LibTextCodec: Implement a Windows-1251 decoder

This encoding (a superset of ascii that adds in the cyrillic alphabet)
is currently the third most used encoding on the web, and because
cyrillic glyphs were added by Dmitrii Trifonov recently, we can now
support it as well :^)

2 files changed, 38 insertions, 0 deletions

diff --git a/Userland/Libraries/LibTextCodec/Decoder.cpp b/Userland/Libraries/LibTextCodec/Decoder.cpp
index f82bb029eb..37a28d5999 100644
--- a/Userland/Libraries/LibTextCodec/Decoder.cpp
+++ b/Userland/Libraries/LibTextCodec/Decoder.cpp
@@ -51,6 +51,14 @@ HebrewDecoder& hebrew_decoder()
     return *decoder;
 }
 
+CyrillicDecoder& cyrillic_decoder()
+{
+    static CyrillicDecoder* decoder = nullptr;
+    if (!decoder)
+        decoder = new CyrillicDecoder;
+    return *decoder;
+}
+
 }
 
 Decoder* decoder_for(const String& a_encoding)
@@ -66,6 +74,8 @@ Decoder* decoder_for(const String& a_encoding)
         return &latin2_decoder();
     if (encoding.equals_ignoring_case("windows-1255"))
         return &hebrew_decoder();
+    if (encoding.equals_ignoring_case("windows-1251"))
+        return &cyrillic_decoder();
     dbgln("TextCodec: No decoder implemented for encoding '{}'", a_encoding);
     return nullptr;
 }
@@ -304,4 +314,27 @@ String HebrewDecoder::to_utf8(const StringView& input)
     return builder.to_string();
 }
 
+String CyrillicDecoder::to_utf8(const StringView& input)
+{
+    static constexpr Array<u32, 128> translation_table = {
+        0x402, 0x403, 0x201A, 0x453, 0x201E, 0x2026, 0x2020, 0x2021, 0x20AC, 0x2030, 0x409, 0x2039, 0x40A, 0x40C, 0x40B, 0x40F,
+        0x452, 0x2018, 0x2019, 0x201C, 0x201D, 0x2022, 0x2013, 0x2014, 0xFFFD, 0x2122, 0x459, 0x203A, 0x45A, 0x45C, 0x45B, 0x45F,
+        0xA0, 0x40E, 0x45E, 0x408, 0xA4, 0x490, 0xA6, 0xA7, 0x401, 0xA9, 0x404, 0xAB, 0xAC, 0xAD, 0xAE, 0x407,
+        0xB0, 0xB1, 0x406, 0x456, 0x491, 0xB5, 0xB6, 0xB7, 0x451, 0x2116, 0x454, 0xBB, 0x458, 0x405, 0x455, 0x457,
+        0x410, 0x411, 0x412, 0x413, 0x414, 0x415, 0x416, 0x417, 0x418, 0x419, 0x41A, 0x41B, 0x41C, 0x41D, 0x41E, 0x41F,
+        0x420, 0x421, 0x422, 0x423, 0x424, 0x425, 0x426, 0x427, 0x428, 0x429, 0x42A, 0x42B, 0x42C, 0x42D, 0x42E, 0x42F,
+        0x430, 0x431, 0x432, 0x433, 0x434, 0x435, 0x436, 0x437, 0x438, 0x439, 0x43A, 0x43B, 0x43C, 0x43D, 0x43E, 0x43F,
+        0x440, 0x441, 0x442, 0x443, 0x444, 0x445, 0x446, 0x447, 0x448, 0x449, 0x44A, 0x44B, 0x44C, 0x44D, 0x44E, 0x44F
+    };
+    StringBuilder builder(input.length());
+    for (unsigned char ch : input) {
+        if (ch < 0x80) { // Superset of ASCII
+            builder.append(ch);
+        } else {
+            builder.append_code_point(translation_table[ch - 0x80]);
+        }
+    }
+    return builder.to_string();
+}
+
 }
diff --git a/Userland/Libraries/LibTextCodec/Decoder.h b/Userland/Libraries/LibTextCodec/Decoder.h
index b7b5c0ab0e..6ea4147b7e 100644
--- a/Userland/Libraries/LibTextCodec/Decoder.h
+++ b/Userland/Libraries/LibTextCodec/Decoder.h
@@ -43,6 +43,11 @@ public:
     virtual String to_utf8(const StringView&) override;
 };
 
+class CyrillicDecoder final : public Decoder {
+public:
+    virtual String to_utf8(const StringView&) override;
+};
+
 Decoder* decoder_for(const String& encoding);
 String get_standardized_encoding(const String& encoding);
 bool is_standardized_encoding(const String& encoding);