Skip to content

Latest commit

 

History

History
42 lines (31 loc) · 2.02 KB

README.md

File metadata and controls

42 lines (31 loc) · 2.02 KB
Raw

Multibyte String Stream Filter Build Status

A php_user_filter implementation for quickly converting stream character sets without loading the entire string into memory.

Note: this library does not have the ability to persist state for stateful character sets, and as such, could potentially lose data. See #1 for an open issue, and this document for more info about stateful character sets.

Example

// Not required if the file was autoloaded (e.g. using composer)
MultibyteStringStream::registerStreamFilter();

$native_file = fopen('iso-8859-1-file.txt', 'r');

stream_filter_append($native_file, 'convert.mbstring.UTF-8/ISO-8859-1');

$unicode_file = fopen('utf8-file.txt', 'w');

stream_copy_to_stream($native_file, $unicode_file);

mbstring-stream also works as a write filter:

$native_file  = fopen('sjis-file.txt', 'r');
$unicode_file = fopen('utf8-file.txt', 'w');

stream_filter_append($unicode_file, 'convert.mbstring.UTF-8/SJIS');

stream_copy_to_stream($native_file, $unicode_file);

Usage

/**
 * resource   $stream        The stream to filter.
 * string     $to_encoding   The encoding to convert to.
 * string     $from_encoding The encoding to convert from. Optional, defaults to mb_internal_encoding()
 * int        $read_write    See http://php.net/manual/en/function.stream-filter-append.php
 * string|int $sub_char      The substitute character to use. Optional, defaults to mb_substitute_character()
 */
stream_filter_append($stream, "convert.mbstring.$to_encoding/$from_encoding", $read_write, $sub_char);

Note: Be careful when using on streams in 'r+' or 'w+' (or similar) modes; by default PHP will assign the filter to both the reading and writing chain. This means it will attempt to convert the data twice - once when writing to the stream, and once when reading from it.