mdcat - render CommonMark Markdown to text terminals
mdcat renders Markdown FILE
s in CommonMark dialect to text terminals with sophisticated formatting.
If no FILE
is given, or if FILE
is '-', it reads from standard input.
If invoked as mdless
automatically use a pager to display the output, see below.
mdcat supports all basic CommonMark syntax plus a few extensions, highlights syntax in code blocks, and shows inline links and even inline images in some terminal programs. In iTerm2 it also adds jump marks for section headings.
See section Terminal support below for a list of supported terminal programs and their features.
To enable formatting extensions such as inline images, mdcat
needs to detect the terminal program, by checking the following environment variables in the given order:
-
$TERM
-
$TERM_PROGRAM
-
$TERMINOLOGY
See section Environment below for a detailed description of each environment variable.
mdcat can render output in a pager; this is the default when run as mdless
.
The environment variables $MDCAT_PAGER
and $PAGER
control the pager used.
Note that common pagers do not support proprietary terminal codes for e.g. image support, so mdcat falls back to pure ANSI formatting when pagination is enabled. In particular this disables all image support which relies on proprietary escape codes.
In iTerm2, Kitty, Terminology and WezTerm mdcat prints inline images. mdcat supports most standard pixel formats by default.
mdcat silently ignores images larger than 100 MiB, under the assumption that images of that size cannot reasonably be rendered in a terminal.
In Terminology mdcat also renders SVG images, using the built-in support of Terminology.
In iTerm2, Kitty and WezTerm mdcat renders SVG images into pixel graphics using the resvg library. Currently this library only supports SVG 1, and only the static subset thereof; see SVG support for details. While this is sufficient for most simple SVG images, complex SVG images may fail to render or render incompletely.
For local SVG files mdcat relies on the file extension to identify SVG images.
For remote images from HTTP(S) URLs mdcat inspects the Content-Type
header to identify SVG images.
- -p
- --paginate
-
Paginate the output of mdcat with a pager like less.
Note: When paginating mdcat only uses standard ANSI formatting and hyperlinks, but no images or other proprietary format codes, because pager programs normally do not support any of these.
This is the default when run as
mdless
. - -P
- --no-pager
-
Do not paginate output.
This is the default when run as
mdcat
. - -c
- --no-colour
-
Disable all colours and other styles.
- --ansi
-
Skip terminal detection and only use ANSI formatting.
- --columns
-
Maximum number of columns to use for text output. Defaults to the size of the underlying terminal.
- -l
- --local
-
Do not access remote resources.
- --fail
-
Fail immediately at the first FILE which fails to read. By default, mdcat continues with the next file.
- --detect-terminal
-
Detect the terminal program, print its name, and exit.
- -h
- --help
-
Show a help message to the user and exit.
- -V
- --version
-
Show the version of mdcat and exit. The long flag also includes information about the builtin features.
mdcat exits with 0 if no error occurred, or 1 otherwise.
If run as mdless
or if --paginate
is given and the pager fails to start mdcat exists with 128.
- TERM
-
mdcat
first checks this variable to identify the terminal program (see section Terminal detection). It understands the following values.-
wezterm
: WezTerm. Note that WezTerm sets$TERM
toxterm-256color
by default, and only useswezterm
for$TERM
if explicitly configured to do so. -
xterm-kitty
: KittyFor all other values
mdcat
proceeds to check$TERM_PROGRAM
.
-
- TERM_PROGRAM
-
If
$TERM
does not conclusively identify the terminal programmdcat
checks this variable next. It understands the following values:-
iTerm.app
: iTerm2 -
WezTerm
: WezTermFor all other values
mdcat
proceeds to check$TERMINOLOGY
.
-
- TERMINOLOGY
-
If this variable is
1
, mdcat assumes that the terminal is Terminology.Otherwise
mdcat
ends terminal detection and assumes that the terminal is only capable of standard ANSI formatting. - COLUMNS
-
The number of character columns on screen.
mdcat only uses this variable if it fails to query the size from the underlying terminal.
- ROWS
-
The number of character rows on screen.
mdcat only uses this variable if it fails to query the size from the underlying terminal.
- MDCAT_PAGER
-
The pager program to use for
mdless
or if--paginate
is given.The pager program must support basic ANSI formatting sequences, like e.g.
less -r
.The value of this variable is subject to shell-like word-splitting. It is not subject to any kind of expansion or substitution (e.g. parameter expansion, process subsitution, etc.).
If set to an empty value, mdcat completely disables pagination.
- PAGER
-
The pager program to use if
$MDCAT_PAGER
is unset.Subject to the same rules as
$MDCAT_PAGER
.If both
$PAGER
and$MDCAT_PAGER
are unset useless -r
as pager. - http_proxy
- https_proxy
- HTTPS_PROXY
- all_proxy
- ALL_PROXY
-
Proxies for HTTP, HTTPS, or both protocols, to use when fetching images.
Each variable provides the proxy for the corresponding protocol as URL, e.g.
http://proxy.example.com:3128
.The lowercase name takes precedence; note that
$http_proxy
deliberately has no uppercase variant. - no_proxy
- NO_PROXY
-
A comma-separated list of host/domain names or IP address not to use a proxy for.
Matches partial hostnames (e.g.
example.org
also disables proxy forwww.example.org
), but always at name boundaries. - MDCAT_LOG
-
Directives to configure output of tracing information.
See https://docs.rs/tracing-subscriber/latest/tracing_subscriber/struct.EnvFilter.html#directives for syntax details; use
MDCAT_LOG=trace
for complete debugging information, andMDCAT_LOG=mdcat::render=trace
to trace rendering only.
mdcat supports version 0.29 of the CommonMark Spec, plus Task lists and strikethrough, through pulldown-cmark.
mdcat does not yet support footnotes and tables. mdcat parses HTML blocks and inline tags but does not apply special rendering; it prints HTML as is.
Unless --no-colour
is given, mdcat translates CommonMark text into ANSI formatted text, with standard SGR formatting codes and hyperlinks.
It uses bold (SGR 1), italic (SGR 3) and strikethrough (SGR 9) formatting, and the standard 4-bit color sequences, as well as OSC 8 for hyperlinks.
It does not use 8-bit or 24-bit color sequences, though this may change in future releases.
Additionally, it uses proprietary escape code if it detects one of the following terminal emulators (see sections Terminal detection and Environment for details):
-
iTerm2: Inline images (iTerm2 protocol) and Marks.
-
Kitty: Inline images (Kitty Graphics protocol).
-
Terminology: Inline images (terminology protocol).
-
WezTerm: Inline images (iTerm2 protocol, see above).
Please report bugs to https://github.com/swsnr/mdcat/issues.
Currently mdcat does not yet wrap text to column limits, and does not provide means to customize styles and colours.
- mdcat hello - world
-
Render markdown in
hello
, then from standard input, then fromworld
. - mdless hello
-
Render markdown from
mdless
through a pager.
Copyright Sebastian Wiesner <[email protected]> and contributors
Binaries are subject to the terms of the Mozilla Public License, v. 2.0. See https://github.com/swsnr/mdcat/blob/main/LICENSE.
Most of the source is subject to the terms of the Mozilla Public License, v. 2.0, unless otherwise noted; some files are subject to the terms of the Apache 2.0 license, see http://www.apache.org/licenses/LICENSE-2.0.