Add Lexer for scdoc

The scdoc markup language is increasingly used by the Wayland
community for writing manpages. Examples are:

* https://git.sr.ht/~sircmpwn/scdoc/tree/master/scdoc.5.scd
* https://github.com/swaywm/sway/blob/master/sway/sway.1.scd
* https://github.com/swaywm/swayidle/blob/master/swayidle.1.scd
* https://git.sr.ht/~sircmpwn/aerc/tree/master/doc/aerc-tutorial.7.scd

3 files changed, 257 insertions, 0 deletions

diff --git a/pygments/lexers/_mapping.py b/pygments/lexers/_mapping.py
index b6aed44f..7dfb2775 100644
--- a/pygments/lexers/_mapping.py
+++ b/pygments/lexers/_mapping.py
@@ -389,6 +389,7 @@ LEXERS = {
     'SassLexer': ('pygments.lexers.css', 'Sass', ('sass',), ('*.sass',), ('text/x-sass',)),
     'ScalaLexer': ('pygments.lexers.jvm', 'Scala', ('scala',), ('*.scala',), ('text/x-scala',)),
     'ScamlLexer': ('pygments.lexers.html', 'Scaml', ('scaml',), ('*.scaml',), ('text/x-scaml',)),
+    'ScdocLexer': ('pygments.lexers.scdoc', 'scdoc', ('scdoc', 'scd'), ('*.scd', '*.scdoc'), ()),
     'SchemeLexer': ('pygments.lexers.lisp', 'Scheme', ('scheme', 'scm'), ('*.scm', '*.ss'), ('text/x-scheme', 'application/x-scheme')),
     'ScilabLexer': ('pygments.lexers.matlab', 'Scilab', ('scilab',), ('*.sci', '*.sce', '*.tst'), ('text/scilab',)),
     'ScssLexer': ('pygments.lexers.css', 'SCSS', ('scss',), ('*.scss',), ('text/x-scss',)),
diff --git a/pygments/lexers/scdoc.py b/pygments/lexers/scdoc.py
new file mode 100644
index 00000000..94e3954d
--- /dev/null
+++ b/pygments/lexers/scdoc.py
@@ -0,0 +1,59 @@
+import re
+
+from pygments.lexer import RegexLexer, include, bygroups, \
+    using, this
+from pygments.token import Text, Comment, Keyword, String, \
+    Generic
+
+
+__all__ = ['ScdocLexer']
+
+
+class ScdocLexer(RegexLexer):
+    """
+    `scdoc` is a simple man page generator for POSIX systems written in C99.
+    https://git.sr.ht/~sircmpwn/scdoc
+
+    .. versionadded:: 2.5
+    """
+    name = 'scdoc'
+    aliases = ['scdoc', 'scd']
+    filenames = ['*.scd', '*.scdoc']
+    flags = re.MULTILINE
+
+    tokens = {
+        'root': [
+            # comment
+            (r'^(;.+\n)', bygroups(Comment)),
+
+            # heading with pound prefix
+            (r'^(#)([^#].+\n)', bygroups(Generic.Heading, Text)),
+            (r'^(#{2})(.+\n)', bygroups(Generic.Subheading, Text)),
+            # bulleted lists
+            (r'^(\s*)([*-])(\s)(.+\n)',
+            bygroups(Text, Keyword, Text, using(this, state='inline'))),
+            # numbered lists
+            (r'^(\s*)(\.+\.)( .+\n)',
+            bygroups(Text, Keyword, using(this, state='inline'))),
+            # quote
+            (r'^(\s*>\s)(.+\n)', bygroups(Keyword, Generic.Emph)),
+            # text block
+            (r'^(```\n)([\w\W]*?)(^```$)', bygroups(String, Text, String)),
+
+            include('inline'),
+        ],
+        'inline': [
+            # escape
+            (r'\\.', Text),
+            # underlines
+            (r'(\s)(_[^_]+_)(\W|\n)', bygroups(Text, Generic.Emph, Text)),
+            # bold
+            (r'(\s)(\*[^\*]+\*)(\W|\n)', bygroups(Text, Generic.Strong, Text)),
+            # inline code
+            (r'`[^`]+`', String.Backtick),
+
+            # general text, must come last!
+            (r'[^\\\s]+', Text),
+            (r'.', Text),
+        ],
+    }
diff --git a/tests/examplefiles/scdoc_manual.scd b/tests/examplefiles/scdoc_manual.scd
new file mode 100644
index 00000000..65a2b366
--- /dev/null
+++ b/tests/examplefiles/scdoc_manual.scd
@@ -0,0 +1,197 @@
+scdoc(5)
+
+# NAME
+
+scdoc - document format for writing manual pages
+
+# SYNTAX
+
+Input files must use the UTF-8 encoding.
+
+## PREAMBLE
+
+Each scdoc file must begin with the following preamble:
+
+	*name*(_section_) ["left\_footer" ["center\_header"]]
+
+*name* is the name of the man page you are writing, and _section_ is the section
+you're writing for (see *man*(1) for information on manual sections).
+
+_left\_footer_ and _center\_header_ are optional arguments which set the text
