Regenerate man page.

doc/djot.1

@@ -1,126 +1,104 @@
-.\" Automatically generated by Pandoc 3.1.2
+.\" Automatically generated by Pandoc 3.1.7
 .\"
-.\" Define V font for inline verbatim, using C font in formats
-.\" that render this, and otherwise B font.
-.ie "\f[CB]x\f[]"x" \{\
-. ftr V B
-. ftr VI BI
-. ftr VB B
-. ftr VBI BI
-.\}
-.el \{\
-. ftr V CR
-. ftr VI CI
-. ftr VB CB
-. ftr VBI CBI
-.\}
-.TH "DJOT" "1" "" "djot 0.2.3" ""
-.hy
+.TH "DJOT" "1" "" "djot 0.2.4" ""
 .SH NAME
-.PP
 djot \[en] converts djot markup.
 .SH SYNOPSIS
-.PP
 djot options [file..]
 .SH DESCRIPTION
-.PP
 djot is a command-line parser for djot markup (https://djot.net).
 It can produce
 .IP \[bu] 2
-an HMTL document (default behavior or \f[V]-t html\f[R])
+an HMTL document (default behavior or \f[CR]-t html\f[R])
 .IP \[bu] 2
-a stream of annotated tokens with byte offsets (\f[V]-t events\f[R])
+a stream of annotated tokens with byte offsets (\f[CR]-t events\f[R])
 .IP \[bu] 2
-an AST in either human-readable or JSON form (\f[V]-t ast\f[R],
-\f[V]-t astpretty\f[R]).
+an AST in either human-readable or JSON form (\f[CR]-t ast\f[R],
+\f[CR]-t astpretty\f[R]).
 .IP \[bu] 2
-a djot document (\f[V]-t djot\f[R])
+a djot document (\f[CR]-t djot\f[R])
 .IP \[bu] 2
-a pandoc AST serialized to JSON (\f[V]-t pandoc\f[R]), which can be read
-by pandoc (version >= 3) and converted to many other formats
+a pandoc AST serialized to JSON (\f[CR]-t pandoc\f[R]), which can be
+read by pandoc (version >= 3) and converted to many other formats
 .PP
-It can also read a JSON-serialized djot AST (\f[V]-f ast\f[R]) or a
-pandoc AST serialized to JSON (\f[V]-f pandoc\f[R]), allowing conversion
-from many other formats.
+It can also read a JSON-serialized djot AST (\f[CR]-f ast\f[R]) or a
+pandoc AST serialized to JSON (\f[CR]-f pandoc\f[R]), allowing
+conversion from many other formats.
 .PP
 Finally, it can apply \f[I]filters\f[R] that modify the parsed AST
 between the parsing and rendering steps.
 This is the primary way in which djot syntax can be customized.
 .SH OPTIONS
 .TP
-\f[V]--to,-t\f[R] \f[I]FORMAT\f[R]
+\f[CR]--to,-t\f[R] \f[I]FORMAT\f[R]
 Specify format to convert to.
-Valid values are \f[V]html\f[R] (the default), \f[V]ast\f[R]
-(JSON-formatted AST), \f[V]astpretty\f[R] (human-readable AST),
-\f[V]events\f[R] (JSON array of events produced by the tokenizer),
-\f[V]djot\f[R] (djot markup), or \f[V]pandoc\f[R] (JSON serialized
+Valid values are \f[CR]html\f[R] (the default), \f[CR]ast\f[R]
+(JSON-formatted AST), \f[CR]astpretty\f[R] (human-readable AST),
+\f[CR]events\f[R] (JSON array of events produced by the tokenizer),
+\f[CR]djot\f[R] (djot markup), or \f[CR]pandoc\f[R] (JSON serialized
 pandoc AST).
 .TP
-\f[V]--from,-f\f[R] \f[I]FORMAT\f[R]
+\f[CR]--from,-f\f[R] \f[I]FORMAT\f[R]
 Specify format to convert from.
