Crate rich_data 

rich-data

Tabular data for rs-rich: one row source contract with adapters behind it, so a file, a stream of records or a query result becomes a table without hand-written glue. Not a port of anything upstream; it builds on rich-ext’s public API only.

• RowSource: column names, an optional Schema (rich_ext’s format-neutral model) and rows of Values, one at a time. Rows holds them in memory and renders through rich_ext’s TableData.
• Adapters: csv (CSV and TSV through the port of Python’s csv sniffer and reader that rich --csv uses), jsonl (JSON Lines), serialize (any serde::Serialize rows), and arrow (Arrow RecordBatches, behind the off-by-default arrow feature).
• er: an ER diagram of a Schema or of SQL DDL, through rs-rich-diagram, behind the off-by-default er feature.
• infer: column types (integers, floats, booleans, dates, timestamps, nulls, text) with the evidence for each, on request.
• stats: per-column count, nulls, distinct values, min, max, mean, median, quantiles and top values, as a table or under each heading.
• window: rows for rich_ext’s virtualised tables: Rows by index, and RowWindow, one window of a forward-only source read in constant memory.
• sql: a query result set (ResultSet) with typed alignment, NULL marked apart from empty text, and a row count.
• profile: a bounded profile of any row source (types, nulls, distinct values, statistics, histograms or top values, and a missing-value map), sampled when the input is long, as JSON or drawn; and of JSON records path by path, with presence and type mixes.
• quality: data quality check results (pass, warn, fail, error, with observed and expected values and failing rows) as a report.

Conditional styles for the tables these rows become are rich_ext::table::rules.

use rich::Console;
use rich_data::{csv, infer::Inferrer, RowSource};

let mut rows = csv::CsvReader::new()
    .read("service,p99,up\nweb,120,true\napi,35.5,false\n")
    .unwrap();
assert_eq!(rows.columns(), ["service", "p99", "up"]);

// Types are only inferred when asked, and every guess carries its evidence.
let inference = Inferrer::new().infer(&rows);
assert_eq!(inference.columns()[1].data_type().to_string(), "float");
assert_eq!(inference.columns()[1].evidence().floats, 2);
inference.apply(&mut rows);

let console = Console::builder().width(40).build();
let out = console.render_export(&rows.to_table_data());
assert!(out.contains("│ api     │ 35.5 │ false │"));

Modules

arrow

Arrow RecordBatches as rows, and Arrow schemas in the format-neutral model (the arrow feature).

csv

CSV and TSV: a port of Python’s csv sniffer and reader, and a CsvReader that turns a file into Rows.

er

ER diagrams from the schema model (#247), behind the er feature.

infer

Column type inference (#267), on request and with its evidence.

jsonl

JSON Lines: one JSON value per line, as rows.

profile

Data profiles (#343, #344, #345, #346): what every column of a CSV, JSON Lines or any other RowSource holds, in one bounded pass.

quality

Data quality results (#269): a check-result model and a report that renders it the way rich_ext’s test report renders test runs.

serialize

Rows from any serde::Serialize values: a struct’s fields (or a map’s keys) are columns, and any other value is one value column.

sql

SQL-shaped result sets (#235): a query’s rows as a database shell shows them.

stats

Column statistics (#261): count, nulls, distinct values, min, max, mean, median, quantiles and the most common values, per column.

window

Windows of rows for rich_ext’s virtualised tables (#260).

Structs

DataError

Why rows could not be read: what went wrong, and where when the source knows (a 1-based line or record number).

Field

One named field: a type, whether it may be null and whether it must be present, its constraints, and optionally the source’s own name for the type, a description and metadata.

RecordSource

Rows from a stream of JSON records.

Rows

Rows held in memory: column names, an optional schema, and the rows.

RowsSource

Rows as a RowSource, from Rows::into_source.

Schema

An ordered list of top-level fields, optionally named (a table, a record type, a document), and the further tables a schema of several holds (SQL DDL’s CREATE TABLEs).

Enums

DataType

A field’s logical type.

Value

A typed table cell.

Traits

RowSource

The row source contract: column names, an optional schema, and rows.

Type Aliases

Row

One row of cells.