats to git with `-HEAD`
suffix.
Subsequent changes will use this new processor for SPARQL and RDF/XML.
|
| | |
|
| | |
|
| | |
|
| |
|
