weblib MDL-22664 html_to_text should not strip images, it should replace them by their alt text.

Also, an new optional argument to html_to_text to control word-wrapping.
This commit is contained in:
Tim Hunt
2010-08-04 17:39:56 +00:00
parent 791325ae6c
commit f47906fbf5
4 changed files with 47 additions and 5 deletions
+4 -1
View File
@@ -219,7 +219,7 @@ class html2text
'-',
'*',
'£',
'EUR', // Euro sign. € ?
'EUR', // Euro sign. € ?
' ' // Runs of spaces, post-handling
);
@@ -237,6 +237,7 @@ class html2text
'/<(a) [^>]*href=("|\')([^"\']+)\2[^>]*>(.*?)<\/a>/i',
// <a href="">
'/<(th)[^>]*>(.*?)<\/th>/i', // <th> and </th>
'/<(img)[^>]*alt=\"([^>"]+)\"[^>]*>/i', // <img> with alt
);
/**
@@ -574,6 +575,8 @@ class html2text
return $this->_strtoupper("\n\n". $matches[2] ."\n\n");
case 'a':
return $this->_build_link_list($matches[3], $matches[4]);
case 'img':
return '[' . $matches[2] . ']';
}
}