nushell/crates/nu-cmd-dataframe/src/dataframe/eager/query_df.rs
Jack Wright f879c00f9d
The ability to specify a schema when using dfr open and dfr into-df (#11634)
# Description

There are times where explicitly specifying a schema for a dataframe is
needed such as:
- Opening CSV and JSON lines files and needing provide more information
to polars to keep it from failing or in a desire to override default
type conversion
- When converting a nushell value to a dataframe and wanting to override
the default conversion behaviors.

This pull requests provides:
- A flag to allow specifying a schema when using dfr into-df
- A flag to allow specifying a schema when using dfr open that works for
CSV and JSON types
- A new command `dfr schema` which displays schema information and will
allow display support schema dtypes

Schema is specified creating a record that has the key value and the
dtype. Examples usages:

```
{a:1, b:{a:2}} | dfr into-df -s {a: u8, b: {a: i32}} | dfr schema
{a: 1, b: {a: [1 2 3]}, c: [a b c]} | dfr into-df -s {a: u8, b: {a: list<u64>}, c: list<str>} | dfr schema
 dfr open -s {pid: i32, ppid: i32, name: str, status: str, cpu: f64, mem: i64, virtual: i64} /tmp/ps.jsonl  | dfr schema
```

Supported dtypes:
null                                                   
bool                                                   
u8                                                     
u16                                                    
u32                                                    
u64                                                    
i8                                                     
i16                                                    
i32                                                    
i64                                                    
f32                                                    
f64                                                    
str                                                    
binary                                                 
date                                                   
datetime[time_unit: (ms, us, ns) timezone (optional)]  
duration[time_unit: (ms, us, ns)]                      
time                                                   
object                                                 
unknown                                                
list[dtype]


structs are also supported but are specified via another record:
{a: u8, b: {d: str}}

Another feature with the dfr schema command is that it returns the data
back in a format that can be passed to provide a valid schema that can
be passed in as schema argument:

<img width="638" alt="Screenshot 2024-01-29 at 10 23 58"
src="https://github.com/nushell/nushell/assets/56345/b49c3bff-5cda-4c86-975a-dfd91d991373">

---------

Co-authored-by: Jack Wright <jack.wright@disqo.com>
2024-01-29 13:26:04 -06:00

109 lines
3.1 KiB
Rust

use super::super::values::NuDataFrame;
use crate::dataframe::values::Column;
use crate::dataframe::{eager::SQLContext, values::NuLazyFrame};
use nu_engine::CallExt;
use nu_protocol::{
ast::Call,
engine::{Command, EngineState, Stack},
Category, Example, PipelineData, ShellError, Signature, Span, SyntaxShape, Type, Value,
};
// attribution:
// sql_context.rs, and sql_expr.rs were copied from polars-sql. thank you.
// maybe we should just use the crate at some point but it's not published yet.
// https://github.com/pola-rs/polars/tree/master/polars-sql
#[derive(Clone)]
pub struct QueryDf;
impl Command for QueryDf {
fn name(&self) -> &str {
"dfr query"
}
fn usage(&self) -> &str {
"Query dataframe using SQL. Note: The dataframe is always named 'df' in your query's from clause."
}
fn signature(&self) -> Signature {
Signature::build(self.name())
.required("sql", SyntaxShape::String, "sql query")
.input_output_type(
Type::Custom("dataframe".into()),
Type::Custom("dataframe".into()),
)
.category(Category::Custom("dataframe".into()))
}
fn search_terms(&self) -> Vec<&str> {
vec!["dataframe", "sql", "search"]
}
fn examples(&self) -> Vec<Example> {
vec![Example {
description: "Query dataframe using SQL",
example: "[[a b]; [1 2] [3 4]] | dfr into-df | dfr query 'select a from df'",
result: Some(
NuDataFrame::try_from_columns(
vec![Column::new(
"a".to_string(),
vec![Value::test_int(1), Value::test_int(3)],
)],
None,
)
.expect("simple df for test should not fail")
.into_value(Span::test_data()),
),
}]
}
fn run(
&self,
engine_state: &EngineState,
stack: &mut Stack,
call: &Call,
input: PipelineData,
) -> Result<PipelineData, ShellError> {
command(engine_state, stack, call, input)
}
}
fn command(
engine_state: &EngineState,
stack: &mut Stack,
call: &Call,
input: PipelineData,
) -> Result<PipelineData, ShellError> {
let sql_query: String = call.req(engine_state, stack, 0)?;
let df = NuDataFrame::try_from_pipeline(input, call.head)?;
let mut ctx = SQLContext::new();
ctx.register("df", &df.df);
let df_sql = ctx
.execute(&sql_query)
.map_err(|e| ShellError::GenericError {
error: "Dataframe Error".into(),
msg: e.to_string(),
span: Some(call.head),
help: None,
inner: vec![],
})?;
let lazy = NuLazyFrame::new(false, df_sql);
let eager = lazy.collect(call.head)?;
let value = Value::custom_value(Box::new(eager), call.head);
Ok(PipelineData::Value(value, None))
}
#[cfg(test)]
mod test {
use super::super::super::test_dataframe::test_dataframe;
use super::*;
#[test]
fn test_examples() {
test_dataframe(vec![Box::new(QueryDf {})])
}
}