Creating and loading tables

tibble()

tibble(mixed ...$columns): DataFrame

Build a DataFrame from named arguments, one per column, as tibble’s tibble(). Same rules as DataFrame::tibble().

Function in Phibble
Returns DataFrame

tribble()

tribble(array $names, mixed ...$cells): DataFrame

Build a DataFrame from column names, then rows, as tibble’s tribble(). Same rules as DataFrame::tribble().

Function in Phibble
Returns DataFrame

as_tibble()

as_tibble(mixed $data, string $nameRepair = 'check_unique'): DataFrame

Convert tables, rows, columns, or collections into a DataFrame, as tibble’s as_tibble(). Same rules as DataFrame::asTibble().

Function in Phibble
Returns DataFrame

enframe()

enframe(    array $values,    ?string $name = 'name',    string $value = 'value',): DataFrame

An array as a two-column DataFrame of names and values, as tibble’s enframe(). Same rules as DataFrame::enframe().

Function in Phibble
Returns DataFrame

tbl()

tbl(PDO $pdo, string $from, array $bindings = []): SqlTable

A lazy SqlTable, as dbplyr’s tbl(). Pass a table name, or a SELECT or WITH query with bindings.

tbl($pdo, ‘visits’); tbl($pdo, ‘SELECT * FROM visits WHERE year = ?’, [2024]);

Function in Phibble
Returns SqlTable
Throws InvalidArgumentException: when bindings come with a table name.

polars()

polars(mixed $data): PolarsTable

A PolarsTable from any Table, or from anything as_tibble() accepts. Requires the php_polars extension.

polars(DataFrame::fromCsv('sales.csv'))->filter(col('units')``->gt(0));

Function in Phibble
Returns PolarsTable
Throws RuntimeException: when the php_polars extension is not loaded.

DataFrame::fromRows()

DataFrame::fromRows(    iterable $rows,    array $types = [],    bool $parseStrings = false,): DataFrame

Build a DataFrame from rows. Column order follows first appearance; keys missing from a row are NA.

Static method of Phibble\DataFrame
Returns DataFrame
Examples Getting started

DataFrame::fromColumns()

DataFrame::fromColumns(    object|array $columns,    array $types = [],): DataFrame

Build a DataFrame from columns: name => list of values, or name => Column.

An object with a public toColumns() method, such as phit’s Phit\Table or another phibble table, is read through that method.

Static method of Phibble\DataFrame
Returns DataFrame
Throws InvalidArgumentException: when an object has no toColumns() method, or it returns something other than an array.
Examples Types, Getting started

DataFrame::fromCsv()

DataFrame::fromCsv(    mixed $file,    array|string $types = [],    string $delimiter = ',',    array $na = ['', 'NA'],    mixed ...$options,): DataFrame

Read delimited text like readr::read_csv(): a path (“.gz” is decompressed), an http:// or https:// URL (downloaded to a temporary file first), or an open stream. Types are guessed from up to guessMax rows; “” and “NA” are NA. More readr options go as named arguments; see Phibble\Io\CsvOptions (colNames, select, skip, nMax, quote, comment, trimWs, skipEmptyRows, guessMax, nameRepair, decimalMark, groupingMark, escapeDouble, escapeBackslash).

Problems (unparseable values, ragged rows) raise one E_USER_WARNING; the data is still returned, with NA where values failed.

Static method of Phibble\DataFrame
Returns DataFrame
Examples Getting started, Reading and writing data, Expressions

DataFrame::fromCsvString()

DataFrame::fromCsvString(    string $csv,    array|string $types = [],    string $delimiter = ',',    array $na = ['', 'NA'],    mixed ...$options,): DataFrame

Read delimited text from a string; options as fromCsv().

Static method of Phibble\DataFrame
Returns DataFrame
Examples Reading and writing data, Types

DataFrame::fromCsvChunked()

DataFrame::fromCsvChunked(    mixed $file,    callable $callback,    int $chunkSize = 10000,    array|string $types = [],    string $delimiter = ',',    array $na = ['', 'NA'],    mixed ...$options,): void

Read delimited text in chunks, as readr::read_delim_chunked(): each chunk of at most $chunkSize rows is passed to $callback with the 1-based index of its first row. Returning false from the callback stops reading. Memory follows the chunk size, not the file’s.

Types not given are guessed from the first chunk (guessMax defaults to the chunk size, as in readr) and every later chunk is parsed as those types. Give types, for example [‘.default’ => ‘c’], to skip guessing. Factors without levels take their levels from each chunk. Each chunk’s problems raise their own warning, with rows numbered from the start of the file. Options are those of fromCsv(), and $file may be a URL as there. With the phit extension loaded, the chunks are read natively by Phit\CsvReader.

Static method of Phibble\DataFrame
Examples Reading and writing data

DataFrame::csvChunks()

DataFrame::csvChunks(    mixed $file,    int $chunkSize = 10000,    array|string $types = [],    string $delimiter = ',',    array $na = ['', 'NA'],    mixed ...$options,): Generator

The chunks fromCsvChunked() reads, as a generator keyed by the 1-based index of each chunk’s first row.

