Skip to content
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

Minor: Move shared testing code into datafusion_common #7334

Merged
merged 2 commits into from
Aug 21, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
151 changes: 151 additions & 0 deletions datafusion/common/src/test_util.rs
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,8 @@

//! Utility functions to make testing DataFusion based crates easier

use std::{error::Error, path::PathBuf};

/// A macro to assert that one string is contained within another with
/// a nice error message if they are not.
///
Expand Down Expand Up @@ -62,3 +64,152 @@ macro_rules! assert_not_contains {
);
};
}

/// Returns the arrow test data directory, which is by default stored
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is moved code -- no changes

/// in a git submodule rooted at `testing/data`.
///
/// The default can be overridden by the optional environment
/// variable `ARROW_TEST_DATA`
///
/// panics when the directory can not be found.
///
/// Example:
/// ```
/// let testdata = datafusion_common::test_util::arrow_test_data();
/// let csvdata = format!("{}/csv/aggregate_test_100.csv", testdata);
/// assert!(std::path::PathBuf::from(csvdata).exists());
/// ```
pub fn arrow_test_data() -> String {
match get_data_dir("ARROW_TEST_DATA", "../../testing/data") {
Ok(pb) => pb.display().to_string(),
Err(err) => panic!("failed to get arrow data dir: {err}"),
}
}

/// Returns the parquet test data directory, which is by default
/// stored in a git submodule rooted at
/// `parquet-testing/data`.
///
/// The default can be overridden by the optional environment variable
/// `PARQUET_TEST_DATA`
///
/// panics when the directory can not be found.
///
/// Example:
/// ```
/// let testdata = datafusion_common::test_util::parquet_test_data();
/// let filename = format!("{}/binary.parquet", testdata);
/// assert!(std::path::PathBuf::from(filename).exists());
/// ```
pub fn parquet_test_data() -> String {
match get_data_dir("PARQUET_TEST_DATA", "../../parquet-testing/data") {
Ok(pb) => pb.display().to_string(),
Err(err) => panic!("failed to get parquet data dir: {err}"),
}
}

/// Returns a directory path for finding test data.
///
/// udf_env: name of an environment variable
///
/// submodule_dir: fallback path (relative to CARGO_MANIFEST_DIR)
///
/// Returns either:
/// The path referred to in `udf_env` if that variable is set and refers to a directory
/// The submodule_data directory relative to CARGO_MANIFEST_PATH
pub fn get_data_dir(
udf_env: &str,
submodule_data: &str,
) -> Result<PathBuf, Box<dyn Error>> {
// Try user defined env.
if let Ok(dir) = std::env::var(udf_env) {
let trimmed = dir.trim().to_string();
if !trimmed.is_empty() {
let pb = PathBuf::from(trimmed);
if pb.is_dir() {
return Ok(pb);
} else {
return Err(format!(
"the data dir `{}` defined by env {} not found",
pb.display(),
udf_env
)
.into());
}
}
}

// The env is undefined or its value is trimmed to empty, let's try default dir.

// env "CARGO_MANIFEST_DIR" is "the directory containing the manifest of your package",
// set by `cargo run` or `cargo test`, see:
// https://doc.rust-lang.org/cargo/reference/environment-variables.html
let dir = env!("CARGO_MANIFEST_DIR");

let pb = PathBuf::from(dir).join(submodule_data);
if pb.is_dir() {
Ok(pb)
} else {
Err(format!(
"env `{}` is undefined or has empty value, and the pre-defined data dir `{}` not found\n\
HINT: try running `git submodule update --init`",
udf_env,
pb.display(),
).into())
}
}

#[cfg(test)]
mod tests {
use super::*;
use std::env;

#[test]
fn test_data_dir() {
let udf_env = "get_data_dir";
let cwd = env::current_dir().unwrap();

let existing_pb = cwd.join("..");
let existing = existing_pb.display().to_string();
let existing_str = existing.as_str();

let non_existing = cwd.join("non-existing-dir").display().to_string();
let non_existing_str = non_existing.as_str();

env::set_var(udf_env, non_existing_str);
let res = get_data_dir(udf_env, existing_str);
assert!(res.is_err());

env::set_var(udf_env, "");
let res = get_data_dir(udf_env, existing_str);
assert!(res.is_ok());
assert_eq!(res.unwrap(), existing_pb);

env::set_var(udf_env, " ");
let res = get_data_dir(udf_env, existing_str);
assert!(res.is_ok());
assert_eq!(res.unwrap(), existing_pb);

env::set_var(udf_env, existing_str);
let res = get_data_dir(udf_env, existing_str);
assert!(res.is_ok());
assert_eq!(res.unwrap(), existing_pb);

env::remove_var(udf_env);
let res = get_data_dir(udf_env, non_existing_str);
assert!(res.is_err());

let res = get_data_dir(udf_env, existing_str);
assert!(res.is_ok());
assert_eq!(res.unwrap(), existing_pb);
}

#[test]
fn test_happy() {
let res = arrow_test_data();
assert!(PathBuf::from(res).is_dir());

let res = parquet_test_data();
assert!(PathBuf::from(res).is_dir());
}
}
18 changes: 17 additions & 1 deletion datafusion/core/src/datasource/physical_plan/csv.rs
Original file line number Diff line number Diff line change
Expand Up @@ -654,9 +654,9 @@ mod tests {
use crate::datasource::physical_plan::chunked_store::ChunkedStore;
use crate::prelude::*;
use crate::test::{partitioned_csv_config, partitioned_file_groups};
use crate::test_util::{aggr_test_schema_with_missing_col, arrow_test_data};
use crate::{scalar::ScalarValue, test_util::aggr_test_schema};
use arrow::datatypes::*;
use datafusion_common::test_util::arrow_test_data;
use futures::StreamExt;
use object_store::local::LocalFileSystem;
use rstest::*;
Expand Down Expand Up @@ -1249,4 +1249,20 @@ mod tests {
}
}
}

/// Get the schema for the aggregate_test_* csv files with an additional filed not present in the files.
fn aggr_test_schema_with_missing_col() -> SchemaRef {
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I moved this out as this is the only place it is used and rewrote it to append the column explicitly rather than copy/paste with one line difference

let fields =
Fields::from_iter(aggr_test_schema().fields().iter().cloned().chain(
std::iter::once(Arc::new(Field::new(
"missing_col",
DataType::Int64,
true,
))),
));

let schema = Schema::new(fields);

Arc::new(schema)
}
}
Loading