dataprof 0.7.0

High-performance data profiler with ISO 8000/25012 quality metrics for CSV, JSON/JSONL, and Parquet files
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

use pyo3::exceptions::PyRuntimeError;
use pyo3::prelude::*;
use std::path::Path;

use super::config::PyProfilerConfig;
use super::types::PyProfileReport;

/// Analyze a file and return a full profile report.
///
/// Format is auto-detected from the file extension unless overridden
/// via `config.format`. Supports CSV, JSON, JSONL, and Parquet.
///
/// The GIL is released during profiling so that progress callbacks
/// (which re-acquire the GIL) can execute on the profiling thread.
#[pyfunction]
#[pyo3(signature = (path, config=None))]
pub fn analyze_file(
    py: Python<'_>,
    path: &str,
    config: Option<&PyProfilerConfig>,
) -> PyResult<PyProfileReport> {
    let profiler = match config {
        Some(cfg) => cfg.to_profiler(),
        None => crate::api::Profiler::new(),
    };

    let path = path.to_string();
    let report = py
        .detach(|| profiler.analyze_file(Path::new(&path)))
        .map_err(|e| PyRuntimeError::new_err(format!("Analysis failed: {}", e)))?;

    Ok(PyProfileReport::new(report))
}