mirror of
https://github.com/skeeto/w64devkit.git
synced 2026-10-24 17:39:18 +03:00
880be39a58
A page for each applet in the kit, plus .so stubs for aliases such as egrep and unxz. Each was written from the pinned busybox-w32 source with the kit's patches and configuration, then checked against the source and by experiment. The NOTES sections cover Windows behavior. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
204 lines
4.2 KiB
Groff
204 lines
4.2 KiB
Groff
.TH BASE64 1 2026-10-19 "w64devkit" "w64devkit Manual"
|
|
.SH NAME
|
|
base64 \- encode or decode base64
|
|
.SH SYNOPSIS
|
|
.B base64
|
|
.RB [ \-w
|
|
.IR col ]
|
|
.RI [ file ]
|
|
.br
|
|
.B base64 \-d
|
|
.RI [ file ]
|
|
.SH DESCRIPTION
|
|
.B base64
|
|
encodes
|
|
.I file
|
|
as base64 on standard output, or with
|
|
.B \-d
|
|
decodes it.
|
|
With no
|
|
.IR file ,
|
|
or when it is
|
|
.BR \- ,
|
|
it reads standard input.
|
|
.PP
|
|
Encoding uses the standard alphabet of RFC 4648: uppercase letters,
|
|
lowercase letters, digits,
|
|
.BR + ,
|
|
and
|
|
.BR / .
|
|
Each 3 input bytes become 4 characters, and the output is padded with
|
|
.B =
|
|
to a multiple of 4 characters.
|
|
The output is broken into lines of
|
|
.I col
|
|
characters, set by
|
|
.BR \-w ,
|
|
and each line, including a shorter last line, ends in LF.
|
|
Empty input produces no output.
|
|
.PP
|
|
Decoding writes the bytes that the input encodes.
|
|
Characters outside the alphabet are skipped, including spaces, line
|
|
breaks, NUL bytes, and the
|
|
.B \-
|
|
and
|
|
.B _
|
|
of the URL-safe alphabet, so base64url input does not decode correctly.
|
|
An
|
|
.B =
|
|
is padding only where it completes a group of 4 characters, and is
|
|
otherwise skipped.
|
|
Decoding continues after padding, so encodings joined with line breaks
|
|
between them decode to the joined data.
|
|
If the input ends inside a group, as unpadded input may,
|
|
.B base64
|
|
writes what it has decoded and fails with
|
|
.BR "truncated input" .
|
|
.SH OPTIONS
|
|
Options may be grouped and may follow
|
|
.IR file .
|
|
The argument of
|
|
.B \-w
|
|
may be attached, as in
|
|
.BR \-w0 ,
|
|
or be the next argument.
|
|
.B \-\-
|
|
ends the options.
|
|
.TP
|
|
.B \-d
|
|
Decode instead of encode.
|
|
.TP
|
|
.B \-i
|
|
Accepted and ignored.
|
|
Decoding always skips characters outside the alphabet.
|
|
.TP
|
|
.BI "\-w " col
|
|
Wrap encoded lines at
|
|
.I col
|
|
characters.
|
|
The default is 76, and 0 disables wrapping, in which case no newline at
|
|
all is written.
|
|
.I col
|
|
is a decimal number from 0 to 2147483647.
|
|
Ignored with
|
|
.BR \-d .
|
|
.TP
|
|
.B \-\-help
|
|
When it is the first argument, print a usage message on standard output
|
|
and exit.
|
|
.SH ENVIRONMENT
|
|
.TP
|
|
.B POSIXLY_CORRECT
|
|
If set, options must precede
|
|
.IR file .
|
|
.PP
|
|
Variables that busybox-w32 reads in every applet, such as those that
|
|
control console output, are described in
|
|
.BR busybox (1).
|
|
.SH EXIT STATUS
|
|
.TP
|
|
.B 0
|
|
Success.
|
|
.TP
|
|
.B 1
|
|
Invalid command line,
|
|
.I file
|
|
could not be opened or read, truncated input, or an error writing
|
|
standard output.
|
|
.TP
|
|
.B 141
|
|
Standard output is a pipe that was closed while
|
|
.B base64
|
|
was writing to it.
|
|
A closed pipe found only when the output is flushed at exit gives status
|
|
1 instead.
|
|
.SH EXAMPLES
|
|
Encode a file:
|
|
.PP
|
|
.RS 4
|
|
.EX
|
|
$ base64 hello.txt
|
|
SGVsbG8sIHdvcmxkIQo=
|
|
.EE
|
|
.RE
|
|
.PP
|
|
Decode a file:
|
|
.PP
|
|
.RS 4
|
|
.EX
|
|
$ base64 \-d hello.b64 >hello.txt
|
|
.EE
|
|
.RE
|
|
.PP
|
|
Encode a file as a single line with no final newline:
|
|
.PP
|
|
.RS 4
|
|
.EX
|
|
$ base64 \-w0 icon.png >icon.txt
|
|
.EE
|
|
.RE
|
|
.PP
|
|
Make a random 32-byte key:
|
|
.PP
|
|
.RS 4
|
|
.EX
|
|
$ head \-c 32 /dev/urandom | base64
|
|
.EE
|
|
.RE
|
|
.SH NOTES
|
|
Differences from GNU base64:
|
|
.IP \(bu 2
|
|
There are no long options, such as
|
|
.B \-\-decode
|
|
or
|
|
.BR \-\-wrap ,
|
|
and no
|
|
.BR \-\-version .
|
|
.IP \(bu 2
|
|
.B \-d
|
|
skips characters outside the alphabet whether or not
|
|
.B \-i
|
|
is given, where GNU fails without
|
|
.BR \-i .
|
|
.PP
|
|
Encoded lines end with LF, not CRLF.
|
|
Input is read in binary mode, so files are encoded byte for byte.
|
|
.PP
|
|
When standard output is a console, decoded data passes through the
|
|
busybox-w32 console layer as text, so binary data does not display
|
|
faithfully.
|
|
Redirect it to a file or a pipe.
|
|
.PP
|
|
Decoding reads the input in pieces that end at a character with code 32
|
|
(space) or less, or after 80 characters, and has two bugs:
|
|
.IP \(bu 2
|
|
A group of 4 characters that continues into the next piece decodes
|
|
wrongly, or fails as truncated input, when a skipped character falls
|
|
inside the group before the break.
|
|
This includes an
|
|
.B =
|
|
that ends a line before the
|
|
.B =
|
|
that starts the next, which an encoder writes for some inputs when it
|
|
wraps at an odd width, such as
|
|
.BR "\-w 3" .
|
|
The last byte of such encoded data is lost without an error.
|
|
.IP \(bu 2
|
|
Characters that follow
|
|
.B =
|
|
padding with no space or line break in between, as in
|
|
.BR YQ==YQ== ,
|
|
can fail as truncated input.
|
|
.PP
|
|
.I file
|
|
may be a Windows path, with a drive letter and either kind of slash, or
|
|
a busybox-w32 device name such as
|
|
.I /dev/null
|
|
or
|
|
.IR /dev/urandom .
|
|
.SH SEE ALSO
|
|
.BR base32 (1),
|
|
.BR busybox (1),
|
|
.BR uudecode (1),
|
|
.BR uuencode (1)
|