Skip to content

Home

Jump to bottom
System Administrator edited this page May 29, 2018 · 20 revisions

layout: default

Hashit, an hashing application

Hashit is an command line hashing application that supports a large varity of fileformats and hashing algorithms written in python(3) for hashing an verifing files.

File Formats it supports:

  • Default: where hash [size] filename

    • used by md5sum

  • Simple File Verification filename hash [size]

    • popular checksum file format

  • BSD-tag style hashname (filename) = hash [size]

    • output from bsd systems hashing commands

One of the reasons i created this program was because i thougth (from README.rst) that the current naming convention for hashing and file verification tools on debian based systems (they did better on bsd but still) where seriosly inefficent by seperating each hash into a different tool (yes i am aware of that these tools are implemented in C and making them into one program could most likely also cause the same kind of confusion) so hashit ships with all hash functions in one program/command where md5 is the default.

See docs/hashes.md for the diffrent types of hashes supported

Usage

see docs/usage.md

Installing

I would recommend installing it from pypi like this

pip(3) install hashit

But you can also install it from snap (linux only)

snap install hashit (--devmode or --classic is recommend)

And if you're using an debian based distro, you can use my ppa

sudo add-apt-repository ppa:javadsm/javads
sudo apt-get update
sudo apt-get install hashit

Technical Notes

There where a memory leak in the check function which caused wrong hashes to be resolved from my observatitions it has something to do with pythons generators, i fixed it by added an read-all mode (normal) and made it default in the check function and other systems to, but by given the application -m you will enable the generator for the initial hash, which so far hasn't shown any memory leaks but the check function will still be running on the default read-all mode.

The check function works by detect/selecting hashtype and file format then applying the data by creating indexes for diffrent values such as the hash, path and filesize. These indexes can then be used on the lists we create from the line in the file using the file formats parser.

Due to some interface problems with snap, is it not posible to access devices than home and external drives. therefore i would recommend you to install it in --devmode but if you want you can also use classic. (Bypass: use sudo)

Due to the way exclude works it is not needed to use a wildcard '*' to exclude specific extentions for that just do '.ext'. it works by doing:

if 'exclude-string`' in 'path':
    remove_from_list('path')

Extra, see extra for more

The hash-classes are built like so all classes added needs this kind of api due to compatibility. See plugins for more

class hashname:
   name = "hashname"

   def __init__(self, data=b''):
       self.data = data
   
   def update(self, data=b''):
       self.data += data

   def copy(self):
       # pass data on
       return hashname(self.data)

   def digest(self):
       # generate hash
       return GENERATED_HASH

   def hexdigest(self):
       # return as hex
       G_HASH = self.digest()
       return convert_to_hex(G_HASH)

Links

extra how to setup hashit (plugins & config)

pydocmd generated with pydocmd

pydocs generated with pydocs

Notes:

  • I interpet N/a as None At All because i can
  • Now -s --string (input) is the default option so it is posix compatible
  • Detect does not work with shake due to its integration
  • I would not recommend using -S --size because then you will have to specify it everytime you check
  • Detect format benchmarks (using timeit on python3 ubuntu):
    • BSD: 7-5 seconds (10**6 times)
    • SFV: 4.2 seconds (10**6 times)
    • N/A: 2.4 seconds (10**6 times)
Clone this wiki locally