hare

commit b8b0327ef2249b846f9f0cde2ac4d0f17af2f9d2
parent 79b55532d74c4b1b0d6b1a1a1846073a4363f60a
Author: Drew DeVault <sir@cmpwn.com>
Date:   Fri,  5 Feb 2021 12:26:56 -0500

encoding: add tests

Diffstat:
M
encoding/utf8/decode.ha
 | 
27
+++++++++++++++++++++++++++
M
encoding/utf8/encode.ha
 | 
15
+++++++++++++++

2 files changed, 42 insertions(+), 0 deletions(-)
diff --git a/encoding/utf8/decode.ha b/encoding/utf8/decode.ha
@@ -60,6 +60,33 @@ export fn next(d: *decoder) (rune | void | more | invalid) = {
 	return r: rune;
 };
 
+@test fn decode() void = {
+	const input = [
+		0xE3u8, 0x81u8, 0x93u8, 0xE3u8, 0x82u8, 0x93u8, 0xE3u8, 0x81u8,
+		0xABu8, 0xE3u8, 0x81u8, 0xA1u8, 0xE3u8, 0x81u8, 0xAFu8, 0x00u8,
+	];
+	const expected = ['こ', 'ん', 'に', 'ち', 'は', '\0'];
+	let decoder = decode(input);
+	for (let i = 0z; i < len(expected); i += 1z) {
+		match (next(&decoder)) {
+			r: rune => assert(r == expected[i]),
+			invalid => abort(),
+			more    => abort(),
+			void    => abort(),
+		};
+	};
+	assert(next(&decoder) is void);
+
+	// TODO: Test more invalid sequences
+	const invalid = [0xA0u8, 0xA1u8];
+	decoder = decode(invalid);
+	assert(next(&decoder) is invalid);
+
+	const incomplete = [0xE3u8, 0x81u8];
+	decoder = decode(incomplete);
+	assert(next(&decoder) is more);
+};
+
 // Returns true if a given string or byte slice contains only valid UTF-8
 // sequences. Note that Hare strings (str) are always valid UTF-8 - if this
 // returns false for a str type, something funny is going on.
diff --git a/encoding/utf8/encode.ha b/encoding/utf8/encode.ha
@@ -24,3 +24,18 @@ export fn encode_rune(r: rune) []u8 = {
 	buf[0] = ch: u8 | first;
 	return buf[..n];
 };
+
+@test fn encode() void = {
+	const expected: [_][]u8 = [
+		[0u8],
+		[0x25u8],
+		[0xE3u8, 0x81u8, 0x93u8],
+	];
+	const inputs = ['\0', '%', 'こ'];
+	for (let i = 0z; i < len(inputs); i += 1z) {
+		const out = encode_rune(inputs[i]);
+		for (let j = 0z; j < len(expected[i]); j += 1z) {
+			assert(out[j] == expected[i][j]);
+		};
+	};
+};