postmodern / hexdump.rb Goto Github PK

Refactor the messy numeric formatting and character conversion code into Classes.

Also support dumping floats and doubles, to match the od utility.

Add an option to skip N bytes before starting to read/hexdump the input data. Similar to zhexdum.

String#unpack should be faster and provide more format options (such as floats). Segments will also have to be padded with \0 bytes.

Once 1.0.0 has been released, work on a Crystal port.

Some hex dumping utilities add one more line. It contains only the first offset part with the actual length of the data. Like this:

000003f0  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00000400  00 00 00 00                                       |................|
00000404

It is helpful as it can be immediately seen how long the data is without needing to count the actually present bytes on the last line of the dump.

Just adding here as a suggestion.

Support decoding types:

• char
• uchar
• uint8 / byte
• int8
• uint16
• int16
• uint32
• int32
• uint64
• int64
• float32
• float64

Stop reading input after N bytes. Similar to zhexdump.

Right now, hexdump uses STDOUT instead of $stdout. According to the documentation, $stddout is the current output and STDOUT is the standard output. IRuby updates $stdout to write all output to a 0mq socket, which gets sent to the user. But since hexdump writes to STDOUT, its output is silently swallowed.

GNU hexdump adds an extra space in between eight columns.

00000000  23 20 68 65 78 64 75 6d  70 2e 72 62 0a 0a 5b 21  |# hexdump.rb..[!|

Add a grouping keyword argument to match GNU hexdump output.

Need screenshots of the ANSI examples.

It should be possible to configure the Hexdump object via a block.

Hexdump.hexdump(...) do |hexdump|
  ...
end

Also add attr_accessors to allow writing to certain settings.

I believe that the two occurrences of ascci in benchmarks/hexdump.rb should read ascii.

[For my own education: what should happen in cases like this - should Hexdump.dump or Dumper.initialize check for unrecognised keys in the options hash and issue a warning, raise an exception, or what?]

Do a pre-release of 1.0.0.

Add the ability to post-process rows of a hexdump, similar to zhexdump.

Switch to using require_relative to load other hexdump/ files to improve load-times.

Add support for ANSI styling hexdumps:

• Standard ANSI colors and maybe some font styling?
• Allow styling:
  • index
  • numeric column
  • chars column
  • char separators? (the | characters around the chars column)
  • specific values in both the numeric and chars columns (ex: 0x00 => :red).

Add an option to toggle off the omission of repeating rows.

Implement grouping of characters together, mainly to match the column grouping. Will only be possible when the encoding is ASCII and should disable the feature if the encoding is UTF-8. This would be similar to pwntool's phd command:
https://twitter.com/ebeip90/status/1263255253337518080

Add a String#to_hexdump method similar to zhexdump.