gemfedwiki

commit 4e759ecf7cdf23d3a40a1a3b963fd88b230d5f13
parent 532d9c2f058f5ba84a83831f4768493bd5ae93ec
Author: William Casarin <jb55@jb55.com>
Date:   Wed, 27 Oct 2021 22:27:52 -0700

revamp link parsing

Diffstat:
M
index.js
 | 
101
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-------

1 file changed, 92 insertions(+), 9 deletions(-)
diff --git a/index.js b/index.js
@@ -7,14 +7,88 @@ function slugify(text)
 	return text.replace(/\s/g, '-').replace(/[^A-Za-z0-9-]/g, '').toLowerCase()
 }
 
-function parse_links(text)
+function consume_until(cursor, c)
 {
+	for (; cursor.pos < cursor.str.length; cursor.pos++) {
+		if (cursor.str[cursor.pos] == c[0])
+			return true
+	}
+	return false
+}
+
+function pull_char(cursor)
+{
+	if (cursor.pos >= cursor.str.length)
+		return null
+	return cursor.str[cursor.pos++]
+}
+
+function consume_char(cursor, match)
+{
+	return pull_char(cursor) === match[0]
+}
+
+function parse_slug_link(cursor, links, root, start)
+{
+	let i = cursor.pos
+	if (!consume_until(cursor, ']')) return false
+	const name = cursor.str.slice(i, cursor.pos)
+	const url = `${root}/${slugify(name)}`
+	if (!consume_char(cursor, ']')) return false
+	if (!consume_char(cursor, ']')) return false
+	const orig = cursor.str.slice(start, cursor.pos)
+	links.push({name, url, orig})
+	return true
+}
+
+function consume_whitespace(cursor)
+{
+	for (; cursor.pos < cursor.str.length; cursor.pos++) {
+		if (cursor.str[cursor.pos] !== ' ')
+			return;
+	}
+}
+
+function parse_link(cursor, links, root)
+{
+	let c, i, url, name, start
+
+	start = cursor.pos
+
+	if (!consume_char(cursor, "["))
+		return false
+
+	if ((c = pull_char(cursor)) && c == '[')
+		return parse_slug_link(cursor, links, root, start)
+
+	cursor.pos--
+	consume_whitespace(cursor)
+	i = cursor.pos
+
+	if (!consume_until(cursor, " ")) return false
+	url = cursor.str.slice(i, cursor.pos++)
+	i = cursor.pos
+
+	if (!consume_until(cursor, "]")) return false
+	name = cursor.str.slice(i, cursor.pos++).trim()
+
+	const orig = cursor.str.slice(start, cursor.pos)
+
+	links.push({ name, url, orig })
+	return true
+}
+
+function parse_links(text, root)
+{
+	let cursor = {pos: 0, str: text}
 	let links = []
-	const matches = text.matchAll(/\[\[([^\]]+)\]\]/g)
-	for (const match of matches) {
-		const name = match[1]
-		const slug = slugify(name)
-		links.push({slug, name})
+
+	for (; cursor.pos < cursor.str.length;) {
+		if (!consume_until(cursor, "["))
+			return links
+
+		if (!parse_link(cursor, links, root))
+			return links
 	}
 
 	return links
@@ -22,19 +96,28 @@ function parse_links(text)
 
 function process_item(root, lines, item)
 {
+	let first = true;
+	let i = 0
+
 	if (item.type === 'paragraph' || item.type === 'markdown' || item.type === 'html') {
 		const text = item.text.trim()
-		const links = parse_links(text)
+		const links = parse_links(text, root)
 
 		lines.push('')
 		lines.push(text)
+		i = lines.length-1
 
 		if (links.length !== 0)
 			lines.push('')
 
-		for (const {slug, name} of links) {
-			lines.push(`=> ${root}/${slug} ${name}`)
+		for (const {url, name, orig} of links) {
+			lines[i] = lines[i].replace(orig, name)
+			if (lines[i] === name)
+				// just remove it if it's the same as the link
+				lines.splice(i, 2)
+			lines.push(`=> ${url} ${name}`)
 		}
+
 	}
 }