Conway-Markdown (CMD) v4.6.0

Conway-Markdown (CMD) is:

A replacement-driven markup language inspired by Markdown.
A demonstration of the folly of throwing regex at a parsing problem.
The result of someone joking that the filenames would look like Windows executables from the 90s.
Implemented in Python 3.{whatever Debian stable is at}.
Licensed under "MIT No Attribution" (MIT-0).

Installation and usage
Authoring CMD files
Standard CMD replacement rules
- Standard queued replacements
- Standard unqueued replacements
CMD replacement rule syntax
CMD placeholders
CMD attribute specifications
Repository links

Installation and usage

Clone the repository of the Python implementation:
```
$ git clone https://github.com/conway-markdown/conway-markdown
```
Since cmd.py is a shitty single-file script, it will not be turned into a proper Python package.

Linux terminals, macOS Terminal, Git BASH for Windows

Make an alias for cmd.py in whatever dotfile you configure your aliases in:
```
alias cmd='path/to/cmd.py'
```

Invoke the alias to convert a CMD file to HTML:

$ cmd [-h] [-v] [-x] [file.cmd ...]

Convert Conway-Markdown (CMD) to HTML.

positional arguments:
  file.cmd       Name of CMD file to be converted. Abbreviate as `file` or
                 `file.` for increased productivity. Omit to convert all CMD
                 files under the working directory.

optional arguments:
  -h, --help     show this help message and exit
  -v, --version  show program's version number and exit
  -x, --verbose  run in verbose mode (prints every replacement applied)

Windows Command Prompt

Add the folder containing cmd.py to the %PATH% variable

Invoke cmd.py to convert a CMD file to HTML:

> cmd.py [-h] [-v] [-x] [file.cmd ...]

Convert Conway-Markdown (CMD) to HTML.

positional arguments:
  file.cmd       Name of CMD file to be converted. Abbreviate as `file` or
                 `file.` for increased productivity. Omit to convert all CMD
                 files under the working directory.

optional arguments:
  -h, --help     show this help message and exit
  -v, --version  show program's version number and exit
  -x, --verbose  run in verbose mode (prints every replacement applied)

WARNING: on Windows, be careful not to run any .cmd files by accident; they might break your computer. God save!

Authoring CMD files

CMD files are parsed thus:

«replacement rules»
«delimiter»
«main content»

«replacement rules» are user-defined CMD replacement rules that will be used in addition to the standard CMD replacement rules.
«delimiter» is the first occurrence of 3-or-more percent signs on its own line. (If no «delimiter» is found in the file, the whole file is parsed as «main content».)
«main content» is the CMD content that will be converted to HTML according to the standard and user-defined CMD replacement rules.

In the implementation:

An empty replacement queue is initialised.
STANDARD_RULES in cmd.py are parsed, and CMD replacement rules are added to the replacement queue accordingly.
«replacement rules» in the CMD file are parsed, and CMD replacement rules are added or inserted into the replacement queue accordingly.
The CMD replacement rules in the replacement queue are applied sequentially to «main content» to convert it to HTML.

To get started with writing «main content», read the listing of standard CMD replacement rules.

To learn about writing user-defined «replacement rules», read about CMD replacement rule syntax.

Standard CMD replacement rules

This section lists the standard CMD replacement rules as defined by the constant string STANDARD_RULES in cmd.py.

Some replacement rules are queued, in that they appear explicitly in the replacement queue.

Other replacement rules are unqueued, in that they do not appear explicitly in the replacement queue. However, they might be called by queued replacements.

Standard queued replacements

0. `#placeholder-markers`

Definition

PlaceholderMarkerReplacement: #placeholder-markers
- queue_position: ROOT

Description

Replaces occurrences of the placeholder marker «U+F8FF» with a placeholder so that the occurrences will not be confounding.

1. `#literals`

Definition

ExtensibleFenceReplacement: #literals
- queue_position: AFTER #placeholder-markers
- syntax_type: INLINE
- allowed_flags:
    u=KEEP_HTML_UNESCAPED
    i=KEEP_INDENTED
    w=REDUCE_WHITESPACE
- prologue_delimiter: <
- extensible_delimiter: `
- content_replacements:
    #escape-html
    #de-indent
    #trim-whitespace
    #reduce-whitespace
    #placeholder-protect
- epilogue_delimiter: >

Syntax

<` «content» `>

«flags»<` «content» `>

«flags»:
- u: keep HTML unescaped (do not apply #escape-html)
- i: keep indented (do not apply #de-indent)
- w: reduce whitespace (apply #reduce-whitespace)
The number of backticks ` may be increased arbitrarily.

Description

Preserves «content» literally.

Examples

Write a literal string:
- CMD: <` Foo `>
- HTML: Foo
- Rendered: Foo
Keep HTML unescaped:
- CMD: u<` Foo `>
- HTML: Foo
- Rendered: Foo
Increase the number of backticks to include content that matches the #literals syntax itself:
- CMD: <`` Literally <`literal`>. ``>
- HTML: Literally <`literal`>.
- Rendered: Literally <`literal`>.

2. `#display-code`

Definition

ExtensibleFenceReplacement: #display-code
- queue_position: AFTER #literals
- syntax_type: BLOCK
- allowed_flags:
    u=KEEP_HTML_UNESCAPED
    i=KEEP_INDENTED
    w=REDUCE_WHITESPACE
- extensible_delimiter: ``
- attribute_specifications: EMPTY
- content_replacements:
    #escape-html
    #de-indent
    #reduce-whitespace
    #code-tag-wrap
    #placeholder-protect
- tag_name: pre

Syntax

``
  «content»
``

``{«attribute specifications»}
  «content»
``

«flags»``
  «content»
``

«flags»``{«attribute specifications»}
  «content»
``

«flags»:
- u: keep HTML unescaped (do not apply #escape-html)
- i: keep indented (do not apply #de-indent)
- w: reduce whitespace (apply #reduce-whitespace)
The number of backticks ` may be increased arbitrarily.
«attribute specifications»: see CMD attribute specifications.

Description

Produces (pre-formatted) display code:

<pre«attribute sequence»><code>«content»
</code></pre>

Examples

Display code:

CMD:

``{.java}
  for (int index = 0; index < count; index++)
  {
    // etc. etc.
  }
``

HTML:

<pre class="java"><code>for (int index = 0; index &lt; count; index++)
{
  // etc. etc.
}
</code></pre>

Rendered:

for (int index = 0; index < count; index++)
{
  // etc. etc.
}

Use #literals with flag u to inject HTML:

CMD:

``
  Injection of <b> element u<` <b>here</b> `>.
``

HTML:

<pre><code>Injection of &lt;b&gt; element <b>here</b>.
</code></pre>

Rendered:
```
Injection of element here.
```

3. `#comments`

Definition

RegexDictionaryReplacement: #comments
- queue_position: AFTER #display-code
* [^\S\n]*
  [<]
    (?P<hashes> [#]+ )
      [\s\S]*?
    (?P=hashes)
  [>]
    -->

Syntax

<# «content» #>

The number of hashes # may be increased arbitrarily.

Description

Remove CMD comments, along with all preceding whitespace.

Examples

#display-code prevails over #comments:

CMD:

``
  Before.
  <# This be a comment. #>
  After.
``

HTML:

<pre><code>Before.
&lt;# This be a comment. #&gt;
After.
</code></pre>

Rendered:

Before.
<# This be a comment. #>
After.

#comments prevail over #inline-code:
- CMD: `Before. <# This be a comment. #> After.`
- HTML: <code>Before. After.</code>
- Rendered: Before. After.
Use #literals to make #inline-code prevail over #comments:
- CMD: ` <`Before. <# This be a comment. #> After.`> `
- HTML: <code>Before. <# This be a comment. #> After.</code>
- Rendered: Before. <# This be a comment. #> After.

