hashmash.1

.TH HASHMASH 1
.SH NAME
hashmash \- identify, and optionally randomize, cryptographic hashes in text files
.SH SYNOPSIS
.B hashmash
[\fB\-r\fR]
.IR file ...
.SH DESCRIPTION
.B hashmash
runs like \fBcat(1)\fR, over \fIstdin\fR or the specified input file(s). By default it
highlights the substrings that look like cryptographic hashes, GUIDs, etc. Optionally
it can also randomize all such strings. This can be handy for generating documentation
without leaking secrets.
.PP
The identification of hash-like substrings doesn't come from a fixed set of patterns.
Rather, trigrams are used to give \fBhashmash\fR some idea of what human words usually
look like, and it considers a hash to be anything that deviates wildly from that.
.PP
Since the trigram distribution it knows about comes from English, it may hit more false
positives with other languages.
.PP
Highlighting is done using ANSI colour codes, if (and only if) \fIstdout\fR is a terminal.
.SH OPTIONS
.TP
.BR \-r ", " \-\-replace\fR
Replace identified cryptographic hashes with a random string. This preserves the length
of the string, and also the class of each character (eg "digit" or "lowercase letter").
For example,
.CW
W6B43240-ad76s==62231DH00
might get randomized into
.CW
Q1L83073-mn13q==03510AP62
\&.
.TP
.BR \-i ", " \-\-in\-place\fR
Do a \fB--replace\fR, but rather than just send the updated output to stdout, actually
edit the input files destructively in-place.
.TP
.BR \-h ", " \-\-help\fR
Show help summary.
.TP
.BR \-V ", " \-\-version\fR
Show version number.
.TP
.BR \-d ", " \-\-debug\fR
An assist during development: highlight in green those substrings that pass the crude
pre-classifier, but which are not classified as hashes by the tangram algorithm.