-Valid values are \f[V]djot\f[R] (the default), \f[V]ast\f[R]
-(JSON-formatted AST), and \f[V]pandoc\f[R] (JSON serialized pandoc AST).
+Valid values are \f[CR]djot\f[R] (the default), \f[CR]ast\f[R]
+(JSON-formatted AST), and \f[CR]pandoc\f[R] (JSON serialized pandoc
+AST).
 .TP
-\f[V]--filter\f[R] \f[I]FILE\f[R]
+\f[CR]--filter\f[R] \f[I]FILE\f[R]
 Read a filter from \f[I]FILE\f[R] and apply it to the parsed AST prior
 to rendering.
 This option may be used multiple times; the filters will be applied in
 the order specified on the command line.
 See FILTERS below for details.
 .TP
-\f[V]--compact\f[R]
+\f[CR]--compact\f[R]
 Use compact JSON for the AST, with no extra spacing for readbility.
 .TP
-\f[V]--width,-w\f[R] \f[I]NUMBER\f[R]
-Text width to which to wrap output of \f[V]-t djot\f[R].
+\f[CR]--width,-w\f[R] \f[I]NUMBER\f[R]
+Text width to which to wrap output of \f[CR]-t djot\f[R].
 If \f[I]NUMBER\f[R] is 0, no wrapping will be done, and line breaks in
 the input will be preserved in the output.
 If it is -1, no wrapping will be done, and line breaks in the input will
 be treated as spaces.
 .TP
-\f[V]--sourcepos,-p\f[R]
+\f[CR]--sourcepos,-p\f[R]
 Include information about the start and end positions of elements in the
 HTML and AST.
 Source positions include the line number, column number (counting UTF-16
 code units, which might not accord with visible columns), and character
 offset in the entire document (again, in UTF-16 code units).
 .TP
-\f[V]--time\f[R]
+\f[CR]--time\f[R]
 Print timings for parsing, filtering, and rendering to stderr.
 .TP
-\f[V]--quiet\f[R]
+\f[CR]--quiet\f[R]
 Suppress warnings.
 .TP
-\f[V]--version\f[R]
+\f[CR]--version\f[R]
 Print version
 .TP
-\f[V]--help,-h\f[R]
+\f[CR]--help,-h\f[R]
 Print this message.
 .SH FILTERS
-.PP
 Filters are JavaScript programs that modify the parsed document prior to
 rendering.
 Here is an example of a filter that capitalizes all the content text in
 a document:
 .IP
-.nf
-\f[C]
+.EX
 // This filter capitalizes regular text, leaving code and URLs unaffected
 return {
   str: (el) => {
     el.text = el.text.toUpperCase();
   }
 }
-\f[R]
-.fi
+.EE
 .PP
-Save this as \f[V]caps.js\f[R] use tell djot to use it using
+Save this as \f[CR]caps.js\f[R] use tell djot to use it using
 .IP
-.nf
-\f[C]
+.EX
 djot --filter caps input.js
-\f[R]
-.fi
+.EE
 .PP
 Note: never run a filter from a source you don\[cq]t trust, without
 inspecting the code carefully.
@@ -132,42 +110,39 @@ document.
 This filter doesn\[cq]t alter the document at all; it just prints the
 list to stderr.
 .IP
-.nf
-\f[C]
+.EX
 return {
   link: (el) => {
     process.stderr:write(el.destination + \[dq]\[rs]n\[dq])
   }
 }
-\f[R]
-.fi
+.EE
 .PP
 A filter walks the document\[cq]s abstract syntax tree, applying
 functions to like-tagged nodes, so you will want to get familiar with
 how djot\[cq]s AST is designed.
-The easiest way to do this is to use \f[V]djot --ast\f[R] or
-\f[V]djot --astpretty\f[R].
+The easiest way to do this is to use \f[CR]djot --ast\f[R] or
+\f[CR]djot --astpretty\f[R].
 .PP
 By default filters do a bottom-up traversal; that is, the filter for a
 node is run after its children have been processed.
