f17276e0c4
The reader's three-pass pipeline (backslash-join -> fold "+"-continuations onto raw lines -> split each folded line on ": ") broke on fields written as bare "Name:" (colon, no trailing space, no value) whose value is supplied entirely by following "+" lines -- the idiomatic recutils pattern for e.g. a "%doc:" field. Folding the continuation onto the raw line destroyed the ": " substring needed for the later split, so parsing such a field crashed with a "missing field separator" error, even on well-formed input. Found by testing against GNU recutils' own tutorial example (books.rec, from the manual's "A Little Example" page), now added as a fixture. Restructured parsing to split each line into a key/value field before folding continuations, so folding happens on the value rather than the raw line; when the preceding value is empty, the continuation becomes the value outright instead of gaining a spurious leading newline. Adds regression coverage (test/cases/io-recutils/0006, 0007) and a docs section covering the empty-value + continuation pattern. Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| cases | ||
| cases-not-suitable-for-ci | ||
| expected | ||
| input | ||
| stdlib | ||
| README.md | ||
Miller regression tests
There are a few files unit-tested with Go's testing package -- a few dozen cases total.
The vast majority of Miller tests, though -- thousands of cases -- are tested by running scripted invocations of mlr with various flags and inputs, comparing against expected output, and checking the exit code back to the shell.
How to run the regression tests, in brief
Note: while this README.md file is within the test/ subdirectory, all paths in this file are written from the perspective of the user being cd'ed into the repository base directory, i.e. this directory's parent directory.
-
mlr regtest --help -
go test github.com/johnkerl/miller/v6/pkg/...— runs the Go unit tests (a few dozen cases).
Items for the duration of the Go port
mlr regtest -c ...runs the C version of Miller from the local checkout
More details
You can alias mr='mlr regtest' for convenience. With no arguments, mr runs all cases under test/cases/. Pass one or more paths to run only those directories or specific .cmd files.
mr— run all regression cases (default path istest/cases/).mr test/cases/foo— run only cases under that directory.mr -v test/cases/foo— same, plus per-command pass/fail; use-vvor-vvvfor more detail.mr -j test/cases/foo/0003— show the Miller command, any script, and actual output for that case (handy for debugging).mr -p test/cases/foo/0003— populate: write or overwriteexpoutandexperrfrom the current run (use when adding or updating expected output).mr -c ...— use the C build of Miller (e.g.-c→../c/mlr) instead of the current executable.
To review populated files before committing, run mr -p on the desired path, then git diff to inspect changes and git reset --hard to discard them.
Creating new cases
- Create a case directory under
test/cases/, e.g.test/cases/my-feature/0001. - Add a
cmdfile containing the Miller command line (one line), e.g.mlr cat test/input/simple.dkvp. - Use shared input under
test/input/, or add a localinputfile in the case directory; incmdyou can use${CASEDIR}so the command refers to the case directory (e.g.mlr cat ${CASEDIR}/input). - Run
mlr regtest -p test/cases/my-feature/0001to generateexpout(andexperrif the command produces stderr). If the command is expected to exit non-zero, add an emptyshould-failfile. - Run
mlr regtest test/cases/my-feature/0001(without-p) to confirm the case passes.
Optional: mlr — DSL script file when the test uses -f/put/filter; env — environment variables to set for the case (unset after).