$kept = []; foreach (DataFrame::csvChunks('feed.tsv', delimiter: "\t", types: ['.default' => 'c']) as $chunk) { $kept[] = $chunk->filter(col('category')``->eq('running')); } $shoes = DataFrame::concat($kept);

Static method of Phibble\DataFrame
Returns Generator<int, DataFrame>
Examples Reading and writing data

DataFrame::fromJson()

DataFrame::fromJson(    string $json,    array $types = [],    ?string $orient = null,): DataFrame

Read JSON: a list of records ([{“a”: 1}, …]) or an object of columns ({“a”: [1, 2]}). The shape is detected unless $orient is ‘records’ or ‘columns’. Missing keys and nulls are NA; nested arrays and objects become list-columns.

Static method of Phibble\DataFrame
Returns DataFrame
Examples Reading and writing data

DataFrame::fromPdo()

DataFrame::fromPdo(    PDO $pdo,    string $sql,    array $bindings = [],    array $types = [],): DataFrame

Run a query and load the result.

Static method of Phibble\DataFrame
Returns DataFrame
Examples Reading and writing data

DataFrame::fromQuery()

DataFrame::fromQuery(    object $builder,    ?object $connection = null,    array $types = [],): DataFrame

Load the result of a Laravel query builder (query or Eloquent) or a Doctrine DBAL QueryBuilder. Neither library is required.

Static method of Phibble\DataFrame
Returns DataFrame

DataFrame::fromCollection()

DataFrame::fromCollection(    Traversable|object|array $collection,    array $types = [],): DataFrame

Build a DataFrame from a Laravel Collection (or anything with all() or toArray(), or any iterable). Items may be arrays, Eloquent models or other objects with toArray(), or plain objects. Laravel is not needed.

Static method of Phibble\DataFrame
Returns DataFrame
Examples Reading and writing data

DataFrame::fromSource()

DataFrame::fromSource(Source $source): DataFrame

Read a Phibble\Source\Source, such as a CsvSource or PdoSource, into a DataFrame.

Static method of Phibble\DataFrame
Returns DataFrame

DataFrame::tibble()

DataFrame::tibble(    array $columns,    string $nameRepair = 'check_unique',): DataFrame

Build a DataFrame column by column, as tibble’s tibble().

• A list is a column; a scalar (or null) is repeated to the common length. Only length-1 values are recycled.
• A closure receives the DataFrame built so far and returns a list or a scalar, so columns can be computed from earlier ones.
• A Table under an integer key is spliced in column by column.

Static method of Phibble\DataFrame
Returns DataFrame

DataFrame::tribble()

DataFrame::tribble(array $names, mixed ...$cells): DataFrame

Build a DataFrame row by row, as tibble’s tribble(). Pass the column names, then either one list per row or the cells in row order.

DataFrame::tribble(['x', 'y'], [1, 'a'], [2, 'b']);

Static method of Phibble\DataFrame
Returns DataFrame

DataFrame::asTibble()

DataFrame::asTibble(    mixed $data,    string $nameRepair = 'check_unique',): DataFrame

Convert tables, rows, columns, collections, or other iterables into a DataFrame, as tibble’s as_tibble().

• A Table is collected.
• A list of associative arrays (or objects) is rows.
• An associative array of lists is columns.
• An object with toColumns(), such as phit’s Phit\Table, is read as columns.
• An object with all() or toArray() (a Laravel Collection) is unwrapped.

Static method of Phibble\DataFrame
Returns DataFrame

DataFrame::enframe()

DataFrame::enframe(    array $values,    ?string $name = 'name',    string $value = 'value',): DataFrame

An array as a two-column DataFrame of names and values, as tibble’s enframe(). A list gets 1-based integer names, as in R.

Static method of Phibble\DataFrame
Returns DataFrame

SqlTable::table()

SqlTable::table(PDO $pdo, string $name): SqlTable

A table (or view) by name.

Static method of Phibble\Sql\SqlTable
Returns SqlTable

SqlTable::query()

SqlTable::query(PDO $pdo, string $sql, array $bindings = []): SqlTable

Any SELECT query with positional (?) or named (:name) bindings.

Static method of Phibble\Sql\SqlTable
Returns SqlTable

SqlTable::fromQuery()

SqlTable::fromQuery(    object $builder,    ?object $connection = null,): SqlTable

A Laravel or Doctrine DBAL query builder, kept lazy.

Static method of Phibble\Sql\SqlTable
Returns SqlTable

SqlTable::fromSource()

SqlTable::fromSource(PdoSource $source): SqlTable

A lazy table over a PdoSource’s query and bindings.

Static method of Phibble\Sql\SqlTable
Returns SqlTable

PolarsTable::fromTable()

PolarsTable::fromTable(Table $table): PolarsTable

Copy any phibble table into Polars, keeping types, groups and factor levels.

Static method of Phibble\Polars\PolarsTable
Returns PolarsTable

PolarsTable::fromPolars()

PolarsTable::fromPolars(    Phpolars\DataFrame|Phpolars\LazyFrame $df,): PolarsTable

Wrap a Polars DataFrame; a LazyFrame is collected.

Static method of Phibble\Polars\PolarsTable
Returns PolarsTable