alamb commented on code in PR #20282:
URL: https://github.com/apache/datafusion/pull/20282#discussion_r2801071136
##########
.github/workflows/extended.yml:
##########
@@ -167,11 +167,19 @@ jobs:
uses: ./.github/actions/setup-builder
with:
rust-version: stable
+ - name: Build sqllogictest binary
+ run: |
+ # Cargo emits test binaries with hashed filenames under
target/.../deps.
+ # We use head to pick the first matching sqllogictests artifact from
the JSON stream.
+ TEST_BIN=$(cargo build --profile release-nonlto --features
backtrace,parquet_encryption --package datafusion-sqllogictest --test
sqllogictests --message-format=json | sed -n
's/.*"executable":"\([^"]*\)".*/\1/p' | head -n 1)
Review Comment:
I am very confused (similar to @nuno-faria ) why this is taking so much less
time
I would be hesitant to merge this until we understand why this would make
the CI run faster. It doesn't make any sense to me
As an interim step, can you try at least splitting the build and run steps
so we can see the timings
So like the build would be
```shell
cargo build --features backtrace,parquet_encryption --profile
release-nonlto --test sqllogictests
```
And then the test can be run with the same command
```shell
cargo test --features backtrace,parquet_encryption --profile
release-nonlto --test sqllogictests -- --include-sqlite
```
That way we can at least see what is taking so long.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
