Bytecode

NAME

MoarVM::Bytecode - Provide introspection into MoarVM bytecode

SYNOPSIS


use MoarVM::Bytecode;

my $M = MoarVM::Bytecode.new($filename);  # or letter or IO or Blob

say $M.hll-name;     # most likely "Raku"

say $M.strings[99];  # the 100th string on the string heap

DESCRIPTION

MoarVM::Bytecode provides an object oriented interface to the MoarVM bytecode format, based on the information provided in docs/bytecode.markdown.

CLASS METHODS

new


my $M = MoarVM::Bytecode.new("c");           # the 6.c setting

my $M = MoarVM::Bytecode.new("foo/bar");     # file as string

my $M = MoarVM::Bytecode.new($filename.IO);  # path as IO object

my $M = MoarVM::Bytecode.new($buf);          # a Buf object

Create an instance of the MoarVM::Bytecode object from a letter (assumed to be a Raku version letter such as "c", "d" or "e"), a filename, an IO::Path or a Buf/Blob object.

blib


my $blib = MoarVM::Bytecode.blib;

Returns an IO::Path of the "blib" directory of the installation of the currently running rakudo executable.

setting


my $setting = MoarVM::Bytecode.setting;

my $setting = MoarVM::Bytecode.setting("d");

Returns an IO::Path of the bytecode file of the given setting letter. Assumes the currently lowest supported setting by default.

INSTANCE METHODS

hexdump


say $M.hexdump($M.string-heap-offset);  # defaults to 256

say $M.hexdump($M.string-heap-offset, 1024);

Returns a hexdump representation of the bytecode from the given byte offset for the given number of bytes (256 by default).

hll-name


say $M.hll-name;     # most likely "Raku"

Returns the HLL language name for this bytecode. Most likely "Raku", or "NQP".

strings


.say for $M.strings[^10];  # The first 10 strings on the string heap

Returns an object that serves as a Positional for all of the strings on the string heap.

version


say $M.version;     # most likely 7

Returns the numeric version of this bytecode. Most likely "7".

PRIMITIVES

bytecode


my $b = $M.bytecode;

Returns the Buf with the bytecode.

slice


dd $M.slice(0, 8).chrs;     # "MOARVM\r\n"

Returns a List of unsigned 32-bit integers from the given offset and number of bytes. Basically a shortcut for $M,bytecode[$offset ..^ $offset + $bytes]. The number of bytes defaults to 256 if not specified.

subbuf


dd $M.subbuf(0, 8).decode;  # "MOARVM\r\n"

Calls subbuf on the bytecode and returns the result. Basically a shortcut for $M.bytecode.subbuf(...).

uint32


my $i = $M.uint32($M.string-heap-offset);

Returns the unsigned 32-bit integer value at the given offset in the bytecode.

HEADER SHORTCUTS

The following methods provide shortcuts to the values in the bytecode header. They are explained in the MoarVM documentation.

sc-dependencies-offset, sc-dependencies-entries, extension-ops-offset, extension-ops-entries, frames-data-offset, frames-data-entries, callsites-data-offset, callsites-data-entries, string-heap-offset, string-heap-entries, sc-data-offset, sc-data-length, bytecode-offset, bytecode-length, annotation-offset, annotation-length, main-entry-frame-index, library-load-frame-index, deserialization-frame-index

AUTHOR

Elizabeth Mattijsen <[email protected]>

COPYRIGHT AND LICENSE

Source can be located at: https://github.com/lizmat/MoarVM-Bytecode . Comments and Pull Requests are welcome.

If you like this module, or what I’m doing more generally, committing to a small sponsorship would mean a great deal to me!

This library is free software; you can redistribute it and/or modify it under the Artistic License 2.0.