4. `#divisions`

Definition

ExtensibleFenceReplacement: #divisions
- queue_position: AFTER #comments
- syntax_type: BLOCK
- extensible_delimiter: ||
- attribute_specifications: EMPTY
- content_replacements:
    #divisions
    #prepend-newline
- tag_name: div

Syntax

||
  «content»
||

||{«attribute specifications»}
  «content»
||

The number of pipes | may be increased arbitrarily.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a division:

<div«attribute sequence»>
«content»
</div>

Examples

Division:

CMD:

||{#test-div-id .test-div-class}
  This be a division.
||

HTML:

<div id="test-div-id" class="test-div-class">
This be a division.
</div>

More pipes for nested divisions:

CMD:

||||{.outer}
  ||{.inner}
    This be a division.
  ||
||||

HTML:

<div class="outer">
<div class="inner">
This be a division.
</div>
</div>

5. `#blockquotes`

Definition

ExtensibleFenceReplacement: #blockquotes
- queue_position: AFTER #divisions
- syntax_type: BLOCK
- extensible_delimiter: ""
- attribute_specifications: EMPTY
- content_replacements:
    #blockquotes
    #prepend-newline
- tag_name: blockquote

Syntax

""
  «content»
""

""{«attribute specifications»}
  «content»
""

The number of double-quotes " may be increased arbitrarily.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a blockquote:

<blockquote«attribute sequence»>
«content»
</blockquote>

Examples

Blockquote:

CMD:

""{#test-blockquote-id .test-blockquote-class}
  This be a blockquote.
""

HTML:

<blockquote id="test-blockquote-id" class="test-blockquote-class">
This be a blockquote.
</blockquote>

6. `#unordered-lists`

Definition

ExtensibleFenceReplacement: #unordered-lists
- queue_position: AFTER #blockquotes
- syntax_type: BLOCK
- extensible_delimiter: ==
- attribute_specifications: EMPTY
- content_replacements:
    #unordered-lists
    #unordered-list-items
    #prepend-newline
- tag_name: ul

Syntax

==
- «item»
«...»
==

=={«attribute specifications»}
-{«attribute specifications»} «item»
«...»
==

The number of equals signs = may be increased arbitrarily.
The item delimiter may be -, +, or *, see #unordered-list-items.
«attribute specifications»: see CMD attribute specifications.

Description

Produces an unordered list:

<ul«attribute sequence»>
<li>
«item»
</li>
«...»
</ul>

Examples

Nested list:

CMD:

======
* A
  ===={style="background: yellow"}
  - A1
  - A2
  ====
- B
  ====
  +{style="color: purple"} B1
  - B2
  ====
======

Rendered:
- A
  - A1
  - A2
- B
  - B1
  - B2

7. `#ordered-lists`

Definition

ExtensibleFenceReplacement: #ordered-lists
- queue_position: AFTER #unordered-lists
- syntax_type: BLOCK
- extensible_delimiter: ++
- attribute_specifications: EMPTY
- content_replacements:
    #ordered-lists
    #ordered-list-items
    #prepend-newline
- tag_name: ol

Syntax

++
1. «item»
«...»
++

++{«attribute specifications»}
1.{«attribute specifications»} «item»
«...»
++

The number of plus signs + may be increased arbitrarily.
The item delimiter may be any run of digits followed by a full stop, see #ordered-list-items.
«attribute specifications»: see CMD attribute specifications.

Description

Produces an ordered list:

<ol«attribute sequence»>
<li>
«item»
</li>
«...»
</ol>

Examples

Any run of digits can be used in the item delimiter:
- CMD:
```
++
1. A
2. B
3. C
0. D
99999999. E
++
```
- Rendered:
  1. A
  2. B
  3. C
  4. D
  5. E

Nested ordered lists:

CMD:

++++
1. This shall be respected.
2. This shall be read aloud if:
  ++{type="a"}
  1. I say so;
  2. I think so; or
  3.{style="color: purple"} Pigs fly.
  ++
++++

Rendered:
1. This shall be respected.
2. This shall be read aloud if:
  1. I say so;
  2. I think so; or
  3. Pigs fly.

Zero-based indexing:
- CMD:
```
++{start=0}
0. Nil
1. One
++
```
- Rendered:
  1. Nil
  2. One

8. `#tables`

Definition

ExtensibleFenceReplacement: #tables
- queue_position: AFTER #ordered-lists
- syntax_type: BLOCK
- extensible_delimiter: ''
- attribute_specifications: EMPTY
- content_replacements:
    #tables
    #table-head
    #table-body
    #table-foot
    #table-rows
    #prepend-newline
- tag_name: table

Syntax

''
|^
  //
    ; «item»
    , «item»
    «...»
  «...»
|:
  «...»
|_
  «...»
''

''
//
  ; «item»
  , «item»
  «...»
«...»
''

''{«attribute specifications»}
|^{«attribute specifications»}
  //{«attribute specifications»}
    ;{«attribute specifications»} «item»
    ,{«attribute specifications»} «item»
    «...»
  «...»
|:{«attribute specifications»}
  «...»
|_{«attribute specifications»}
  «...»
''

The number of single-quotes ' may be increased arbitrarily.
Table parts:
- |^ for table head, see #table-head.
- |: for table body, see #table-body.
- |: for table foot, see #table-foot.
Table cells:
- ; for table headers, see #table-headers.
- , for table data, see #table-data.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a table:

<table«attribute sequence»>
«...»
</table>

Examples

Table without parts:

CMD:
```
''
//
  ; A
  , 1
//
  ; B
  , 2
''
```

HTML:

<table>
<tr>
<th>A</th>
<td>1</td>
</tr>
<tr>
<th>B</th>
<td>2</td>
</tr>
</table>

Rendered:

A 1

B 2

A	1
B	2

Table with head and body:

CMD:

''
|^
  //
    ; A
    ; B
|:
  //
    , 1
    , 2
  //
    , First
    , Second
''

HTML:

<table>
<thead>
<tr>
<th>A</th>
<th>B</th>
</tr>
</thead>
<tbody>
<tr>
<td>1</td>
<td>2</td>
</tr>
<tr>
<td>First</td>
<td>Second</td>
</tr>
</tbody>
</table>

Rendered:

A B

1 2

First Second

A	B
1	2
First	Second

Cell merging

CMD:

''
//
  ,{c3} 3*1
  , 14
//
  , 21
  , 22
  , 23
  , 24
//
  ,{rowspan=2 colspan="2"} 2*2
  , 33
  ,{r2} 2*1
//
  , 43
''

Rendered

3*1 14

21 22 23 24

2*2 33 2*1

43

9. `#headings`

Definition

HeadingReplacement: #headings
- queue_position: AFTER #tables
- attribute_specifications: EMPTY

Syntax

# «content»

#{«attribute specifications»} «content»

The number of hashes # may be between 1 and 6.
«content» may be optionally closed by hashes.
For continuation of «content», indent the continuation more than the leading hashes.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a heading:

<h«hash count»>«content»</h«hash count»>

Examples

Empty headings:

CMD:
```
#
##
###
####
#####
######
```

HTML:

<h1></h1>
<h2></h2>
<h3></h3>
<h4></h4>
<h5></h5>
<h6></h6>

Non-empty headings require whitespace after the hashes:
- CMD:
```
## OK
#lacks-whitespace
```
- HTML:
```
<h2>OK</h2>
#lacks-whitespace
```

Optional closing hashes:

CMD:

## Closed ##
## Fewer closing hashes is OK #
## More closing hashes is OK ###

HTML:

<h2>Closed</h2>
<h2>Fewer closing hashes is OK</h2>
<h2>More closing hashes is OK</h2>

Continuation:

CMD:

##{.interesting-heading-class}
  This heading is so long, I have used continuation.
  Second continuation line.
This line is not a continuation due to insufficient indentation.

HTML:

<h2 class="interesting-heading-class">
This heading is so long, I have used continuation.
Second continuation line.</h2>
This line is not a continuation due to insufficient indentation.

10. `#paragraphs`

Definition

ExtensibleFenceReplacement: #paragraphs
- queue_position: AFTER #headings
- syntax_type: BLOCK
- extensible_delimiter: --
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS
- content_replacements:
    #prepend-newline
- tag_name: p

Syntax

--
«content»
--

--{«attribute specifications»}
«content»
--

The number of hyphens - may be increased arbitrarily.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a paragraph:

<p«attribute sequence»>
«content»
</p>

Examples

Paragraph:

CMD:
```
--
The quick brown fox etc. etc.
--
```
HTML:
```

The quick brown fox etc. etc.

