Convert::BinHex

CRC computation

macbinary_crc DATA, SEED

Compute the MacBinary-II-style CRC for the given DATA, with the CRC seeded to SEED. Normally, you start with a SEED of 0, and you pump in the previous CRC as the SEED if you're handling a lot of data one chunk at a time. That is:

    $crc = 0;
    while (<STDIN>) {
        $crc = macbinary_crc($_, $crc);
    }
    

Note: Extracted from the mcvert utility (Doug Moore, April '87), using a "magic array" algorithm by Jim Van Verth for efficiency. Converted to Perl5 by Eryq. Untested.

binhex_crc DATA, SEED

Compute the HQX-style CRC for the given DATA, with the CRC seeded to SEED. Normally, you start with a SEED of 0, and you pump in the previous CRC as the SEED if you're handling a lot of data one chunk at a time. That is:

    $crc = 0;
    while (<STDIN>) {
        $crc = binhex_crc($_, $crc);
    }
    

Note: Extracted from the mcvert utility (Doug Moore, April '87), using a "magic array" algorithm by Jim Van Verth for efficiency. Converted to Perl5 by Eryq.

Conversion

bin2hex

Class method, constructor. Return a converter object. Just creates a new instance of "Convert::BinHex::Bin2Hex"; see that class for details.

hex2bin

Class method, constructor. Return a converter object. Just creates a new instance of "Convert::BinHex::Hex2Bin"; see that class for details.

Construction

new PARAMHASH

Class method, constructor. Return a handle on a BinHex'able entity. In general, the data and resource forks for such an entity are stored in native format (binary) format.

Parameters in the PARAMHASH are the same as header-oriented method names, and may be used to set attributes:

    $HQX = new Convert::BinHex filename => "icon.gif",
                               type    => "GIFB",
                               creator => "CNVS";
    

open PARAMHASH

Class method, constructor. Return a handle on a new BinHex'ed stream, for parsing. Params are:

Get/set header information

creator [VALUE]

Instance method. Get/set the creator of the file. This is a four-character string (though I don't know if it's guaranteed to be printable ASCII!) that serves as part of the Macintosh's version of a MIME "content-type".

For example, a document created by "Canvas" might have creator "CNVS".

data [PARAMHASH]

Instance method. Get/set the data fork. Any arguments are passed into the new() method of "Convert::BinHex::Fork".

filename [VALUE]

Instance method. Get/set the name of the file.

flags [VALUE]

Instance method. Return the flags, as an integer. Use bitmasking to get as the values you need.

header_as_string

Return a stringified version of the header that you might use for logging/debugging purposes. It looks like this:

    X-HQX-Software: BinHex 4.0 (Convert::BinHex 1.102)
    X-HQX-Filename: Something_new.eps
    X-HQX-Version: 0
    X-HQX-Type: EPSF
    X-HQX-Creator: ART5
    X-HQX-Data-Length: 49731
    X-HQX-Rsrc-Length: 23096
    

As some of you might have guessed, this is RFC-822-style, and may be easily plunked down into the middle of a mail header, or split into lines, etc.

requires [VALUE]

Instance method. Get/set the software version required to convert this file, as extracted from the comment that preceded the actual binhex'ed data; e.g.:

    (This file must be converted with BinHex 4.0)
    

In this case, after parsing in the comment, the code:

    $HQX->requires;
    

would get back "4.0".

resource [PARAMHASH]

Instance method. Get/set the resource fork. Any arguments are passed into the new() method of "Convert::BinHex::Fork".

type [VALUE]

Instance method. Get/set the type of the file. This is a four-character string (though I don't know if it's guaranteed to be printable ASCII!) that serves as part of the Macintosh's version of a MIME "content-type".

For example, a GIF89a file might have type "GF89".

version [VALUE]

Instance method. Get/set the version, as an integer.

Decode, high-level

read_comment

Instance method. Skip past the opening comment in the file, which is of the form:

   (This file must be converted with BinHex 4.0)
    

As per RFC-1741, this comment must immediately precede the BinHex data, and any text before it will be ignored.

You don't need to invoke this method yourself; "read_header()" will do it for you. After the call, the version number in the comment is accessible via the "requires()" method.

read_header

Instance method. Read in the BinHex file header. You must do this first!

read_data [NBYTES]

Instance method. Read information from the data fork. Use it in an array context to slurp all the data into an array of scalars:

    @data = $HQX->read_data;
    

Or use it in a scalar context to get the data piecemeal:

    while (defined($data = $HQX->read_data)) {
       # do stuff with $data
    }
    

The NBYTES to read defaults to 2048.

read_resource [NBYTES]

Instance method. Read in all/some of the resource fork. See "read_data()" for usage.

Encode, high-level

encode OUT

Encode the object as a BinHex stream to the given output handle OUT. OUT can be a filehandle, or any blessed object that responds to a "print()" message.

The leading comment is output, using the "requires()" attribute.

Convert::BinHex::Bin2Hex

A BINary-to-HEX converter. This kind of conversion requires a certain amount of state information; it cannot be done by just calling a simple function repeatedly. Use it like this:

    # Create and use a "translator" object:
    my $B2H = Convert::BinHex->bin2hex;    # get a converter object
    while (<STDIN>) {
        print STDOUT $B2H->next($_);          # convert some more input
    }
    print STDOUT $B2H->done;               # no more input: finish up
    # Re-use the object:
    $B2H->rewind;                 # ready for more action!
    while (<MOREIN>) { ...

