aboutsummaryrefslogtreecommitdiffstats
path: root/test
AgeCommit message (Collapse)AuthorFilesLines
2024-01-09Avoid regressions in clang nullability checksDavid Robillard2-4/+8
Clang 15 (and still as of 16) lost the ability to understand null checks in conditionals, which is supposed to suppress these warnings. For now, work around some, and suppress others. The suppression boilerplate here is noisy and ugly, and hopefully temporary. It should be removed once the issue is fixed in clang. See https://github.com/llvm/llvm-project/issues/63018
2023-10-06Tolerate reduce floating point precision on Apple SiliconDavid Robillard1-2/+2
This is... strange, and slightly worrying, although it's already known that this conversion code isn't exact in all cases. The next major version will fix that (and I've confirmed that these issues aren't present there), but that's a major change so, for now, just reduce the very large number in the test case slightly so that it passes on these chips as well.
2023-09-23Avoid use of deprecated meson string formatting with filesDavid Robillard1-10/+29
Meson annoyingly doesn't provide access to the path of file objects (but it should). It was possible to do that by formatting a string with a file argument, but now this triggers a loud warning that claims it has always been broken and will be deprecated. So, avoid this by juggling things around and maintaining unchecked path strings instead, and hope that they're all passed to files() at some point so they can't get stale.
2023-09-22Clean up meson formattingDavid Robillard1-35/+73
Formatted automatically with muon, with some manual changes to make things format more nicely.
2023-05-03Make serd_reader_read_chunk() work with NQuadsDavid Robillard1-7/+101
2023-05-03Remove test output file after executionDavid Robillard1-0/+1
2023-04-30Suppress cross-compilation warnings in MacOS buildsDavid Robillard1-0/+5
2023-04-30Add long "help" and "version" options to serdiDavid Robillard1-0/+2
Although serdi still "officially" has a BSD-style interface, meson uses --version to find the version of executables. So, support that, and also add --help, since users will often try that for unknown commands.
2023-04-16Gracefully handle errors when writing textDavid Robillard3-1/+164
2023-04-06Add test for different comment termination casesDavid Robillard3-0/+13
2023-04-06Improve pretty-printing of lists and inline subjectsDavid Robillard13-10/+87
2023-04-06Remove support for writing Turtle named inline nodes extensionDavid Robillard2-18/+0
2023-04-06Make URI writing stricter by defaultDavid Robillard2-1/+39
2023-04-06Gracefully handle boolean subject and predicate errorsDavid Robillard6-2/+17
2023-04-06Check formatting of project Turtle filesDavid Robillard1-0/+11
2023-04-06Use portable names for all test targetsDavid Robillard1-8/+8
2023-04-06Simplify stdout capturing codeDavid Robillard1-4/+2
This is slightly less excruciatingly slow on Windows, sometimes, maybe.
2023-04-06Add pylint testDavid Robillard1-1/+25
2023-04-06Fix flake8 warnings and add testDavid Robillard2-1/+7
2023-04-06Simplify command-line option test namesDavid Robillard1-2/+2
2023-04-06Simplify naming of W3C test directoriesDavid Robillard1025-8/+9
This matches the names used in <https://github.com/w3c/rdf-tests.git>.
2023-04-06Update TriG test suiteDavid Robillard4-18/+34
This suite can't be exactly copied because it has modifications to support exact streaming tests, but this commit incorporates some upstream changes manually.
2023-04-06Update Turtle test suiteDavid Robillard19-21/+123
This suite can't be exactly copied because it has modifications to support exact streaming tests, but this commit incorporates some upstream changes manually.
2023-04-06Update NQuads test suiteDavid Robillard7-13/+29
2023-04-06Update NTriples test suiteDavid Robillard3-2/+18
2023-04-06Use more consistent meson formattingDavid Robillard1-2/+2
2023-04-06Use new meson strict warning systemDavid Robillard2-21/+15
Meson includes warning_level=everything as of version 1.0.0, which supercedes maintaining redundant lists of warning flags here.
2023-04-05Add tests for EOF within UTF-8 charactersDavid Robillard3-0/+16
2023-04-05Clean up remaining stale test suite dataDavid Robillard20-300/+307
2023-04-05Add a simple abbreviation test suiteDavid Robillard9-0/+94
2023-04-05Move big test cases to a separate suiteDavid Robillard10-21/+38
2023-04-05Factor out a suite of "perfect" reversible testsDavid Robillard63-20141/+80428
2023-04-05Clean up "good" test suiteDavid Robillard49-50262/+20276
2023-04-05Separate test suites and simplify licensing metadataDavid Robillard1226-25/+24
2023-04-05Replace complicated test support codeDavid Robillard3-416/+116
Everything covered by the old hairy runner script is now covered by more focused test suites, so the "thru" pass can be eliminated without losing significant coverage.
2023-04-05Improve test suite coverageDavid Robillard2-2/+11
2023-04-05Test full URI preservation separatelyDavid Robillard4-0/+33
2023-04-05Use a consistent naming scheme for "good" testsDavid Robillard6-13/+16
2023-04-05Run lax suite with the simpler runner scriptDavid Robillard3-21/+52
2023-04-05Test adding blank node prefixes separatelyDavid Robillard2-6/+13
2023-04-05Improve writer error handlingDavid Robillard1-1/+1
2023-04-05Fix relative URI creationDavid Robillard5-17/+13
2023-04-05Test URI and blank node manipulation features separatelyDavid Robillard17-10/+227
2023-04-05Clean up relative URI unit testsDavid Robillard1-31/+156
2023-04-05Clean up Python scripts and add a formatting testDavid Robillard5-93/+102
2023-04-05Use shorter URIs in lax testsDavid Robillard2-12/+12
2023-04-05Fix potential memory leaks when a write is abortedDavid Robillard1-0/+49
Also clean up and simplify writer context management in general.
2023-04-05Strengthen URI utility unit testsDavid Robillard1-6/+29
2023-04-05Fix pretty-printing of anonymous subjectsDavid Robillard6-0/+125
2023-04-05Add pretty-printing test suiteDavid Robillard31-109/+618
The earlier "test" was just hitting the code without actually checking the output. This new suite is a set of pretty-printed documents which serd must reproduce exactly to pass. This should make it easy to add cases in the future, since each case is just a document, as it should look.
generated by cgit v1.2.1 (git 2.18.0) at 2024-09-19 09:22:07 +0000