```
Rendered:
The quick brown fox etc. etc.

Paragraphs cannot be nested:

CMD:

----
Before.
--
Not a nested paragraph.
--
After.
----

HTML:

<p>
Before.
--
Not a nested paragraph.
--
After.
</p>

Rendered:
Before. -- Not a nested paragraph. -- After.

11. `#inline-code`

Definition

ExtensibleFenceReplacement: #inline-code
- queue_position: AFTER #paragraphs
- syntax_type: INLINE
- allowed_flags:
    u=KEEP_HTML_UNESCAPED
    i=KEEP_INDENTED
    w=REDUCE_WHITESPACE
- extensible_delimiter: `
- attribute_specifications: EMPTY
- prohibited_content: ANCHORED_BLOCKS
- content_replacements:
    #escape-html
    #de-indent
    #trim-whitespace
    #reduce-whitespace
    #placeholder-protect
- tag_name: code

Syntax

` «content» `

`{«attribute specifications»} «content» `

«flags»` «content» `

«flags»`{«attribute specifications»} «content» `

«flags»:
- u: keep HTML unescaped (do not apply #escape-html)
- i: keep indented (do not apply #de-indent)
- w: reduce whitespace (apply #reduce-whitespace)
The number of backticks ` may be increased arbitrarily.
«attribute specifications»: see CMD attribute specifications.

Description

Produces inline code:

<code«attribute sequence»>«content»</code>

Examples

Code:
- CMD: ` `
- HTML: <code> </code>
- Rendered:  
Code containing backticks:
- CMD: ``A `backticked` word.``
- HTML: <code>A `backticked` word.</code>
- Rendered: A `backticked` word.
Use #literals with flag u to inject HTML:
- CMD: `Some u<`bold`> code`
- HTML: <code>Some bold code</code>
- Rendered: Some bold code

12. `#boilerplate`

Definition

RegexDictionaryReplacement: #boilerplate
- queue_position: AFTER #inline-code
* \A -->
    <!DOCTYPE html>
    <html lang="%lang">
      <head>
        <meta charset="utf-8">
        %head-elements-before-viewport
        <meta name="viewport" content="%viewport-content">
        %head-elements-after-viewport
        <title>%title</title>
        <style>
          %styles
        </style>
      </head>
      <body>\n
* \Z -->
      </body>
    </html>\n

Description

Wraps content in the HTML5 boilerplate:

<!DOCTYPE html> through <body> at the start
</body></html> at the end

The default boilerplate properties %lang, %head-elements-before-viewport, %viewport-content, %head-elements-after-viewport, %title, and %styles are set in #boilerplate-properties.

13. `#boilerplate-properties`

Definition

OrdinaryDictionaryReplacement: #boilerplate-properties
- queue_position: AFTER #boilerplate
* %lang --> en
* %head-elements-before-viewport -->
* %viewport-content --> width=device-width, initial-scale=1
* %head-elements-after-viewport -->
* %title --> Title
* %styles -->

Description

Makes replacements for the default boilerplate properties:

Property	Default value
`%lang`	`en`
`%head-elements-before-viewport`	(empty)
`%viewport-content`	`width=device-width, initial-scale=1`
`%head-elements-after-viewport`	(empty)
`%title`	`Title`
`%styles`	(empty)

Examples

Defaults:

CMD:

# %title
This document hath `lang` equal to <code>%lang</code>.

HTML (including boilerplate):

<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Title</title>
</head>
<body>
<h1>Title</h1>
This document hath <code>lang</code> equal to <code>en</code>.
</body>
</html>

Override the defaults:

CMD:

OrdinaryDictionaryReplacement: #.boilerplate-properties-override
- queue_position: BEFORE #boilerplate-properties
* %lang --> en-AU
* %head-elements-before-viewport --> <meta name="author" content="Me">
* %title --> Overridden title

%%%

# %title
This document hath `lang` equal to <code>%lang</code>.

HTML (including boilerplate):

<!DOCTYPE html>
<html lang="en-AU">
<head>
<meta charset="utf-8">
<meta name="author" content="Me">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Overridden title</title>
</head>
<body>
<h1>Overridden title</h1>
This document hath <code>lang</code> equal to <code>en-AU</code>.
</body>
</html>

14. `#cmd-properties`

Definition

OrdinaryDictionaryReplacement: #cmd-properties
- queue_position: AFTER #boilerplate-properties
* %cmd-version --> CMD_VERSION
* %cmd-name --> CMD_NAME
* %cmd-basename --> CMD_BASENAME
* %clean-url --> CLEAN_URL
- concluding_replacements:
    #placeholder-protect

Description

Makes replacements for CMD properties:

Property	Description
`%cmd-version`	`__version__` in `cmd.py` (currently `4.6.0`)
`%cmd-name`	CMD file name, relative to working directory, without extension
`%cmd-basename`	CMD file name, without path, without extension
`%clean-url`	`%cmd-name`, with `%cmd-basename` removed if it equals `index`

15. `#boilerplate-protect`

Definition

RegexDictionaryReplacement: #boilerplate-protect
- queue_position: AFTER #cmd-properties
* <style>[\s]*?</style>[\s]* -->
* <style>[\s\S]*?</style> --> \g<0>
* <head>[\s\S]*?</head> --> \g<0>
- concluding_replacements:
    #reduce-whitespace
    #placeholder-protect

Description

Protects boilerplate elements:

Removes empty <style> element
Protects <style> element
Protects <head> element

16. `#backslash-escapes`

Definition

OrdinaryDictionaryReplacement: #backslash-escapes
- queue_position: AFTER #boilerplate-protect
* \\ --> \
* \" --> "
* \# --> #
* \& --> &amp;
* \' --> '
* \( --> (
* \) --> )
* \* --> *
* \< --> &lt;
* \> --> &gt;
* \[ --> [
* \] --> ]
* \_ --> _
* \{ --> {
* \| --> |
* \} --> }
* "\ " --> " "
* \t --> "	"
- concluding_replacements:
    #placeholder-protect

Description

Applies backslash escapes:

Escaped	Unescaped
`\\`	`\`
`\"`	`"`
`\#`	`#`
`\&`	`&`
`\'`	`'`
`\(`	`(`
`\)`	`)`
`\*`	`*`
`\<`	`<`
`\>`	`>`
`\[`	`[`
`\]`	`]`
`\_`	`_`
`\{`	`{`
`\\|`	`\|`
`\}`	`}`
`\`	(space)
`\t`	(tab)

17. `#backslash-continuations`

Definition

RegexDictionaryReplacement: #backslash-continuations
- queue_position: AFTER #backslash-escapes
* \\ \n [^\S\n]* -->

Description

Applies backslash continuation.

18. `#reference-definitions`

Definition

ReferenceDefinitionReplacement: #reference-definitions
- queue_position: AFTER #backslash-continuations
- attribute_specifications: EMPTY

Syntax

[«label»]: «uri»
[«label»]: <«uri»>

[«label»]: «...» "«title»"
[«label»]: «...» '«title»'

[«label»]{«attribute specifications»}: «...»

For continuation of «uri» or «title», indent the continuation more than the leading square bracket.
«attribute specifications»: see CMD attribute specifications.

Description

Defines a reference, to be used by #referenced-images or #referenced-links.

19. `#specified-images`

Definition

SpecifiedImageReplacement: #specified-images
- queue_position: AFTER #reference-definitions
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS

Syntax

![«alt text»](«src»)
![«alt text»](<«src»>)

![«alt text»](«...» "«title»")
![«alt text»](«...» '«title»')

![«alt text»]{«attribute specifications»}(«...»)

«attribute specifications»: see CMD attribute specifications.

Description

Produces an image:

<img«attribute sequence»>

Here, «attribute sequence» is the sequence of attributes built from

«alt text»
«src»
«title»
«attribute specifications»

parsed in that order.

