Simplify schema by using XXZ[0-9] instead of XXZ[A-J] for leading digits

Author: ad-si

Elm/DoubleXEncode.elm

@@ -232,80 +232,6 @@ charDecode char =
             '\u{0000}'
 
 
-hexDigitEncode : Char -> Char
-hexDigitEncode digit =
-    case digit of
-        '0' ->
-            'A'
-
-        '1' ->
-            'B'
-
-        '2' ->
-            'C'
-
-        '3' ->
-            'D'
-
-        '4' ->
-            'E'
-
-        '5' ->
-            'F'
-
-        '6' ->
-            'G'
-
-        '7' ->
-            'H'
-
-        '8' ->
-            'I'
-
-        '9' ->
-            'J'
-
-        _ ->
-            '\u{0000}'
-
-
-hexDigitDecode : Char -> Char
-hexDigitDecode digit =
-    case digit of
-        'A' ->
-            '0'
-
-        'B' ->
-            '1'
-
-        'C' ->
-            '2'
-
-        'D' ->
-            '3'
-
-        'E' ->
-            '4'
-
-        'F' ->
-            '5'
-
-        'G' ->
-            '6'
-
-        'H' ->
-            '7'
-
-        'I' ->
-            '8'
-
-        'J' ->
-            '9'
-
-        _ ->
-            '\u{0000}'
-
-
 hexShiftEncode : Char -> Char
 hexShiftEncode digit =
     case digit of
@@ -482,7 +408,7 @@ doubleXEncodeWithOptions encodeOptions string =
                 |> String.fromList
 
         encodeDigit digit =
-            String.fromList [ 'X', 'X', 'Z', hexDigitEncode digit ]
+            String.fromList [ 'X', 'X', 'Z', digit ]
     in
     if encodeOptions.encodeLeadingDigit then
         case String.toList string of
@@ -573,9 +499,6 @@ doubleXDecode text =
                 ( noXX
                     |> String.dropLeft 1
                     |> String.left 1
-                    |> String.toList
-                    |> List.map hexDigitDecode
-                    |> String.fromList
                 , noXX |> String.dropLeft 2
                 )
 

Elm/Test.elm

@@ -35,6 +35,18 @@ main =
         expectEncodeGql a b =
             getEncoderTest doubleXEncodeGql a b
 
+        expectDecode a b =
+            if doubleXDecode a /= b then
+                "❌ DECODE ERROR: "
+                    ++ a
+                    ++ " | "
+                    ++ doubleXDecode a
+                    ++ " /= "
+                    ++ b
+
+            else
+                "✅ " ++ a ++ ": " ++ doubleXDecode a ++ " == " ++ b
+
         blnsFiltered =
             blns
                 |> List.filter
@@ -80,7 +92,9 @@ main =
     , expectEncode "Multi Byte Emoji: 👨\u{200D}🦲"
         "MultiXX0ByteXX0EmojiXXGXX0XXbpegiXXacaanXXbpjlc"
     , expectEncodeGql "__Schema" "XXRXXRSchema"
-    , expectEncodeGql "0test" "XXZAtest"
+    , expectEncodeGql "also__middle" "alsoXXRXXRmiddle"
+    , expectEncodeGql "0test" "XXZ0test"
+    , expectDecode "alsoXXZ0middle" "also0middle"
     , "========== ENCODE - DECODE =========="
     , blnsFiltered |> List.map testEncodeDecode |> String.join ""
     , "========== ENCODE GQL - DECODE =========="

Haskell/DoubleXEncode.hs

@@ -109,38 +109,6 @@ charDecode = \case
   _ -> '\0'
 
 
-hexDigitEncode :: Char -> Char
-hexDigitEncode = \case
-  '0' -> 'A'
-  '1' -> 'B'
-  '2' -> 'C'
-  '3' -> 'D'
-  '4' -> 'E'
-  '5' -> 'F'
-  '6' -> 'G'
-  '7' -> 'H'
-  '8' -> 'I'
-  '9' -> 'J'
-
-  _ -> '\0'
-
-
-hexDigitDecode :: Char -> Char
-hexDigitDecode = \case
-  'A' -> '0'
-  'B' -> '1'
-  'C' -> '2'
-  'D' -> '3'
-  'E' -> '4'
-  'F' -> '5'
-  'G' -> '6'
-  'H' -> '7'
-  'I' -> '8'
-  'J' -> '9'
-
-  _ -> '\0'
-
-
 hexShiftEncode :: Char -> Char
 hexShiftEncode = \case
   '0' -> 'a'
@@ -220,7 +188,7 @@ doubleXEncodeWithOptions encodeOptions text =
       & T.pack
 
     encodeDigit digit =
-      T.pack ['X', 'X', 'Z', hexDigitEncode digit]
+      T.pack ['X', 'X', 'Z', digit]
   in
     if encodeOptions&encodeLeadingDigit
     then
@@ -306,9 +274,6 @@ doubleXDecode text =
                 (noXX
                     & T.drop 1
                     & T.take 1
-                    & T.unpack
-                    <&> hexDigitDecode
-                    & T.pack
                 , T.drop 2 noXX
                 )
 

