Fixed quoting

crobox · May 28, 2024 · 1c5925a · 1c5925a
1 parent 32eca06
commit 1c5925a
Show file tree

Hide file tree

Showing 2 changed files with 42 additions and 23 deletions.
diff --git a/dsl/src/main/scala/com/crobox/clickhouse/dsl/language/SplitMergeFunctionTokenizer.scala b/dsl/src/main/scala/com/crobox/clickhouse/dsl/language/SplitMergeFunctionTokenizer.scala
@@ -8,23 +8,31 @@ trait SplitMergeFunctionTokenizer {
   def tokenizeSplitMergeFunction(col: SplitMergeFunction[_])(implicit ctx: TokenizeContext): String = col match {
     case SplitByChar(sep: StringColMagnet[_], col: StringColMagnet[_]) =>
       // Some small optimizations
-      val separator = tokenizeColumn(sep.column)
-      if (separator.length == 3) {
-        val s = separator.charAt(1).toInt
+      val separator = sep.column match {
+        case c: Const[_] => c.const.asInstanceOf[String]
+      }
+
+      if (separator.length == 1) {
+        val s = separator.charAt(0).toInt
 
         // https://en.wikipedia.org/wiki/List_of_Unicode_characters
-        if (s >= 32 && s <= 126) {
-          s"splitByChar($separator,${tokenizeColumn(col.column)})"
+        // 34 == DoubleQuote ("), 39 == Single Quote ('),
+        // 47 == Forward Slash (/), 92 == Backward Slash (\\)
+        // 96 == Grave Accent (` under tilde)
+        if (s == 34 || s == 39 || s == 92 || s == 96) {
+          s"splitByChar(char($s), ${tokenizeColumn(col.column)})"
+        } else if (s >= 32 && s <= 126) {
+          s"splitByChar(${tokenizeColumn(sep.column)}, ${tokenizeColumn(col.column)})"
         } else {
-          s"splitByChar(char($s),${tokenizeColumn(col.column)})"
+          s"splitByChar(char($s), ${tokenizeColumn(col.column)})"
         }
       } else {
-        s"splitByString($separator,${tokenizeColumn(col.column)})"
+        s"splitByString(${tokenizeColumn(sep.column)}, ${tokenizeColumn(col.column)})"
       }
     case SplitByString(sep: StringColMagnet[_], col: StringColMagnet[_]) =>
-      s"splitByString(${tokenizeColumn(sep.column)},${tokenizeColumn(col.column)})"
+      s"splitByString(${tokenizeColumn(sep.column)}, ${tokenizeColumn(col.column)})"
     case ArrayStringConcat(col: ArrayColMagnet[_], sep: StringColMagnet[_]) =>
-      s"arrayStringConcat(${tokenizeColumn(col.column)},${tokenizeColumn(sep.column)})"
+      s"arrayStringConcat(${tokenizeColumn(col.column)}, ${tokenizeColumn(sep.column)})"
     case AlphaTokens(col: StringColMagnet[_]) => s"alphaTokens(${tokenizeColumn(col.column)})"
   }
 }
diff --git a/dsl/src/test/scala/com/crobox/clickhouse/dsl/language/SplitMergeFunctionTokenizerTest.scala b/dsl/src/test/scala/com/crobox/clickhouse/dsl/language/SplitMergeFunctionTokenizerTest.scala
@@ -1,28 +1,39 @@
 package com.crobox.clickhouse.dsl.language
 
 import com.crobox.clickhouse.dsl._
-import com.crobox.clickhouse.{dsl, DslTestSpec}
+import com.crobox.clickhouse.{DslTestSpec, dsl}
 
 class SplitMergeFunctionTokenizerTest extends DslTestSpec {
   val FIELD_DELIMITER: Char = '\u001F'
 
