phit

Data & Workflow

Statistics, Modeling, & Visualization

CSV functions

On this page

The native CSV reader, which requires the extension. See Reading CSV files for options, types, missing values, and errors.

Reading a file

read_csv()

Reads a delimited text file into typed columns, with the records, names, guessed types, values, and problem messages of phibble’s PHP reader and readr::read_delim().

  • Dates are int days since 1970-01-01, and datetimes are int microseconds since the epoch, in UTC. NA values are null.
  • failed counts values that did not parse under the column’s type; they are null. Skipped columns are null.
Phit\read_csv(    string $path,    array $options = [],): array
Parameter Type Default Description
$path string required A readable file (bytes are passed through; no encoding validation)
$options array [] readr’s defaults when omitted

Returns array{names: list<string>, columns: list<null\|array{type: 'bool'\|'int'\|'float'\|'string'\|'date'\|'datetime', values: list<bool\|int\|float\|string\|null>, failed: int}>, problems: list<string>}

Errors ValueError: a missing or unreadable file, or an invalid option.

See also Details

Reading in chunks

CsvReader

Reads a CSV file a chunk of rows at a time, as phibble’s DataFrame::csvChunks() (after readr::read_delim_chunked()): only the current chunk is held. Options are read_csv()’s; types not given are guessed from the first chunk (guess_max of its rows) and every later chunk is parsed as those types. Problem rows are numbered from the start of the file.

CsvReader::__construct()

Opens the file and reads its header line.

public function __construct(string $path, array $options = [])
Parameter Description
$path A readable file
$options

Errors ValueError: missing or unreadable file, or an invalid option.

CsvReader::names()

Column names: the header’s, or X1..Xn with col_names false; [] for an empty file.

public function names(): array

Returns list<string>

CsvReader::next()

The next chunk of at most $rows rows in read_csv()’s shape, with the 1-based position of its first row and its row count; null once no rows remain. A chunk without rows comes only to carry problems found after the last row.

public function next(int $rows): ?array

Returns null\|array{position: int, rows: int, names: list<string>, columns: list<null\|array{type: 'bool'\|'int'\|'float'\|'string'\|'date'\|'datetime', values: list<bool\|int\|float\|string\|null>, failed: int}>, problems: list<string>}

Errors ValueError: rows below 1, or a read error.