Examples

Basic usage:

CMD:

![Rembrandt painting: The Anatomy Lesson of Dr Nicolaes Tulp.](rembrandt-anatomy.jpg)

HTML:

<img alt="Rembrandt painting: The Anatomy Lesson of Dr Nicolaes Tulp." src="rembrandt-anatomy.jpg">

Rendered:

Set width:

CMD:

![Rembrandt painting: The Anatomy Lesson of Dr Nicolaes Tulp.]{w=120}(rembrandt-anatomy.jpg)

HTML:

<img alt="Rembrandt painting: The Anatomy Lesson of Dr Nicolaes Tulp." src="rembrandt-anatomy.jpg" width="120">

Rendered:

Empty alt text for decorative images, or images where adjacent text already describes the image:
- CMD:
```
![](/favicon-16x16.png) Conway-Markdown is dumb.
```
- HTML:
```
<img alt="" src="/favicon-16x16.png"> Conway-Markdown is dumb.
```
- Rendered:
 Conway-Markdown is dumb.

20. `#referenced-images`

Definition

ReferencedImageReplacement: #referenced-images
- queue_position: AFTER #specified-images
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS

Syntax

![«alt text»][«label»]
![«alt text»]

![«alt text»]{«attribute specifications»}[«label»]
![«alt text»]{«attribute specifications»}

«label»: must correspond to a defined reference definition, up to case. If «label» is omitted, «alt text» is used in its stead.
«attribute specifications»: see CMD attribute specifications.

Description

Produces an image:

<img«attribute sequence»>

Here, «attribute sequence» is the sequence of attributes built from

«alt text»
«src» equal to «uri» of the reference definition defined for «label»
«title» of the reference definition defined for «label»
«attribute specifications» of the reference definition defined for «label»
«attribute specifications»

parsed in that order.

Examples

Basic usage:

CMD:

[moses]: rembrandt-moses.jpg

![Rembrandt painting: Moses Breaking the Tablets of the Law.][moses]

HTML:

<img alt="Rembrandt painting: Moses Breaking the Tablets of the Law." src="rembrandt-moses.jpg">

Rendered:

Use «alt text» for «label»:

CMD:

[Conway-Markdown logo.]: /favicon-32x32.png "Conway-Markdown is dumb."

![Conway-Markdown logo.]

HTML:

<img alt="Conway-Markdown logo." src="/favicon-32x32.png" title="Conway-Markdown is dumb.">

Rendered:

«label» is case-insensitive:

CMD:

[image-label-case]: insensitive.png

[Hooray.][image-label-case]
[Hooray.][image-label-CASE]
[Hooray.][ImAGe-laBEl-CAsE]

HTML:

<a href="insensitive.png">Hooray.</a>
<a href="insensitive.png">Hooray.</a>
<a href="insensitive.png">Hooray.</a>

Later reference definitions prevail:

CMD:

[image-label]{.class1}: file1.png
[image-label]{.class2}: file2.png

[Second definition wins.][image-label]

HTML:

<a href="file2.png" class="class2">Second definition wins.</a>

21. `#explicit-links`

Definition

ExplicitLinkReplacement: #explicit-links
- queue_position: AFTER #referenced-images
- allowed_flags:
    b=ANGLE_BRACKET_WRAP
    s=SUPPRESS_SCHEME
- attribute_specifications: EMPTY
- content_replacements:
    #suppress-scheme
- concluding_replacements:
    #angle-bracket-wrap

Syntax

<«uri»>

<{«attribute specifications»} «uri»>

«flags»<«uri»>

«flags»<{«attribute specifications»} «uri»>

«flags»:
- b: wrap in angle brackets (apply #angle-bracket-wrap)
- s: suppress scheme (apply #suppress-scheme)
«attribute specifications»: see CMD attribute specifications.

Description

Produces a link:

<a«attribute sequence»>«uri»</a>

Or:

&lt;<a«attribute sequence»>«uri»</a>&gt;

Here, «attribute sequence» is the sequence of attributes built from

«uri» as href
«attribute specifications»

parsed in that order.

Examples

Basic usage:
- CMD: <https://example.com>
- HTML: <a href="https://example.com">https://example.com</a>
- Rendered: https://example.com
Wrap in angle brackets:
- CMD: b<https://example.com>
- HTML: <<a href="https://example.com">https://example.com</a>>
- Rendered: <https://example.com>
Suppress scheme:
- CMD: s<https://example.com>
- HTML: <a href="https://example.com">example.com</a>
- Rendered: example.com
¿Por qué no los dos?
- CMD: bs<https://example.com>
- HTML: <<a href="https://example.com">example.com</a>>
- Rendered: <example.com>
Invitation to spammers:
- CMD: s<mailto:mail@example.com>
- HTML: <a href="mailto:mail@example.com">mail@example.com</a>
- Rendered: mail@example.com

22. `#specified-links`

Definition

SpecifiedLinkReplacement: #specified-links
- queue_position: AFTER #explicit-links
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS

Syntax

[«link text»](«href»)
[«link text»](<«href»>)

[«link text»](«...» "«title»")
[«link text»](«...» '«title»')

[«link text»]{«attribute specifications»}(«...»)

«attribute specifications»: see CMD attribute specifications.

Description

Produces a link:

<a«attribute sequence»>«link text»</a>

Here, «attribute sequence» is the sequence of attributes built from

«href»
«title»
«attribute specifications»

parsed in that order.

Examples

Basic usage:

CMD:

[Wikipedia](https://en.wikipedia.org/wiki/Main_Page)

HTML:

<a href="https://en.wikipedia.org/wiki/Main_Page">Wikipedia</a>

Rendered:
Wikipedia

Title:

CMD:

[Wikipedia](
  https://en.wikipedia.org/wiki/Main_Page
  "Wikipedia, the free encyclopedia"
)

HTML:

<a href="https://en.wikipedia.org/wiki/Main_Page" title="Wikipedia, the free encyclopedia">Wikipedia</a>

Rendered:
Wikipedia

Override href

CMD:

[Beware]{href=https://example.com/evil/}(https://example.com/good/)

HTML:

<a href="https://example.com/evil/">Beware</a>

Rendered:
Beware

23. `#referenced-links`

Definition

ReferencedLinkReplacement: #referenced-links
- queue_position: AFTER #specified-links
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS

Syntax

[«link text»][«label»]
[«link text»]

[«link text»]{«attribute specifications»}[«label»]
[«link text»]{«attribute specifications»}

«label»: must correspond to a defined reference definition, up to case. If «label» is omitted, «link text» is used in its stead.
«attribute specifications»: see CMD attribute specifications.

Description

Produces a link:

<a«attribute sequence»>«link text»</a>

Here, «attribute sequence» is the sequence of attributes built from

«href» equal to «uri» of the reference definition defined for «label»
«title» of the reference definition defined for «label»
«attribute specifications» of the reference definition defined for «label»
«attribute specifications»

parsed in that order.

Examples

Basic usage:

CMD:

[wiki]: https://en.wikipedia.org/wiki/Main_Page

[Wikipedia's main page.][wiki]

HTML:

<a href="https://en.wikipedia.org/wiki/Main_Page">Wikipedia's main page.</a>

Rendered:
Wikipedia's main page.

Use «link text» for «label»:

CMD:

[Wikipedia]: https://en.wikipedia.org/wiki/Main_Page

[Wikipedia]

HTML:

<a href="https://en.wikipedia.org/wiki/Main_Page">Wikipedia</a>

Rendered:
Wikipedia

«label» is case-insensitive:

CMD:

[link-label-case]: https://example.com

[Hooray.][link-label-case]
[Hooray.][link-label-CASE]
[Hooray.][LiNK-laBEl-CAsE]

HTML:

<a href="https://example.com">Hooray.</a>
<a href="https://example.com">Hooray.</a>
<a href="https://example.com">Hooray.</a>

Later reference definitions prevail:

CMD:

[link-label]{.class1}: https://example.com/1
[link-label]{.class2}: https://example.com/2

[Second definition wins.][link-label]

HTML:

<a href="https://example.com/2" class="class2">Second definition wins.</a>

24. `#inline-semantics`

Definition

InlineAssortedDelimitersReplacement: #inline-semantics
- queue_position: AFTER #referenced-links
- delimiter_conversion:
    __=b
    _=i
    **=strong
    *=em
    ''=cite
    ""=q
- attribute_specifications: EMPTY
- prohibited_content: BLOCKS

Syntax

__«b content»__
_«i content»_
**«strong content»**
*«em content»*
''«cite content»''
""«q content»""

«...»{«attribute specifications»} «content»«...»

|«...»«content»«...»

|«...»{«attribute specifications»} «content»«...»