Haskell/Test.hs

@@ -38,6 +38,14 @@ main = do
     expectEncodeGql a b =
       getEncoderTest doubleXEncodeGql a b
 
+    expectDecode a b = T.putStrLn $
+        if doubleXDecode a /= b
+        then
+            "❌ DECODE ERROR: "
+            <> a <> " | " <> doubleXDecode a <> " /= " <> b
+        else
+            "✅ " <> a <> ": " <> doubleXDecode a <> " == " <> b
+
   expectEncode "camelCaseId" "camelCaseId"
   expectEncode "snake_case_id" "snake_case_id"
   expectEncode "__Schema" "__Schema"
@@ -55,7 +63,9 @@ main = do
     "MultiXX0ByteXX0EmojiXXGXX0XXbpegiXXacaanXXbpjlc"
 
   expectEncodeGql "__Schema" "XXRXXRSchema"
-  expectEncodeGql "0test" "XXZAtest"
+  expectEncodeGql "also__middle" "alsoXXRXXRmiddle"
+  expectEncodeGql "0test" "XXZ0test"
+  expectDecode "alsoXXZ0middle" "also0middle"
 
   -- https://github.com/minimaxir/big-list-of-naughty-strings
   txtFile <- readFile "Haskell/blns.txt"
@@ -85,10 +95,12 @@ main = do
         <> "\""
       else "✅"
 
-  T.putStrLn "\n\n========== ENCODE - DECODE =========="
+  T.putStrLn "\n\n============ ENCODE - DECODE ============"
   blnsFiltered <&> testEncodeDecode <&> T.putStr & sequence_
 
   T.putStrLn "\n\n========== ENCODE GQL - DECODE =========="
   blnsFiltered <&> testEncodeGqlDecode <&> T.putStr & sequence_
 
+  T.putStrLn "\n\n============== HASKELL END ==============\n\n"
+
   pure ()

JavaScript/index.ts

@@ -48,25 +48,6 @@ const charDecode = Object.fromEntries(
     .map(([k, v]) => [v, k])
 )
 
-const hexDigitEncode = {
-  "0": "A",
-  "1": "B",
-  "2": "C",
-  "3": "D",
-  "4": "E",
-  "5": "F",
-  "6": "G",
-  "7": "H",
-  "8": "I",
-  "9": "J",
-}
-
-const hexDigitDecode = Object.fromEntries(
-  Object
-    .entries(hexDigitEncode)
-    .map(([k, v]) => [v, k])
-)
-
 const hexShiftEncode = {
   "0": "a",
   "1": "b",
@@ -102,9 +83,8 @@ export function doubleXEncode (
 ): string {
   if (options.encodeLeadingDigit) {
     const firstChar = str.slice(0, 1)
-    const firstDigitHex = hexDigitEncode[firstChar]
-    const firstCharEncoded = firstDigitHex
-      ? "XXZ" + firstDigitHex
+    const firstCharEncoded = /^[0-9]$/.test(firstChar)
+      ? "XXZ" + firstChar
       : firstChar
 
     options.encodeLeadingDigit = false
@@ -148,7 +128,7 @@ export function doubleXDecode (str: string): string {
   const strNorm = str.replaceAll("XXXXXX", "XXYXXY")
 
   return strNorm
-    .split(/(XXY|XXZ[A-J]|XX[0-9A-W]|XX[a-p]{5})/)
+    .split(/(XXY|XXZ[0-9]|XX[0-9A-W]|XX[a-p]{5})/)
     .filter(Boolean)  // Remove empty strings
     .map(word =>
       word.startsWith("XX")
@@ -164,7 +144,7 @@ export function doubleXDecode (str: string): string {
               )
             )
           : (word.slice(2, 3) == "Z")
-            ? hexDigitDecode[word.slice(3, 4)]
+            ? word.slice(3, 4)
             : charDecode[word.slice(2, 3)]
         )
       : word

makefile

@@ -1,3 +1,12 @@
+.PHONY: help
+help: makefile
+	@tail -n +4 makefile | grep ".PHONY"
+
+
+.PHONY: test
+test: test-elm test-haskell test-javascript
+
+
 .PHONY: test-elm
 test-elm:
 		cd Elm && elm make Test.elm --output=elm-tests.js
@@ -10,8 +19,4 @@ test-haskell:
 
 .PHONY: test-javascript
 test-javascript:
-		deno run JavaScript/test.ts
-
-
-.PHONY: test
-test: test-elm test-haskell test-javascript
+		deno --unstable run JavaScript/test.ts

readme.md

@@ -35,7 +35,7 @@ With encoding of leading digit and double underscore activated
 
 Input | Output
 ------|-------
-`1FileFormat` | `XXZAFileFormat`
+`1FileFormat` | `XXZ1FileFormat`
 `__index__` | `XXRXXRindexXXRXXR`
 
 
@@ -53,7 +53,7 @@ The encoding scheme is based on the following rules:
     `a` to `p` instead of `0` to `f`.
 
 If the optional leading digit encoding is enabled,
-a leading digit is encoded as `XXZ[A-J]`, where `A` is `0` and `J` is `9`.
+a leading digit is encoded as `XXZ[0-9]`.
 
 If the optional double underscore encoding is enabled,
 double underscores are encoded as `XXRXXR`.