From aa2edcdf68676c3839593309af97b335cb8388b8 Mon Sep 17 00:00:00 2001
From: Linnnus <linnnus@users.noreply.github.com>
Date: Sun, 4 Feb 2024 18:45:16 +0100
Subject: feat(creole): escape special HTML characters

---
 src/creole-test.c | 10 +++++-----
 src/creole.c      | 32 +++++++++++++++++++++++++++++++-
 2 files changed, 36 insertions(+), 6 deletions(-)

diff --git a/src/creole-test.c b/src/creole-test.c
index 835ee42..3020c51 100644
--- a/src/creole-test.c
+++ b/src/creole-test.c
@@ -10,6 +10,11 @@
 struct {
 	const char *name, *input, *output;
 } tests[] = {
+	{
+		.name    =  "Basic paragraph markup",
+		.input   =  "Basic paragraph test with <, >, & and \"",
+		.output  =  "<p>Basic paragraph test with &lt;, &gt;, &amp; and &quot;</p>"
+	},
 	{
 		.name    =  "Two paragraphs next to each other.",
 		.input   =  "Hello,\n\nworld!",
@@ -51,11 +56,6 @@ struct {
 		.output  =  "<p>======= Header =</p>"
 	},
 #if 0
-	{
-		.name    =  "Basic paragraph markup",
-		.input   =  "Basic paragraph test with <, >, & and \"",
-		.output  =  "<p>Basic paragraph test with &lt;, &gt;, &amp; and &quot;</p>"
-	},
 	{
 		.name    =  "Simple unordered list",
 		.input   =  "* list item\n*list item 2",
diff --git a/src/creole.c b/src/creole.c
index b3d1a87..7472a3b 100644
--- a/src/creole.c
+++ b/src/creole.c
@@ -6,6 +6,7 @@
 #include <stdbool.h>
 #include <stdio.h>
 #include <stdlib.h>
+#include <string.h>
 
 #define LENGTH(x)  (sizeof(x)/sizeof((x)[0]))
 
@@ -14,6 +15,7 @@
 void process(const char *begin, const char *end, bool new_block, FILE *out);
 int do_headers(const char *begin, const char *end, bool new_block, FILE *out);
 int do_paragraph(const char *begin, const char *end, bool new_block, FILE *out);
+int do_replacements(const char *begin, const char *end, bool new_block, FILE *out);
 
 // A parser takes a (sub)string and returns the number of characters consumed, if any.
 //
@@ -21,7 +23,7 @@ int do_paragraph(const char *begin, const char *end, bool new_block, FILE *out);
 // The sign of the return value determines whether a new block should begin, after the consumed text.
 typedef int (* parser_t)(const char *begin, const char *end, bool new_block, FILE *out);
 
-static parser_t parsers[] = { do_headers, do_paragraph };
+static parser_t parsers[] = { do_headers, do_paragraph, do_replacements };
 
 int do_headers(const char *begin, const char *end, bool new_block, FILE *out) {
 	if (!new_block) { // Headers are block-level elements.
@@ -87,6 +89,34 @@ found_double_newline:
 	return -(stop - begin);
 }
 
+static struct {
+	const char *from, *to;
+} replacements[] = {
+	// Escaped special characters
+	// TODO (e.g. "\\{" should become "{" so users can escape and void it being parsed as markup
+	// Characters that have special meaning in HTML
+	{"<", "&lt;"},
+	{">", "&gt;"},
+	{"\"", "&quot;"},
+	{"&", "&amp;"},
+};
+
+int do_replacements(const char *begin, const char *end, bool new_block, FILE *out)
+{
+	for (unsigned i = 0; i < LENGTH(replacements); ++i) {
+		size_t length = strlen(replacements[i].from);
+		if (end - begin < length) {
+			continue;
+		}
+		if (strncmp(replacements[i].from, begin, length) == 0) {
+			fputs(replacements[i].to, out);
+			return length;
+		}
+	}
+
+	return 0;
+}
+
 void process(const char *begin, const char *end, bool new_block, FILE *out) {
 	const char *p = begin;
 	while (p < end) {
-- 
cgit v1.2.3