«attribute specifications»: see CMD attribute specifications.
The leading optional pipe is to be used as a disambiguator in some edge cases. If present, it indicates that the delimiter directly after it is opening rather than closing.
The opening delimiter must not be followed by whitespace, nor by </ (the beginning of a closing tag).
The closing delimiter must not be preceded by whitespace, nor by a pipe |.

Description

Produces an inline semantic:

<b«attribute sequence»>«b content»</b>
<i«attribute sequence»>«i content»</i>
<strong«attribute sequence»>«strong content»</strong>
<em«attribute sequence»>«em content»</em>
<cite«attribute sequence»>«cite content»</cite>
<q«attribute sequence»>«q content»</q>

Examples

In HTML5,  and  are not deprecated, see W3C: Using  and  elements.

Bring to attention :
1. Keywords:
 - CMD: Meals are served with __rice__ or __pasta__.
 - HTML: Meals are served with rice or pasta.
 - Rendered: Meals are served with rice or pasta.
Idiomatic offset :
1. Foreign phrases:
 - CMD: Write out _{lang=la} Romani ite domum_ 100 times.
 - HTML: Write out Romani ite domum 100 times.
 - Rendered: Write out Romani ite domum 100 times.
2. Translator-supplied words in the King James Bible:
 - CMD: I _{.translator-supplied} am_ the LORD.
 - HTML: I am the LORD.
 - Rendered: I am the LORD.
3. Thoughts:
 - CMD: _Screw 'em._
 - HTML: Screw 'em.
 - Rendered: Screw 'em.
Strong importance :
1. Warnings:
 - CMD: **{lang=de} Achtung!**
 - HTML: Achtung!
 - Rendered: Achtung!
Stress emphasis :
1. Stress a particular word in a sentence:
 - CMD: I don't know *who* took it.
 - HTML: I don't know who took it.
 - Rendered: I don't know who took it.
Citation <cite>:
1. Book titles:
 - CMD: ''Nineteen Eighty-Four'' is already upon us.
 - HTML: <cite>Nineteen Eighty-Four</cite> is already upon us.
 - Rendered: Nineteen Eighty-Four is already upon us.
Quotation <q>:
1. Inline quotes:
 - CMD: ""It wasn't me.""
 - HTML: <q>It wasn't me.</q>
 - Rendered: It wasn't me.

Nesting and disambiguation:

Pattern	CMD	HTML
1-1	`foo`	`<em>foo</em>`
2-2	`foo`	`<strong>foo</strong>`
3-3 (12-21)	`*foo*`	`<em><strong>foo</strong></em>`
12-3 (12-21)	`foo bar**`	`<em>foo <strong>bar</strong></em>`
3-21 (12-21)	`*foo bar*`	`<em><strong>foo</strong> bar</em>`
2\|1-3 (21-12)	`*\|foo***`	`<strong><em>foo</em></strong>`
21-3 (21-12)	`*foo bar***`	`<strong>foo <em>bar</em></strong>`
3-12 (21-12)	`**foo bar**`	`<strong><em>foo</em> bar</strong>`
4-4 (22-22)	`**foo**`	`<strong><strong>foo</strong></strong>`
1\|2\|1-4 (121-121)	`\|\|foo****`	`<em><strong><em>foo</em></strong></em>`

25. `#escape-idle-html`

Definition

RegexDictionaryReplacement: #escape-idle-html
- queue_position: AFTER #inline-semantics
* [&]
  (?!
    (?:
      [a-zA-Z]{1,31}
        |
      [#] (?: [0-9]{1,7} | [xX] [0-9a-fA-F]{1,6} )
    )
    [;]
  )
    --> &amp;
* [<] (?= [\s] ) --> &lt;

Description

Escapes idle HTML:

Replaces non-entity ampersand with &.
Replaces opening angle brackets before whitespace with <

26. `#whitespace`

Definition

ReplacementSequence: #whitespace
- queue_position: AFTER #escape-idle-html
- replacements:
    #reduce-whitespace

Description

Reduces whitespace. See #reduce-whitespace.

27. `#placeholder-unprotect`

Definition

PlaceholderUnprotectionReplacement: #placeholder-unprotect
- queue_position: AFTER #whitespace

Description

Unprotects all placeholder strings.

Standard unqueued replacements

`#placeholder-protect`

Definition

PlaceholderProtectionReplacement: #placeholder-protect

Description

Protects a string with a placeholder.

Dependants

#literals
#display-code
#inline-code
#cmd-properties
#boilerplate-protect
#backslash-escapes
#angle-bracket-wrap

`#de-indent`

Definition

DeIndentationReplacement: #de-indent
- negative_flag: KEEP_INDENTED

Description

Removes the longest common indentation in a string.

Dependants

#literals
#display-code
#inline-code

`#escape-html`

Definition

OrdinaryDictionaryReplacement: #escape-html
- negative_flag: KEEP_HTML_UNESCAPED
* & --> &amp;
* < --> &lt;
* > --> &gt;

Description

Escapes &, <, and > as their respective HTML ampersand entities.

Dependants

#literals
#display-code
#inline-code

`#trim-whitespace`

Definition

RegexDictionaryReplacement: #trim-whitespace
* \A [\s]* -->
* [\s]* \Z -->

Description

Removes whitespace at the very start and very end of the string.

Dependants

#literals
#table-headers
#table-data
#inline-code

`#reduce-whitespace`

Definition

RegexDictionaryReplacement: #reduce-whitespace
- positive_flag: REDUCE_WHITESPACE
* ^ [^\S\n]+ -->
* [^\S\n]+ $ -->
* [\s]+ (?= <br> ) -->
* [\n]+ --> \n

Description

Removes leading horizontal whitespace on each line.
Removes trailing horizontal whitespace on each line.
Removes whitespace before  .
Collapsed multiple consecutive newlines into a single newline.

Dependants

#literals
#display-code
#inline-code
#boilerplate-protect
#whitespace

`#code-tag-wrap`

Definition

RegexDictionaryReplacement: #code-tag-wrap
* \A --> <code>
* \Z --> </code>

Description

Wraps a string in opening and closing code tags.

Dependants

#display-code

`#prepend-newline`

Definition

RegexDictionaryReplacement: #prepend-newline
* \A --> \n

Description

Adds a newline at the very start of a string.

Dependants

#divisions
#blockquotes
#unordered-list-items
#unordered-lists
#ordered-list-items
#ordered-lists
#table-rows
#table-head
#table-body
#table-foot
#tables
#paragraphs

`#unordered-list-items`

Definition

PartitioningReplacement: #unordered-list-items
- starting_pattern: [-+*]
- attribute_specifications: EMPTY
- content_replacements:
    #prepend-newline
- ending_pattern: [-+*]
- tag_name: li

Description

Partitions content into list items based on leading occurrences of -, +, or *.

Dependants

#unordered-lists

`#ordered-list-items`

Definition

PartitioningReplacement: #ordered-list-items
- starting_pattern: [0-9]+ [.]
- attribute_specifications: EMPTY
- content_replacements:
    #prepend-newline
- ending_pattern: [0-9]+ [.]
- tag_name: li

Description

Partitions content into list items based on leading occurrences of a run of digits followed by a full stop.

Dependants

#ordered-lists

`#mark-table-headers-for-preceding-table-data`

Definition

RegexDictionaryReplacement: #mark-table-headers-for-preceding-table-data
* \A --> ;{}
# Replaces `<th«attributes_sequence»>` with `;{}<th«attributes_sequence»>`,
# so that #table-data will know to stop before it.

Description

Replaces <th«attributes_sequence»> with ;{}<th«attributes_sequence»> so that #table-data will know to stop before it.

