Skip to content

Instantly share code, notes, and snippets.

@kevinmichaelchen
Created August 25, 2015 22:07
Show Gist options
  • Select an option

  • Save kevinmichaelchen/79948b9d0b448a51c472 to your computer and use it in GitHub Desktop.

Select an option

Save kevinmichaelchen/79948b9d0b448a51c472 to your computer and use it in GitHub Desktop.
Decoding split-up email attachment filenames
class Test {
public static void main( String[] args )
{
final String contentDisposition = "name: Content-Disposition\n" +
"value: attachment;\n" +
"\tsize=1123785;\n" +
"\tmodification-date=\"Thu, 20 Aug 2015 17:27:44 GMT\";\n" +
"\tfilename*0=\"=?utf-8?B?TmltciAoNCDDlyA0KSBhbmQgKDYgw5cgNikgaGlnaCBtb2JpbGl0eS\";\n" +
"\tfilename*1=\"B2ZWhp?= cles.pdf\";\n" +
"\tcreation-date=\"Thu, 20 Aug 2015 17:27:44 GMT\"";
System.out.println( contentDisposition );
System.out.println( "" );
String[] lines = contentDisposition.split( "\n" );
String encodedString = "";
String unencodedTail = "";
Pattern filenamePattern = Pattern.compile( "filename\\*\\d+" );
Pattern firstFilenamePattern = Pattern.compile( "filename\\*0" );
for ( String line : lines )
{
// remove the tabs on each line beginning
line = line.trim();
// if the pattern matches...
if ( filenamePattern.matcher( line ).find() )
{
boolean firstFilenamePart = firstFilenamePattern.matcher( line ).find();
boolean lastFilenamePart = !firstFilenamePart && line.contains( " " );
// remove the crap before the first quote
// convert
// filename*0="=?utf-8?B?Tmlt"
// to
// =?utf-8?B?Tmlt"
int firstQuoteIndex = line.indexOf( "\"" );
line = line.substring( firstQuoteIndex + 1 );
// the first part of the filename is special, since it tells us
// if there's an encoding and also specifies which encoding
// e.g., filename*0="=?utf-8?B?Tmlt
if ( firstFilenamePart )
{
// remove encoding prefix
final String encodingPrefix = "=?utf-8?";
int encodingPrefixIndex = line.indexOf( encodingPrefix );
line = line.substring( encodingPrefixIndex + encodingPrefix.length() );
// check for base 64
if ( line.startsWith( "B?" ) )
{
line = line.substring( 2 );
}
// check for QP encoding
else if ( line.startsWith( "Q?" ) )
{
line = line.substring( 2 );
}
// alert me if I'm missing anything
else
{
throw new RuntimeException( "Unspecified filename encoding for line: " + line );
}
}
if ( !lastFilenamePart )
{
// get everything until the next quote
int nextQuoteIndex = line.indexOf( "\"" );
line = line.substring( 0, nextQuoteIndex );
encodedString += line;
}
else {
// e.g., B2ZWhp?= cles.pdf"
// get stuff between quote and space, e.g., B2ZWhp?=
int firstSpaceIndex = line.indexOf( " " );
encodedString += line.substring( 0, firstSpaceIndex );
// add unencoded string after space
line = line.substring( firstSpaceIndex + 1 );
int nextQuoteIndex = line.indexOf( "\"" );
unencodedTail += line.substring( 0, nextQuoteIndex );
}
}
}
System.out.println( "res: " + encodedString );
System.out.println( "tail: " + unencodedTail );
byte[] bytes = Base64.decodeBase64( encodedString );
String result = new String( bytes ) + unencodedTail;
System.out.println( "result: " + result );
}
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment