feat(single): Implement autolinks

1 files changed, 35 insertions, 2 deletions

diff --git a/single.awk b/single.awk
index 3568844..0add85e 100644
--- a/single.awk
+++ b/single.awk
@@ -18,8 +18,8 @@
 #     [x] Images
 #         [x] Inline
 #         [x] Reference
-# [ ] Other
-#     [ ] Autolinks
+# [.] Other
+#     [x] Autolinks
 #     [ ] Escapes
 #
 # [ ] Non-standard features
@@ -72,6 +72,15 @@ function restore_match() {
 	RLENGTH = backup_rlength
 }
 
+BEGIN {
+	## This is used to convert email autolinks into a series of hex values
+	## Escaped output of: awk 'BEGIN {for (i = 32; i <= 126; ++i){ printf("%c", i) } }'
+	ascii = " !\"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_`abcdefghijklmnopqrstuvwxyz{|}~"
+	for (i = 1; i < 97; ++i) {
+		char_to_dec[char(ascii, i)] = 31 + i
+	}
+}
+
 {
 	not_paragraph_p1 = not_paragraph
 	not_paragraph = 0
@@ -95,6 +104,30 @@ function restore_match() {
 	next
 }
 
+# Automatic links
+/<[^ \t<>]+>/ {
+	## URL: <uri_scheme:...>
+	while (match($0, /<[A-Za-z][-A-Za-z0-9+.]*:[^ \t<>]+>/)) {
+		href = substr($0, RSTART + 1, RLENGTH - 2)
+		sub_match("<a href=\"" href "\">" href "</a>")
+	}
+
+	## Email addresses, with support for quoted local part "..."@...
+	## and ip address literal ...@[...]
+	while (match($0, /<([^ \t<>]+|".+")@([-A-Za-z0-9]+\.[^ \t<>]+|\[[0-9A-Fa-f.:]+\])>/)) {
+		href = substr($0, RSTART + 1, RLENGTH - 2)
+		href_masked = ""
+		## The original markdown blogpost calls for convertions of all
+		## email characters into their hex values. Allegedly, this fools
+		## many autoharvesting spam bots.
+		for (i = length(href); i > 0; --i) {
+			href_masked = sprintf("&#x%X;", char_to_dec[char(href, i)]) href_masked
+		}
+		## href="mailto:..."
+		sub_match("<a href=\"&#x6D;&#x61;&#x69;&#x6C;&#x74;&#x6F;&#x3A;" href_masked "\">" href_masked "</a>")
+	}
+}
+
 # Blockquotes
 blockquote_level > 0 && empty0 {
 	for (P1 = P1 ORS; blockquote_level > 0; --blockquote_level) {