Dependants

#table-headers

`#table-headers`

Definition

PartitioningReplacement: #table-headers
- starting_pattern: [;]
- attribute_specifications: EMPTY
- content_replacements:
    #trim-whitespace
- ending_pattern: [;,]
- tag_name: th
- concluding_replacements:
    #mark-table-headers-for-preceding-table-data

Description

Partitions content into table headers based on leading occurrences of ; up to the next leading ; or ,.

Dependants

#table-rows

`#table-data`

Definition

PartitioningReplacement: #table-data
- starting_pattern: [,]
- attribute_specifications: EMPTY
- content_replacements:
    #trim-whitespace
- ending_pattern: [;,]
- tag_name: td

Description

Partitions content into table data based on leading occurrences of , up to the next leading ; or ,.

Dependants

#table-rows

`#unmark-table-headers-for-preceding-table-data`

Definition

RegexDictionaryReplacement: #unmark-table-headers-for-preceding-table-data
* ^ [;] \{ \} <th (?P<bracket_or_placeholder_marker> [>\uF8FF] )
    -->
  <th\g<bracket_or_placeholder_marker>
# Replaces `;{}<th«attributes_sequence»>` with `<th«attributes_sequence»>`
# so that #mark-table-headers-for-preceding-table-data is undone.

Description

Replaces ;{}<th«attributes_sequence»> with <th«attributes_sequence»> so that #mark-table-headers-for-preceding-table-data is undone.

Dependants

#table-rows

`#table-rows`

Definition

PartitioningReplacement: #table-rows
- starting_pattern: [/]{2}
- attribute_specifications: EMPTY
- ending_pattern: [/]{2}
- content_replacements:
    #table-headers
    #table-data
    #unmark-table-headers-for-preceding-table-data
    #prepend-newline
- tag_name: tr

Description

Partitions content into table rows based on leading occurrences of //.

Dependants

#table-head
#table-body
#table-foot
#tables

`#table-head`

Definition

PartitioningReplacement: #table-head
- starting_pattern: [|][\^]
- attribute_specifications: EMPTY
- ending_pattern: [|][:_]
- content_replacements:
    #table-rows
    #prepend-newline
- tag_name: thead

Description

Partitions content into table heads based on leading occurrences of |^ up to the next leading |: or |_.

Dependants

#tables

`#table-body`

Definition

PartitioningReplacement: #table-body
- starting_pattern: [|][:]
- attribute_specifications: EMPTY
- ending_pattern: [|][_]
- content_replacements:
    #table-rows
    #prepend-newline
- tag_name: tbody

Description

Partitions content into table bodies based on leading occurrences of |: up to the next leading |_.

Dependants

#tables

`#table-foot`

Definition

PartitioningReplacement: #table-foot
- starting_pattern: [|][_]
- attribute_specifications: EMPTY
- content_replacements:
    #table-rows
    #prepend-newline
- tag_name: tfoot

Description

Partitions content into table feet based on leading occurrences of |_.

Dependants

#tables

`#suppress-scheme`

Definition

RegexDictionaryReplacement: #suppress-scheme
- positive_flag: SUPPRESS_SCHEME
* \A [\S]+ [:] (?: [/]{2} )? -->

Description

Suppresses the scheme (including the colon and possibly two slashes) of a URI.

Dependants

#explicit-links

`#angle-bracket-wrap`

Definition

RegexDictionaryReplacement: #angle-bracket-wrap
- positive_flag: ANGLE_BRACKET_WRAP
* \A --> &lt;
* \Z --> &gt;
- concluding_replacements:
    #placeholder-protect

Description

Wraps a string in angle brackets.

Dependants

#explicit-links

CMD replacement rule syntax

This section gives the syntax for writing user-defined CMD replacement rules, which go before the «delimiter» when authoring CMD files.

In CMD replacement rule syntax, a line must be one of the following:

Whitespace-only.
Ends the definition of a replacement rule.
A comment:
```
# «comment»
```
Is ignored.
A rules inclusion:
```
< «included_file_name»
```
Includes the content of «included_file_name» as CMD replacement rules.
- If «included_file_name» begins with a slash, it is parsed relative to the working directory.
- If «included_file_name» does not begin with a slash, it is parsed relative to the file in which the inclusion is invoked.
The convention is for «included_file_name» to have the extension .cmdr.
A class declaration:
```
«ClassName»: #«id»
```
Begins the definition of a replacement rule with class «ClassName».
- «ClassName» must be one of ReplacementSequence, PlaceholderMarkerReplacement, PlaceholderProtectionReplacement, PlaceholderUnprotectionReplacement, DeIndentationReplacement, OrdinaryDictionaryReplacement, RegexDictionaryReplacement, FixedDelimitersReplacement, ExtensibleFenceReplacement, PartitioningReplacement, InlineAssortedDelimitersReplacement, HeadingReplacement, ReferenceDefinitionReplacement, SpecifiedImageReplacement, ReferencedImageReplacement, ExplicitLinkReplacement, SpecifiedLinkReplacement, or ReferencedLinkReplacement.
- «id» may only contain lower-case letters, digits, hyphens, and full stops. The convention is for user-defined replacement rules to have «id» beginning with a full stop (whereas the standard rules have «id» beginning with a letter or digit).
The start of an attribute declaration:
```
- «name»: «value»
```
Declares an attribute for the replacement that is currently being defined.
- «name» must be one of the valid attribute names for the replacement that is currently being defined.
- «value» must be a suitable attribute value for that attribute name.
The start of an substitution declaration:
```
* «pattern» --> «substitute»
```
Declares a substitution for the replacement that is currently being defined (which must be an OrdinaryDictionaryReplacement or a RegexDictionaryReplacement).
- The number of hyphens in the delimiter --> may be arbitrarily increased should «pattern» contain a run of hyphens followed by a closing angle-bracket.
A continuation (of an attribute declaration or substitution declaration), beginning with whitespace.

`ReplacementSequence`

Syntax

ReplacementSequence: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
replacements: sequence of replacements to be applied

Description

Defines a replacement rule that applies a sequence of replacement rules.

Standard rules usage

#whitespace

`PlaceholderMarkerReplacement`

Syntax

PlaceholderMarkerReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»

queue_position: position in the replacement queue

Description

Defines a rule for replacing the placeholder marker with a placeholder.

Standard rules usage

#placeholder-markers

`PlaceholderProtectionReplacement`

Syntax

PlaceholderProtectionReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»

queue_position: position in the replacement queue

Description

Defines a replacement rule for protecting strings with a placeholder.

Standard rules usage

#placeholder-protect

`PlaceholderUnprotectionReplacement`

Syntax

PlaceholderUnprotectionReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»

queue_position: position in the replacement queue

Description

Defines a replacement rule for restoring placeholders to their strings.

Standard rules usage

#placeholder-unprotect

`DeIndentationReplacement`

Syntax

DeIndentationReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- positive_flag: (def) NONE | «FLAG_NAME»
- negative_flag: (def) NONE | «FLAG_NAME»

queue_position: position in the replacement queue
positive_flag: name of the flag that must be enabled for the replacement to be applied
negative_flag: name of the flag that must not be enabled for the replacement to be applied

Description

Defines a replacement rule for de-indentation.

Standard rules usage

#de-indent

`OrdinaryDictionaryReplacement`

Syntax

OrdinaryDictionaryReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- positive_flag: (def) NONE | «FLAG_NAME»
- negative_flag: (def) NONE | «FLAG_NAME»
- apply_mode: (def) SIMULTANEOUS | SEQUENTIAL
* "«pattern»" | '«pattern»' | «pattern»
    -->
  CMD_VERSION | CMD_NAME | CMD_BASENAME | CLEAN_URL |
          "«substitute»" | '«substitute»' | «substitute»
[...]
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
positive_flag: name of the flag that must be enabled for the replacement to be applied
negative_flag: name of the flag that must not be enabled for the replacement to be applied
apply_mode: whether the substitutions are to be applied simultaneously or sequentially
Patterns may be double-quoted, single-quoted, or bare.