+positioned at those locations in the generated man page, and *must* be
+surrounded with double quotes.
+
+## SECTION HEADERS
+
+Each section of your man page should begin with something similar to the
+following:
+
+	# HEADER NAME
+
+Subsection headers are also understood - use two hashes. Each header must have
+an empty line on either side.
+
+## PARAGRAPHS
+
+Begin a new paragraph with an empty line.
+
+## LINE BREAKS
+
+Insert a line break by ending a line with \+\+.
+
+The result looks++
+like this.
+
+## FORMATTING
+
+Text can be made *bold* or _underlined_ with asterisks and underscores: \*bold\*
+or \_underlined\_. Underscores in the_middle_of_words will be disregarded.
+
+## INDENTATION
+
+You may indent lines with tab characters (*\\t*) to indent them by 4 spaces in
+the output. Indented lines may not contain headers.
+
+	The result looks something like this.
+
+	You may use multiple lines and most _formatting_.
+
+Deindent to return to normal, or indent again to increase your indentation
+depth.
+
+## LISTS
+
+You may start bulleted lists with dashes (-), like so:
+
+```
+- Item 1
+- Item 2
+	- Subitem 1
+	- Subitem 2
+- Item 3
+```
+
+The result looks like this:
+
+- Item 1
+- Item 2
+	- Subitem 1
+	- Subitem 2
+- Item 3
+
+You may also extend long entries onto another line by giving it the same indent
+level, plus two spaces. They will be rendered as a single list entry.
+
+```
+- Item 1 is pretty long so let's
+  break it up onto two lines
+- Item 2 is shorter
+	- But its children can go on
+	  for a while
+```
+
+- Item 1 is pretty long so let's
+  break it up onto two lines
+- Item 2 is shorter
+	- But its children can go on
+	  for a while
+
+## NUMBERED LISTS
+
+Numbered lists are similar to normal lists, but begin with periods (.) instead
+of dashes (-), like so:
+
+```
+. Item 1
+. Item 2
+. Item 3,
+  with multiple lines
+```
+
+. Item 1
+. Item 2
+. Item 3,
+  with multiple lines
+
+## TABLES
+
+To begin a table, add an empty line followed by any number of rows.
+
+Each line of a table should start with | or : to start a new row or column
+respectively (or space to continue the previous cell on multiple lines),
+followed by [ or - or ] to align the contents to the left, center, or right,
+followed by a space and the contents of that cell.  You may use a space instead
+of an alignment specifier to inherit the alignment of the same column in the
+previous row.
+
+The first character of the first row is not limited to | and has special
+meaning. [ will produce a table with borders around each cell. | will produce a
+table with no borders. ] will produce a table with one border around the whole
+table.
+
+To conclude your table, add an empty line after the last row.
+
+```
+[[ *Foo*
+:- _Bar_
+:-
+|  *Row 1*
+:  Hello
+:] world!
+|  *Row 2*
+:  こんにちは
+:  世界
+   !
+```
+
+[[ *Foo*
+:- _Bar_
+:-
+|  *Row 1*
+:  Hello
+:] world!
+|  *Row 2*
+:  こんにちは
+:  世界
+   !
+
+## LITERAL TEXT
+
+You may turn off scdoc formatting and output literal text with escape codes and
+literal blocks. Inserting a \\ into your source will cause the subsequent symbol
+to be treated as a literal and copied directly to the output. You may also make
+blocks of literal syntax like so:
+
+```
+\```
+_This formatting_ will *not* be interpreted by scdoc.
+\```
+```
+
+These blocks will be indented one level. Note that literal text is shown
+literally in the man viewer - that is, it's not a means for inserting your own
+roff macros into the output. Note that \\ is still interpreted within literal
+blocks, which for example can be useful to output \``` inside of a literal
+block.
+
+## COMMENTS
+
+Lines beginning with ; and a space are ignored.
+
+```
+; This is a comment
+```
+
+# CONVENTIONS
+
+By convention, all scdoc documents should be hard wrapped at 80 columns.
+
+# SEE ALSO
+
+*scdoc*(1)
+
+# AUTHORS
+
+Maintained by Drew DeVault <sir@cmpwn.com>. Up-to-date sources can be found at
+https://git.sr.ht/~sircmpwn/scdoc and bugs/patches can be submitted by email to
+~sircmpwn/public-inbox@lists.sr.ht.