On each iteration, "next()" (and "done()") may return either a decent-sized non-empty string (indicating that more converted data is ready for you) or an empty string (indicating that the converter is waiting to amass more input in its private buffers before handing you more stuff to output.

Note that "done()" always converts and hands you whatever is left.

This may have been a good approach. It may not. Someday, the converter may also allow you give it an object that responds to read(), or a FileHandle, and it will do all the nasty buffer-filling on its own, serving you stuff line by line:

    # Someday, maybe...
    my $B2H = Convert::BinHex->bin2hex(\*STDIN);
    while (defined($_ = $B2H->getline)) {
        print STDOUT $_;
    }

Someday, maybe. Feel free to voice your opinions.

Convert::BinHex::Hex2Bin

A HEX-to-BINary converter. This kind of conversion requires a certain amount of state information; it cannot be done by just calling a simple function repeatedly. Use it like this:

    # Create and use a "translator" object:
    my $H2B = Convert::BinHex->hex2bin;    # get a converter object
    while (<STDIN>) {
        print STDOUT $H2B->next($_);          # convert some more input
    }
    print STDOUT $H2B->done;               # no more input: finish up
    # Re-use the object:
    $H2B->rewind;                 # ready for more action!
    while (<MOREIN>) { ...

On each iteration, "next()" (and "done()") may return either a decent-sized non-empty string (indicating that more converted data is ready for you) or an empty string (indicating that the converter is waiting to amass more input in its private buffers before handing you more stuff to output.

Note that "done()" always converts and hands you whatever is left.

Note that this converter does not find the initial "BinHex version" comment. You have to skip that yourself. It only handles data between the opening and closing ":".

Convert::BinHex::Fork

A fork in a Macintosh file.

    # How to get them...
    $data_fork = $HQX->data;      # get the data fork
    $rsrc_fork = $HQX->resource;  # get the resource fork
    # Make a new fork:
    $FORK = Convert::BinHex::Fork->new(Path => "/tmp/file.data");
    $FORK = Convert::BinHex::Fork->new(Data => $scalar);
    $FORK = Convert::BinHex::Fork->new(Data => \@array_of_scalars);
    # Get/set the length of the data fork:
    $len = $FORK->length;
    $FORK->length(170);        # this overrides the REAL value: be careful!
    # Get/set the path to the underlying data (if in a disk file):
    $path = $FORK->path;
    $FORK->path("/tmp/file.data");
    # Get/set the in-core data itself, which may be a scalar or an arrayref:
    $data = $FORK->data;
    $FORK->data($scalar);
    $FORK->data(\@array_of_scalars);
    # Get/set the CRC:
    $crc = $FORK->crc;
    $FORK->crc($crc);

Design issues

BinHex needs a stateful parser

Unlike its cousins base64 and uuencode, BinHex format is not amenable to being parsed line-by-line. There appears to be no guarantee that lines contain 4n encoded characters... and even if there is one, the BinHex compression algorithm interferes: even when you can decode one line at a time, you can't necessarily decompress a line at a time.

For example: a decoded line ending with the byte "\x90" (the escape or "mark" character) is ambiguous: depending on the next decoded byte, it could mean a literal "\x90" (if the next byte is a "\x00"), or it could mean n-1 more repetitions of the previous character (if the next byte is some nonzero "n").

For this reason, a BinHex parser has to be somewhat stateful: you cannot have code like this:

    #### NO! #### NO! #### NO! #### NO! #### NO! ####
    while (<STDIN>) {            # read HEX
        print hexbin($_);          # convert and write BIN
    }
    

unless something is happening "behind the scenes" to keep track of what was last done. The dangerous thing, however, is that this approach will seem to work, if you only test it on BinHex files which do not use compression and which have 4n HEX characters on each line.

Since we have to be stateful anyway, we use the parser object to keep our state.

We need to be handle large input files

Solutions that demand reading everything into core don't cut it in my book. The first MPEG file that comes along can louse up your whole day. So, there are no size limitations in this module: the data is read on-demand, and filehandles are always an option.

Boy, is this slow!

A lot of the byte-level manipulation that has to go on, particularly the CRC computing (which involves intensive bit-shifting and masking) slows this module down significantly. What is needed perhaps is an optional extension library where the slow pieces can be done more quickly... a Convert::BinHex::CRC, if you will. Volunteers, anyone?

Even considering that, however, it's slower than I'd like. I'm sure many improvements can be made in the HEX-to-BIN end of things. No doubt I'll attempt some as time goes on...

How it works

Since BinHex is a layered format, consisting of...

      A Macintosh file [the "BIN"]...
         Encoded as a structured 8-bit bytestream, then...
            Compressed to reduce duplicate bytes, then...
               Encoded as 7-bit ASCII [the "HEX"]

...there is a layered parsing algorithm to reverse the process. Basically, it works in a similar fashion to stdio's fread():

       0. There is an internal buffer of decompressed (BIN) data,
          initially empty.
       1. Application asks to read() n bytes of data from object
       2. If the buffer is not full enough to accommodate the request:
            2a. The read() method grabs the next available chunk of input
                data (the HEX).
            2b. HEX data is converted and decompressed into as many BIN
                bytes as possible.
            2c. BIN bytes are added to the read() buffer.
            2d. Go back to step 2a. until the buffer is full enough
                or we hit end-of-input.

The conversion-and-decompression algorithms need their own internal buffers and state (since the next input chunk may not contain all the data needed for a complete conversion/decompression operation). These are maintained in the object, so parsing two different input streams simultaneously is possible.