 |
|
 PhpDig.net |
What is PhpDig?
PhpDig is a PHP MySQL based Web Spider & Search Engine. |
recode_file
Description
recode_file() acts as a filter between two file handles - reading all of the content from the current offset of the file position indicator for input_file , filtering it through recode, and writing it to output_file at the current offset of the file position indicator. Both file handles must be opened by calls to fopen() . input_file must have been opened in a mode that allows reading, while output_file must have been opened in a mode that allows writing.
The recode argument should consist of the name of the character set from which to convert, followed by two periods and the name of the target character set; for example, to convert from ASCII to base 64, the argument is "ASCII..64".
If file handle is unusable or the recode request cannot be handled by recode, the function returns FALSE .
Caution
This function may return FALSE in some situations where it successfully recodes the character set.
See also
To recode a single string
Other functions are available for building internationalized applications in PHP:
see the GNU Gettext functions.
Example
Example 1119. Fetch a news story from a web page and strip all the accents out of the text
Note
A warning such as recode: Ambiguous output in step `ISO-8859-1..ASCII-BS' may be generated by this code. This error indicates that the file contains a mix of UNIX-style (\n) and Windows-style (\r\n) line endings. The recode should work properly despite the error.
<?php
# Set the URL of the site to retrieve data from
$base_URL = "http://example.com/new_stories";
# Make a numeric date string (ie. 20010915) for today
$date = date ('Ymd');
# Build a filename for the first news story for the given date
$html_file = $date.'0001.html';
# Open a file handle to the URL
$input = fopen ("$base_URL/$html_file", 'r');
# recode_file doesn't like file handles that point to remote files
# Circumvent this by using a temp file to store the URLs content
$tmp = tmpfile ();
# Write all data from the URL to the temp file
while (! feof ($input)) {
# Read in a kilobyte of data from the URL
$data = fgets ($input, 1024);
# Convert any Windows-style line ends (\r\n) to Unix-style line ends (\n)
# We do this because recode does not like it when we mix line ends within a file
$data = str_replace ("\r\n", "\n", $data);
# write the data to the temp file
fwrite ($tmp, $data);
}
fclose ($input);
# Return to the start of the temp file before reading data from it
rewind ($tmp);
# Open a handle to a new local file
$output = fopen ("./example.com/news_stories/$html_file", 'w');
# Convert from ASCII to flat (ASCII without diacritics or underlines)
# Suppress possible errors with a single @ symbol
@ recode_file ('us..flat', $tmp, $output);
# Close the temp file
# Note: When a file opened by tmpfile() is closed, it's automatically deleted
fclose ($tmp);
fclose ($output);
?>