Implement lists

scdoc.1.scd

@@ -1,17 +1,24 @@
-scdoc(5)
+scdoc(1)
 
 # NAME
 
-scdoc - syntax description for scdoc markup language
+scdoc - tool for generating *roff*(7) manual pages
+
+# SYNOPSIS
+
+*scdoc* < _input_ > _output_
 
 # DESCRIPTION
 
 scdoc is a tool designed to make the process of writing man pages more
-friendly. It converts scdoc files into roff macros, which can then be converted
+friendly. It reads scdoc syntax from stdin and writes roff to stdout, suitable
-to man pages or a number of other formats. The syntax is inspired by, but not
+for reading with *man*(1).
-directly taken from, markdown. Input files *must* use the UTF-8 encoding.
+
+# SYNTAX
+
+Input files must use the UTF-8 encoding.
 
-# PREAMBLE
+## PREAMBLE
 
 Each scdoc file must begin with the following preamble:
 
@@ -20,7 +27,7 @@ Each scdoc file must begin with the following preamble:
 The *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).
 
-# SECTION HEADERS
+## SECTION HEADERS
 
 Each section of your man page should begin with something similar to the
 following:
@@ -30,39 +37,47 @@ following:
 Subsection headers are also understood - use two hashes. Each header must have
 an empty line on either side.
 
-# PARAGRAPHS
+## PARAGRAPHS
 
 Begin a new paragraph with an empty line.
 
-# FORMATTING
+## FORMATTING
 
 Text can be made *bold* or _underlined_ with asterisks and underscores: \*bold\*
 or \_underlined\_.
 
-# INDENTATION
+## INDENTATION
 
 You may indent lines with tab characters (*\\t*) to indent them by 4 spaces in
 the output. Indented lines may not contain headers.
 
-# LISTS
+	The result looks something like this.
+
+	You may use multiple lines and most _formatting_.
+
+Deindent to return to normal.
+
+## LISTS
 
 You may start bulleted lists with dashes, like so:
 
 ```
 - Item 1
 - Item 2
+	- Subitem 1
+	- Subitem 2
 - Item 3
 ```
 
-You may also use numbered lists like so:
+The result looks like this:
 
-```
+- Item 1
-1. Item 1
+- Item 2
-2. Item 2
+	- Subitem 1
-3. Item 3
+	- Subitem 2
-```
+- Item 3
 
-# LITERAL TEXT
+## 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
@@ -78,3 +93,9 @@ _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.
+
+# 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
+sir@cmpwn.com.

src/main.c

@@ -1,5 +1,6 @@
 #include <assert.h>
 #include <ctype.h>
+#include <stdbool.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <time.h>
@@ -97,6 +98,9 @@ static void parse_text(struct parser *p) {
 		case '_':
 			parse_format(p, FORMAT_UNDERLINE);
 			break;
+		case '\n':
+			utf8_fputch(p->output, ch);
+			return;
 		case '.':
 			if (!i) {
 				// Escape . if it's the first character
@@ -108,9 +112,6 @@ static void parse_text(struct parser *p) {
 			utf8_fputch(p->output, ch);
 			break;
 		}
-		if (ch == '\n') {
-			break;
-		}
 		++i;
 	}
 }
@@ -146,46 +147,75 @@ static void parse_heading(struct parser *p) {
 	}
 }
 
-static int parse_indent(struct parser *p) {
+static int parse_indent(struct parser *p, int *indent) {
 	int i = 0;
 	uint32_t ch;
 	while ((ch = parser_getch(p)) == '\t') {
 		++i;
 	}
 	parser_pushch(p, ch);
+	if (i == *indent - 1) {
+		roff_macro(p, "RE", NULL);
+	} else if (i == *indent + 1) {
+		roff_macro(p, "RS", "4", NULL);
+	} else if (i != *indent && ch == '\t') {
+		parser_fatal(p, "(De)indented by an amount greater than 1");
+	}
+	*indent = i;
 	return i;
 }
 
-static void parse_document(struct parser *p) {
+static void parse_list(struct parser *p, int *indent) {
 	uint32_t ch;
-	int indent = 0;
+	if ((ch = parser_getch(p)) != ' ') {
+		parser_fatal(p, "Expected space before start of list entry");
+	}
+	roff_macro(p, "IP", "\\(bu", "4", NULL);
+	parse_text(p);
 	do {
-		int i = parse_indent(p);
+		parse_indent(p, indent);
-		if (i == indent - 1) {
+		if ((ch = parser_getch(p)) == UTF8_INVALID) {
-			roff_macro(p, "RE", NULL);
-		} else if (i == indent + 1) {
-			roff_macro(p, "RS", "4", NULL);
-		} else if (i != indent && ch == '\t') {
-			parser_fatal(p, "(De)indented by an amount greater than 1");
-		}
-		indent = i;
-
-		ch = parser_getch(p);
-		if (ch == UTF8_INVALID) {
 			break;
 		}
-
+		switch (ch) {
-		if (indent != 0) {
+		case ' ':
-			// Only text is allowed at this point
+			break;
-			parser_pushch(p, ch);
+		case '-':
+			if ((ch = parser_getch(p)) != ' ') {
+				parser_fatal(p, "Expected space before start of list entry");
+			}
+			fprintf(p->output, ".sp -1\n");
+			roff_macro(p, "IP", "\\(bu", "4", NULL);
 			parse_text(p);
-			continue;
+			break;
+		default:
+			fprintf(p->output, "\n");
+			parser_pushch(p, ch);
+			return;
 		}
+	} while (ch != UTF8_INVALID);
+}
 
+static void parse_document(struct parser *p) {
+	uint32_t ch;
+	int indent = 0;
+	do {
+		parse_indent(p, &indent);
+		if ((ch = parser_getch(p)) == UTF8_INVALID) {
+			break;
+		}
 		switch (ch) {
 		case '#':
+			if (indent != 0) {
+				parser_pushch(p, ch);
+				parse_text(p);
+				break;
+			}
 			parse_heading(p);
 			break;
+		case '-':
+			parse_list(p, &indent);
+			break;
 		case ' ':
 			parser_fatal(p, "Tabs are required for indentation");
 			break;