-
Notifications
You must be signed in to change notification settings - Fork 462
contrib/database/sql: fix bug where options were always overwritten by register options #1904
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
…y register options
spans := mt.FinishedSpans() | ||
// the number of spans should be 0 since we specified the WithChildSpansOnly option | ||
assert.Len(t, spans, 0) | ||
}) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please add an empty line between test cases here and throughout 🙌
@@ -295,11 +348,14 @@ func TestRegister(_ *testing.T) { | |||
} | |||
|
|||
func TestNamingSchema(t *testing.T) { | |||
newGenSpansFunc := func(t *testing.T, driverName string) namingschematest.GenSpansFn { | |||
newGenSpansFunc := func(t *testing.T, driverName string, registerOverride string) namingschematest.GenSpansFn { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can registerOverride
be a bool ? It seems like it's either register-override
or empty string ""
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it's a good suggestion, however I've set it as a string
to keep it consistent with the serviceOverride string
option I've been adding to all these tests. The reason I used a string originally is because later this value is used for the test assertions, so I thought it would preferable to give control to the caller so they know what value to assert for.
I don't have a strong preference for one or the other, but if possible I would like to keep it consistent with the rest of the TestNamingSchema
tests from other packages.
…ch as queuename tags contrib: upgrade labstack/echo/v4 from v4.2.0 to v4.9.0 (#1891) ci: fix flaky lint job (#1892) contrib/elasticsearch: use naming schema (#1897) ci: introduce golangci (#1898) appsec: suspicious request blocking (#1797) Co-authored-by: Julio Guerra <julio@datadog.com> ci/golangci-lint: skip google.golang.org/grpc.v12 (#1899) .github/workflows: run ASM and RC system-tests scenarios (#1900) contrib/hashicorp/vault: use naming schema (#1868) contrib/database/sql: add WithIgnoreQueryTypes option (#1823) Co-authored-by: Zarir Hamza <zarir.hamza@datadoghq.com> Co-authored-by: Rodrigo Argüello <rodrigo.arguello@datadoghq.com> contrib/database/sql: use naming schema (#1895) internal/appsec: add server.request.method address (#1893) Signed-off-by: Eliott Bouhana <eliott.bouhana@datadoghq.com> Co-authored-by: François Mazeau <francois.mazeau@datadoghq.com> internal/appsec/dyngo: atomic instrumentation swapping (#1873) Co-authored-by: François Mazeau <francois.mazeau@datadoghq.com> go.mod: datadog-agent/pkg/remoteconfig/state 7.45.0-rc.1 (#1902) internal/version: bump to v1.51.0 (#1912) ddtrace/tracer: don't set empty tracestate propagation tag (#1910) go.mod: github.com/DataDog/datadog-agent/pkg/obfuscate 7.45.0-rc.1 (#1916) appsec: add blocking SDK body operation (#1901) * Modifying the appsec api: adding appsec.MonitorParsedHTTPBody an error as return value * Adding a call to the WAF to check for security event synchronously with a call to appsec.MonitorParsedHTTPBody on the body passed as parameter * Removing the call to the WAF done on the body an the end of a request because we moved it. * Refactoring the waf addresses storage and access Signed-off-by: Eliott Bouhana <eliott.bouhana@datadoghq.com> ddtrace/{opentelemetry,opentracer}: add telemetry (#1909) internal/appsec: fix user ID event detection (#1918) internal/telemetry: track tracer init time metric (#1896) Co-authored-by: Andrew Glaude <andrew.glaude@datadoghq.com> internal/appsec/remoteconfig: fix rules overrides (#1921) contrib/mongodb: use naming schema (#1908) contrib/syndtr/goleveldb/leveldb: use naming schema (#1914) contrib/tidwall/buntdb: use naming schema (#1913) internal/appsec: do not ignore the appsec events rate limiter (#1927) remoteconfig: remove empty products and don't override appsec rules data (#1925) contrib/kafka: refactor tests (#1907) contrib/google.golang.org/grpc: use naming schema (#1919) contrib/twitchtv/twirp: use naming schema (#1920) contrib/http: use naming schema (#1929) ddtrace/tracer: reset decision maker during fallback behavior of w3c header extraction (#1933) contrib/cassandra: use naming schema (#1911) Co-authored-by: Diana Shevchenko <40775148+dianashevchenko@users.noreply.github.com> contrib/redis: use naming schema (#1906) Co-authored-by: Andrew Glaude <andrew.glaude@datadoghq.com> ci/system-tests: more scenarios with parallel jobs (#1938) ci: update linter job and add bodyclose (#1942) contrib/redis/go-redis.v9: support v9 (#1730) Add support for new go-redis version v9. It does 2 things: Copy existing version 8 files to a new path, /redis/go-redis.v9. Make changes to support version 9. Fixes #1710 format and rerun go tidy get rid of prints add topLevelRegion assertions remove confusing named return values and todo comment ddtrace/tracer: ensure access to trace tags is concurrency-safe (#1948) Spancontext marshaling was accessing tracer internal structures without a lock, resulting in a data race and panic. This commit adds a few methods to trace to allow safe access to the tags and propagatingTags members of trace to the marshaling code. Fixes #1944 ddtrace/tracer: mark context updated when SetUser is called (#1949) Fixes a minor logic mistake when setting a user on a span lint and add default switch case refactor resourceNameKey and value assignments restructure functions to be left aligned use internal logger, be less verbose with function names go back to normal switch type and format Set keyTraceID128 on first span in the chunk only (#1946) go.mod: upgrade go-libddwaf to v1.2.0 (#1953) Co-authored-by: Julio Guerra <julio@datadog.com> contrib/database/sql: fix bug where options were always overwritten by register options (#1904) Co-authored-by: Diana Shevchenko <40775148+dianashevchenko@users.noreply.github.com> ci/smoke-tests: update the go.sum file after go get -u (#1957) contrib/net/http: don't set empty string values as span tags (#1956) Do not set span fields when they are not configured so the tracer can put the defaults in. use normal string then derefence rever go.mod and go.sum changes contrib/internal/httptrace: remove naming schema from init (#1960) contrib/graphql: use naming schema (#1926) internal/telemetry: trim the dependencies version prefix v (#1963) contrib/aws: use naming schema (#1931) contrib/cloud.google.com/go/pubsub.v1: use naming schema (#1937) go mod tidy lint and fix test
…y register options (#1904) Co-authored-by: Diana Shevchenko <40775148+dianashevchenko@users.noreply.github.com>
What does this PR do?
Open
orOpenDB
functions were always ignored and replaced by the value provided during theRegister
driver call (tested in the newTestOpenOptions
test cases).Register
instead of the default<driverName>.db
is also used in naming schema v1 (tested in the newTestNamingSchema/PostgresWithRegisterOverride
test case).Motivation
Bug fixes.
Describe how to test/QA your changes
Reviewer's Checklist