Substitutes may be a keyword, double-quoted, single-quoted, or bare. Possible keywords are:

Keyword	Description
`CMD_VERSION`	`__version__` in `cmd.py` (currently `4.6.0`)
`CMD_NAME`	CMD file name, relative to working directory, without extension
`CMD_BASENAME`	CMD file name, without path, without extension
`CLEAN_URL`	`CMD_NAME`, with `CMD_BASENAME` removed if it equals `index`

concluding_replacements: sequence of replacements to be applied after substitution

Description

Defines a replacement rule for a dictionary of ordinary substitutions.

Standard rules usage

#escape-html
#boilerplate-properties
#cmd-properties
#backslash-escapes

`RegexDictionaryReplacement`

Syntax

RegexDictionaryReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- positive_flag: (def) NONE | «FLAG_NAME»
- negative_flag: (def) NONE | «FLAG_NAME»
* "«pattern»" | '«pattern»' | «pattern»
    -->
  CMD_VERSION | CMD_NAME | CMD_BASENAME | CLEAN_URL |
          "«substitute»" | '«substitute»' | «substitute»
[...]
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
positive_flag: name of the flag that must be enabled for the replacement to be applied
negative_flag: name of the flag that must not be enabled for the replacement to be applied
apply_mode: whether the substitutions are to be applied simultaneously or sequentially
Patterns may be double-quoted, single-quoted, or bare.

Substitutes may be a keyword, double-quoted, single-quoted, or bare. Possible keywords are:

Keyword	Description
`CMD_VERSION`	`__version__` in `cmd.py` (currently `4.6.0`)
`CMD_NAME`	CMD file name, relative to working directory, without extension
`CMD_BASENAME`	CMD file name, without path, without extension
`CLEAN_URL`	`CMD_NAME`, with `CMD_BASENAME` removed if it equals `index`

Patterns and substitutes are parsed according to Python regex syntax with flags=re.ASCII | re.MULTILINE | re.VERBOSE.
concluding_replacements: sequence of replacements to be applied after substitution

Description

Defines a replacement rule for a dictionary of regex substitutions.

Standard rules usage

#trim-whitespace
#reduce-whitespace
#code-tag-wrap
#comments
#prepend-newline
#mark-table-headers-for-preceding-table-data
#unmark-table-headers-for-preceding-table-data
#boilerplate
#boilerplate-protect
#backslash-continuations
#suppress-scheme
#angle-bracket-wrap
#escape-idle-html

`FixedDelimitersReplacement`

Syntax

FixedDelimitersReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- syntax_type: BLOCK | INLINE (mandatory)
- allowed_flags: (def) NONE | «letter»=«FLAG_NAME» [...]
- opening_delimiter: «string» (mandatory)
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS
- content_replacements: (def) NONE | #«id» [...]
- closing_delimiter: «string» (mandatory)
- tag_name: (def) NONE | «name»
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
syntax_type: whether the syntax type is block (delimiters must be on their own lines) or inline (delimiters may be anywhere)
allowed_flags: flag letters and corresponding flag names
opening_delimiter: the opening delimiter
attribute_specifications: whether, after the opening delimiter, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the content between the opening and closing delimiters
content_replacements: sequence of replacements to be applied to the content between the opening and closing delimiters
closing_delimiter: the closing delimiter
tag_name: the tag name of the element to be constructed
concluding_replacements: sequence of replacements to be applied after construction of the element

Description

Defines a fixed-delimiters replacement rule.

Standard rules usage

(None)

`ExtensibleFenceReplacement`

Syntax

ExtensibleFenceReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- syntax_type: BLOCK | INLINE (mandatory)
- allowed_flags: (def) NONE | «letter»=«FLAG_NAME» [...]
- prologue_delimiter: (def) NONE | «string»
- extensible_delimiter: «character_repeated» (mandatory)
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS
- content_replacements: (def) NONE | #«id» [...]
- epilogue_delimiter: (def) NONE | «string»
- tag_name: (def) NONE | «name»
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
syntax_type: whether the syntax type is block (delimiters must be on their own lines) or inline (delimiters may be anywhere)
allowed_flags: flag letters and corresponding flag names
prologue_delimiter: the delimiter that must appear before the opening extensible delimiter
extensible_delimiter: the opening and closing extensible delimiter (a character repeated)
attribute_specifications: whether, after the opening extensible delimiter, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the content between the opening and closing extensible delimiters
content_replacements: sequence of replacements to be applied to the content between the opening and closing extensible delimiters
epilogue_delimiter: the delimiter that must appear after the closing extensible delimiter
tag_name: the tag name of the element to be constructed
concluding_replacements: sequence of replacements to be applied after construction of the element

Description

Defines a generalised extensible-fence-style replacement rule. (Inspired by the repeatable backticks of John Gruber's Markdown.)

Standard rules usage

#literals
#display-code
#divisions
#blockquotes
#unordered-lists
#ordered-lists
#tables
#paragraphs
#inline-code

`PartitioningReplacement`

Syntax

PartitioningReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- starting_pattern: «regex» (mandatory)
- attribute_specifications: (def) NONE | EMPTY | «string»
- content_replacements: (def) NONE | #«id» [...]
- ending_pattern: (def) NONE | «regex»
- tag_name: (def) NONE | «name»
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
starting_pattern: pattern that marks the start of a partition
attribute_specifications: whether, after the starting pattern, there may be CMD attribute specifications supplied, and if so, what the default specification is
content_replacements: sequence of replacements to be applied to the content from the starting pattern up to the ending pattern
ending_pattern: pattern, not consumed, that marks the end of a partition
tag_name: the tag name of the element to be constructed
concluding_replacements: sequence of replacements to be applied after construction of the element

Description

Defines a generalised partitioning replacement rule. (Does partitioning by consuming everything from a starting pattern up to but not including an ending pattern.)

Standard rules usage

#unordered-list-items
#ordered-list-items
#table-headers
#table-data
#table-rows
#table-head
#table-body
#table-foot

`InlineAssortedDelimitersReplacement`

Syntax

InlineAssortedDelimitersReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- delimiter_conversion:
    «character» | «character_doubled»=«tag_name» [...] (mandatory)
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS

queue_position: position in the replacement queue
delimiter_conversion: a sequence of conversions from single-character or double-character delimiters to tag names
attribute_specifications: whether, after each starting delimiter, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the content between the opening and closing delimiters

Description

Defines an inline assorted-delimiters replacement rule. (A generalisation of the replacement that processes * and ** into  and  in Markdown.)

Standard rules usage

#inline-semantics

`HeadingReplacement`

Syntax

HeadingReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»

queue_position: position in the replacement queue
attribute_specifications: whether, after the declaring hashes, there may be CMD attribute specifications supplied, and if so, what the default specification is

Description

Defines a replacement rule for headings. (A generalisation of the replacement that processes ATX headings in Markdown.)

Standard rules usage

#headings

`ReferenceDefinitionReplacement`

Syntax

ReferenceDefinitionReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»

queue_position: position in the replacement queue
attribute_specifications: whether, between the square-bracketed label and the declaring colon, there may be CMD attribute specifications supplied, and if so, what the default specification is

Description

Defines a replacement rule for consuming reference definitions. (A generalisation of the replacement that processes [«label»]: «uri» "«title»" in Markdown.)

Standard rules usage

#reference-definitions

`SpecifiedImageReplacement`

Syntax

SpecifiedImageReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS

queue_position: position in the replacement queue
attribute_specifications: whether, between the square-bracketed alt text and the round-bracketed URI & title, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the square-bracketed alt text

Description

Defines a replacement rule for specified images. (A generalisation of the replacement that processes ![«alt text»](«uri» "«title»") in Markdown.)

Standard rules usage

#specified-images

`ReferencedImageReplacement`

Syntax

ReferencedImageReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS

queue_position: position in the replacement queue
attribute_specifications: whether, after the square-bracketed alt text, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the square-bracketed alt text and the square-bracketed label

Description

Defines a replacement rule for referenced images. (A generalisation of the replacement that processes ![«alt text»][«label»] in Markdown.)