   it should "splitByChar using special character" in {
 
     toSQL(select(dsl.splitByChar(FIELD_DELIMITER.toString, "abcd"))) should matchSQL(
-      "SELECT splitByChar(char(31),'abcd')"
+      "SELECT splitByChar(char(31), 'abcd')"
     )
-    toSQL(select(dsl.splitByChar(31.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(31),'abcd')")
-    toSQL(select(dsl.splitByChar(32.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(' ','abcd')")
-    toSQL(select(dsl.splitByChar(126.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar('~','abcd')")
-    toSQL(select(dsl.splitByChar(127.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(127),'abcd')")
-    toSQL(select(dsl.splitByChar(128.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(128),'abcd')")
-    toSQL(select(dsl.splitByChar(159.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(159),'abcd')")
+    toSQL(select(dsl.splitByChar(31.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(31), 'abcd')")
+    toSQL(select(dsl.splitByChar(32.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(' ', 'abcd')")
+    toSQL(select(dsl.splitByChar(126.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar('~', 'abcd')")
+    toSQL(select(dsl.splitByChar(127.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(127), 'abcd')")
+    toSQL(select(dsl.splitByChar(128.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(128), 'abcd')")
+    toSQL(select(dsl.splitByChar(159.toChar.toString, "abcd"))) should matchSQL("SELECT splitByChar(char(159), 'abcd')")
 
-    toSQL(select(dsl.splitByChar(",", "abcd"))) should matchSQL("SELECT splitByChar(',','abcd')")
-    toSQL(select(dsl.splitByChar("a", "abcd"))) should matchSQL("SELECT splitByChar('a','abcd')")
-    toSQL(select(dsl.splitByChar("ab", "abcd"))) should matchSQL("SELECT splitByString('ab','abcd')")
-    toSQL(select(dsl.splitByChar("L", "abcd"))) should matchSQL("SELECT splitByChar('L','abcd')")
-    toSQL(select(dsl.splitByChar("$", "abcd"))) should matchSQL("SELECT splitByChar('$','abcd')")
-    toSQL(select(dsl.splitByChar("-", "abcd"))) should matchSQL("SELECT splitByChar('-','abcd')")
+    toSQL(select(dsl.splitByChar(",", "abcd"))) should matchSQL("SELECT splitByChar(',', 'abcd')")
+    toSQL(select(dsl.splitByChar("a", "abcd"))) should matchSQL("SELECT splitByChar('a', 'abcd')")
+    toSQL(select(dsl.splitByChar("L", "abcd"))) should matchSQL("SELECT splitByChar('L', 'abcd')")
+    toSQL(select(dsl.splitByChar("$", "abcd"))) should matchSQL("SELECT splitByChar('$', 'abcd')")
+    toSQL(select(dsl.splitByChar("-", "abcd"))) should matchSQL("SELECT splitByChar('-', 'abcd')")
+    toSQL(select(dsl.splitByChar("!", "abcd"))) should matchSQL("SELECT splitByChar('!', 'abcd')")
+
+    // special 'not quoted' characters, see ClickhouseStatement.UnquotedIdentifier
+    toSQL(select(dsl.splitByChar("\"", "abcd"))) should matchSQL("SELECT splitByChar(char(34), 'abcd')")
+    toSQL(select(dsl.splitByChar("'", "abcd"))) should matchSQL("SELECT splitByChar(char(39), 'abcd')")
+    toSQL(select(dsl.splitByChar("`", "abcd"))) should matchSQL("SELECT splitByChar(char(96), 'abcd')")
+    toSQL(select(dsl.splitByChar("\\", "abcd"))) should matchSQL("SELECT splitByChar(char(92), 'abcd')")
+    toSQL(select(dsl.splitByChar("/", "abcd"))) should matchSQL("SELECT splitByChar('/', 'abcd')")
+
+    // Multiple characters
+    toSQL(select(dsl.splitByChar("ab", "abcd"))) should matchSQL("SELECT splitByString('ab', 'abcd')")
+    toSQL(select(dsl.splitByChar("a'", "abcd"))) should matchSQL("SELECT splitByString('a\\'', 'abcd')")
   }
 }