Skip to content

Commit

Permalink
Fix some typos in docs
Browse files Browse the repository at this point in the history
  • Loading branch information
fkmy authored and onsi committed Nov 5, 2024
1 parent 7512021 commit b101e0c
Showing 1 changed file with 4 additions and 4 deletions.
8 changes: 4 additions & 4 deletions docs/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -3438,7 +3438,7 @@ These settings control the amount of information emitted with each spec. By def

The two verbose settings are most helpful when debugging spec suites. They make Ginkgo emit the detailed timeline information for _every_ spec regardless of failure or success. When running in series with `-v` or `-vv` mode Ginkgo will stream out the timeline in real-time while specs are running. A real-time stream isn't possible when running in parallel (the [streams would be interleaved](https://www.youtube.com/watch?v=jyaLZHiJJnE)); instead Ginkgo emits all this information about each spec right after it completes.

Very-verbose mode contains additional information over verbose mode. In particular, `-vv` timelines indicate when individual nodes start and end and also include the full failure descriptions for _every_ failure encountered by the spec. Verbose mode does not include the node start/end events (though this can be turned on with `--show-node-events`) and does not include detailed failure information for anything other than the first (primary) failure. (Additional/subseuqent failures typically occur in clean-up nodes and are not as relevant as the primary failure that occurs in a subject or setup node).
Very-verbose mode contains additional information over verbose mode. In particular, `-vv` timelines indicate when individual nodes start and end and also include the full failure descriptions for _every_ failure encountered by the spec. Verbose mode does not include the node start/end events (though this can be turned on with `--show-node-events`) and does not include detailed failure information for anything other than the first (primary) failure. (Additional/subsequent failures typically occur in clean-up nodes and are not as relevant as the primary failure that occurs in a subject or setup node).

When you [filter specs](#filtering-specs) using Ginkgo's various filtering mechanism Ginkgo usually emits a single cyan `S` for each skipped spec. If you run with the very-verbose setting, however, Ginkgo will emit the description and location information of every skipped spec. This can be useful if you need to debug your filter queries and can be paired with `--dry-run`.

Expand Down Expand Up @@ -3734,14 +3734,14 @@ When run with `--cover`, Ginkgo will generate a single `coverprofile.out` file t

Ginkgo also honors the `--output-dir` flag when generating coverprofiles. If you specify `--output-dir` the generated coverprofile will be placed in the requested directory. If you also specify `--keep-separate-coverprofiles` individual package coverprofiles will be placed in the requested directory and namespaced with a prefix that contains the name of the package in question.

Finally, when running a suite that has [programatically focused specs](#focused-specs) (i.e. specs with the `Focus` decorator or with nodes prefixed with an `F`) Ginkgo exits the suite early with a non-zero exit code. This interferes with `go test`'s profiling code and prevents profiles from being generated. Ginkgo will tell you this has happened. If you want to profile just a subset of your suite you'll need to use a different [mechanism](#filtering-specs) to filter your specs.
Finally, when running a suite that has [programmatically focused specs](#focused-specs) (i.e. specs with the `Focus` decorator or with nodes prefixed with an `F`) Ginkgo exits the suite early with a non-zero exit code. This interferes with `go test`'s profiling code and prevents profiles from being generated. Ginkgo will tell you this has happened. If you want to profile just a subset of your suite you'll need to use a different [mechanism](#filtering-specs) to filter your specs.

#### Other Profiles
Running `ginkgo` with any of `--cpuprofile=X`, `--memprofile=X`, `--blockprofile=X`, and `--mutexprofile=X` will generate corresponding profile files for suite that runs. Doing so will also preserve the test binary generated by Ginkgo to enable users to use `go tool pprof <BINARY> <PROFILE>` to analyze the profile.

By default, the test binary and various profile files are stored in the individual directories of any suites that Ginkgo runs. If you specify `--output-dir`, however, then these assets are moved to the requested directory and namespaced with a prefix that contains the name of the package in question.

As with coverage computation, these profiles will not generate a file if a suite includes programatically focused specs (see the discussion [above](#computing-coverage)).
As with coverage computation, these profiles will not generate a file if a suite includes programmatically focused specs (see the discussion [above](#computing-coverage)).

## Ginkgo and Gomega Patterns
So far we've introduced and described the majority of Ginkgo's capabilities and building blocks. Hopefully the previous chapters have helped give you a mental model for how Ginkgo specs are written and run.
Expand Down Expand Up @@ -4478,7 +4478,7 @@ Describe("Change book font-size", func() {
}()

// now we wait for the `done` channel to close. Note that we neither pass in a context nor set an explicit timeout
// in this case `Eventually` `will use Gomega's default global timeout (1 second, unless overriden by the user)
// in this case `Eventually` `will use Gomega's default global timeout (1 second, unless overridden by the user)
Eventually(done).Should(BeClosed())
})
})
Expand Down

0 comments on commit b101e0c

Please sign in to comment.