 It is possible to do a top-down travel, though, and even to run separate
 actions on entering a node (before processing the children) and on
 exiting (after processing the children).
 To do this, associate the node\[cq]s tag with a table containing
-\f[V]enter\f[R] and/or \f[V]exit\f[R] functions.
-The \f[V]enter\f[R] function is run when we traverse \f[I]into\f[R] the
-node, before we traverse its children, and the \f[V]exit\f[R] function
+\f[CR]enter\f[R] and/or \f[CR]exit\f[R] functions.
+The \f[CR]enter\f[R] function is run when we traverse \f[I]into\f[R] the
+node, before we traverse its children, and the \f[CR]exit\f[R] function
 is run after we have traversed the node\[cq]s children.
-For a top-down traversal, you\[cq]d just use the \f[V]enter\f[R]
+For a top-down traversal, you\[cq]d just use the \f[CR]enter\f[R]
 functions.
 If the tag is associated directly with a function, as in the first
 example above, it is treated as an \[ga]exit\[cq] function.
 .PP
 The following filter will capitalize text that is nested inside
 emphasis, but not other text:
 .IP
-.nf
-\f[C]
+.EX
 // This filter capitalizes the contents of emph
 // nodes instead of italicizing them.
 let capitalize = 0;
@@ -187,14 +162,12 @@ return {
       }
    }
 }
-\f[R]
-.fi
+.EE
 .PP
 Here is a simple filter that changes letter enumerated lists to
 roman-numbered:
 .IP
-.nf
-\f[C]
+.EX
 // Changes letter-enumerated lists to roman-numbered
 return {
   list: (e) => {
@@ -205,14 +178,12 @@ return {
     }
   }
 }
-\f[R]
-.fi
+.EE
 .PP
 A single filter may return a table with multiple tables, which will be
 applied sequentially:
 .IP
-.nf
-\f[C]
+.EX
 // This filter includes two sub-filters, run in sequence
 return [
   { // first filter changes (TM) to trademark symbol
@@ -226,22 +197,20 @@ return [
     }
   }
 ]
-\f[R]
-.fi
+.EE
 .PP
 The filters we\[cq]ve looked at so far modify nodes in place by changing
-one of their properties (\f[V]text\f[R]).
+one of their properties (\f[CR]text\f[R]).
 Sometimes we\[cq]ll want to replace a node with a different kind of
 node, or with several nodes, or to delete a node.
-In these cases we can end the filter function with a \f[V]return\f[R].
+In these cases we can end the filter function with a \f[CR]return\f[R].
 If a single AST node is returned, it will replace the element the filter
 is processing.
 If an array of AST nodes is returned, they will be spliced in to replace
 the element.
 If an empty array is returned, the element will be deleted.
 .IP
-.nf
-\f[C]
+.EX
 // This filter replaces certain Symb nodes with
 // formatted text.
 const substitutions = {
@@ -258,38 +227,33 @@ return {
     }
   }
 }
-\f[R]
-.fi
+.EE
 .IP
-.nf
-\f[C]
+.EX
 // This filter replaces all Image nodes with their descriptions.
 return {
   image: (e) => {
     return e.children;
   }
 }
-\f[R]
-.fi
+.EE
 .PP
 It is possible to inhibit traversal into the children of a node, by
-having the \f[V]enter\f[R] function return an object with the property
-\f[V]stop\f[R].
-The contents of \f[V]stop\f[R] will be used as the regular return value.
+having the \f[CR]enter\f[R] function return an object with the property
+\f[CR]stop\f[R].
+The contents of \f[CR]stop\f[R] will be used as the regular return
+value.
 This can be used, for example, to prevent the contents of a footnote
 from being processed:
 .IP
-.nf
-\f[C]
+.EX
 return {
  footnote: {
    enter: (e) => {
      return {stop: [e]};
     }
   }
 }
-\f[R]
-.fi
+.EE
 .SH AUTHORS
-.PP
 John MacFarlane (<[email protected]>).