Standard rules usage

#referenced-images

`ExplicitLinkReplacement`

Syntax

ExplicitLinkReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- allowed_flags: (def) NONE | «letter»=«FLAG_NAME» [...]
- attribute_specifications: (def) NONE | EMPTY | «string»
- content_replacements: (def) NONE | #«id» [...]
- concluding_replacements: (def) NONE | #«id» [...]

queue_position: position in the replacement queue
allowed_flags: flag letters and corresponding flag names
attribute_specifications: whether, after the opening angle bracket, there may be CMD attribute specifications supplied, and if so, what the default specification is
content_replacements: sequence of replacements to be applied to the content between the opening and closing angle brackets
concluding_replacements: sequence of replacements to be applied after construction of the element

Description

Defines a replacement rule for explicit links. (A generalisation of the replacement that processes <«uri»> in Markdown.)

Standard rules usage

#explicit-links

`SpecifiedLinkReplacement`

Syntax

SpecifiedLinkReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS

queue_position: position in the replacement queue
attribute_specifications: whether, between the square-bracketed link text and the round-bracketed URI & title, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the square-bracketed link text

Description

Defines a replacement rule for specified links. (A generalisation of the replacement that processes [«link text»](«uri» "«title»") in Markdown.)

Standard rules usage

#specified-links

`ReferencedLinkReplacement`

Syntax

ReferencedLinkReplacement: #«id»
- queue_position: (def) NONE | ROOT | BEFORE #«id» | AFTER #«id»
- attribute_specifications: (def) NONE | EMPTY | «string»
- prohibited_content: (def) NONE | BLOCKS | ANCHORED_BLOCKS

queue_position: position in the replacement queue
attribute_specifications: whether, after the square-bracketed link text, there may be CMD attribute specifications supplied, and if so, what the default specification is
prohibited_content: whether block tags or anchored block tags are prohibited in the square-bracketed link text and the square-bracketed label

Description

Defines a replacement rule for referenced links. (A generalisation of the replacement that processes [«link text»][«label»] in Markdown.)

Standard rules usage

#referenced-links

CMD placeholders

There are many instances in which the result of a replacement should not be altered further by replacements to follow. To protect a string from further alteration, it is temporarily replaced by a placeholder consisting of code points in the main Unicode Private Use Area.

Specifically, the placeholder for a string is of the following form:

«marker»«run_characters»«marker»

«marker» is «U+F8FF».
«run_characters» are between «U+E000» and «U+E0FF» each representing a Unicode byte of the string.

It is assumed that the user will not define replacement rules that tamper with strings of the form «marker»«run_characters»«marker».

Example

Consider the string £3.
Its code points are U+00A3 and U+0033.
The corresponding Unicode bytes are \xC2\xA3 and \x33.
The «run_characters» are therefore «U+E0C2»«U+E0A3»«U+E033».
The full placeholder is therefore «U+F8FF»«U+E0C2»«U+E0A3»«U+E033»«U+F8FF».

CMD attribute specifications

When a CMD replacement rule is defined with attribute_specifications not equal to NONE, it allows HTML attributes to be specified by CMD attribute specifications enclosed in curly brackets.

CMD attribute specifications may be of the following forms:

«name»="«quoted value (whitespace allowed)»"
«name»=«bare-value»
#«id»
.«class»
r«rowspan»
c«colspan»
w«width»
h«height»
-«delete-name»
«boolean-name»

In the two forms with an explicit equals sign, the following abbreviations are allowed for «name»:

# for id
. for class
l for lang
r for rowspan
c for colspan
w for width
h for height
s for style

Examples

Behaviour for the standard rule #inline-code:
- CMD: `{#foo .bar l=en-AU title="baz"} test`
- HTML: <code id="foo" class="bar" lang="en-AU" title="baz">test</code>
Empty attribute specifications:
- CMD: `{} test`
- HTML: <code>test</code>
Non-class values will supersede earlier ones:
- CMD: `{#1 #2 title=A title=B} test`
- HTML: <code id="2" title="B">test</code>
class values will accumulate:
- CMD: `{.a .b class=c .d} test`
- HTML: <code class="a b c d">test</code>
Delete class to reset it:
- CMD: `{.a .b -class class=c .d} test`
- HTML: <code class="c d">test</code>

Repository links

This page's CMD: index.cmd
This page's HTML: index.html
This page's repository: conway-markdown.github.io
Python implementation's repository: conway-markdown

Conway-Markdown (CMD) v4.6.0

Contents

Installation and usage

Linux terminals, macOS Terminal, Git BASH for Windows

Windows Command Prompt

Authoring CMD files

Standard CMD replacement rules

Standard queued replacements

0. #placeholder-markers

1. #literals

2. #display-code

3. #comments

4. #divisions

5. #blockquotes

6. #unordered-lists

7. #ordered-lists

8. #tables

9. #headings

10. #paragraphs

11. #inline-code

12. #boilerplate

13. #boilerplate-properties

14. #cmd-properties

15. #boilerplate-protect

16. #backslash-escapes

17. #backslash-continuations

18. #reference-definitions

19. #specified-images

20. #referenced-images

21. #explicit-links

22. #specified-links

23. #referenced-links

24. #inline-semantics

25. #escape-idle-html

26. #whitespace

27. #placeholder-unprotect

Standard unqueued replacements

#placeholder-protect

#de-indent

#escape-html

#trim-whitespace

#reduce-whitespace

#code-tag-wrap

#prepend-newline

#unordered-list-items

#ordered-list-items

#mark-table-headers-for-preceding-table-data

#table-headers

#table-data

#unmark-table-headers-for-preceding-table-data

#table-rows

#table-head

#table-body

#table-foot

#suppress-scheme

#angle-bracket-wrap

CMD replacement rule syntax

ReplacementSequence

PlaceholderMarkerReplacement

PlaceholderProtectionReplacement

PlaceholderUnprotectionReplacement

DeIndentationReplacement

OrdinaryDictionaryReplacement

RegexDictionaryReplacement

FixedDelimitersReplacement

ExtensibleFenceReplacement

PartitioningReplacement

InlineAssortedDelimitersReplacement

HeadingReplacement

ReferenceDefinitionReplacement

SpecifiedImageReplacement

ReferencedImageReplacement

ExplicitLinkReplacement

SpecifiedLinkReplacement

ReferencedLinkReplacement

CMD placeholders

Example

CMD attribute specifications

Examples

Repository links

0. `#placeholder-markers`

1. `#literals`

2. `#display-code`

3. `#comments`

4. `#divisions`

5. `#blockquotes`

6. `#unordered-lists`

7. `#ordered-lists`

8. `#tables`

9. `#headings`

10. `#paragraphs`

11. `#inline-code`

12. `#boilerplate`

13. `#boilerplate-properties`

14. `#cmd-properties`

15. `#boilerplate-protect`

16. `#backslash-escapes`

17. `#backslash-continuations`

18. `#reference-definitions`

19. `#specified-images`

20. `#referenced-images`

21. `#explicit-links`

22. `#specified-links`

23. `#referenced-links`

24. `#inline-semantics`

25. `#escape-idle-html`

26. `#whitespace`

27. `#placeholder-unprotect`

`#placeholder-protect`

`#de-indent`

`#escape-html`

`#trim-whitespace`

`#reduce-whitespace`

`#code-tag-wrap`

`#prepend-newline`

`#unordered-list-items`

`#ordered-list-items`

`#mark-table-headers-for-preceding-table-data`

`#table-headers`

`#table-data`

`#unmark-table-headers-for-preceding-table-data`

`#table-rows`

`#table-head`

`#table-body`

`#table-foot`

`#suppress-scheme`

`#angle-bracket-wrap`

`ReplacementSequence`

`PlaceholderMarkerReplacement`

`PlaceholderProtectionReplacement`

`PlaceholderUnprotectionReplacement`

`DeIndentationReplacement`

`OrdinaryDictionaryReplacement`

`RegexDictionaryReplacement`

`FixedDelimitersReplacement`

`ExtensibleFenceReplacement`

`PartitioningReplacement`

`InlineAssortedDelimitersReplacement`

`HeadingReplacement`

`ReferenceDefinitionReplacement`

`SpecifiedImageReplacement`

`ReferencedImageReplacement`

`ExplicitLinkReplacement`

`SpecifiedLinkReplacement`

`ReferencedLinkReplacement`