Add raw strings to BCQL.

If you want to match a literal string, not a regular expression,
you had to escape various special characters using a backslash.
Now, you can use a raw string to achieve the same:

  [lemma=r'<']

This frees clients who want to match literally from having to know
the exact escaping rules for Lucene's regex engine.

query-parser/src/main/java/nl/inl/blacklab/queryParser/corpusql/CorpusQueryLanguageParser.java

@@ -66,14 +66,23 @@ String chopEnds(String input) {
     }
 
     String getStringBetweenQuotes(String input) throws SingleQuotesException {
+        boolean isRaw = input.charAt(0) == 'r';
+        if (isRaw)
+            input = input.substring(1);
+
         String quoteUsed = input.substring(0, 1);
         input = chopEnds(input); // eliminate quotes
         if (!allowSingleQuotes && quoteUsed.equals("\'"))
             throw new SingleQuotesException();
 
         // Unescape ONLY the quotes found around this string
         // Leave other escaped characters as-is for Lucene's regex engine
-        return StringUtil.unescapeQuote(input, quoteUsed);
+        String quotedUnescaped = StringUtil.unescapeQuote(input, quoteUsed);
+        if (isRaw) {
+            // We want to find this string as-is; create a regex that will match this
+            return StringUtil.escapeLuceneRegexCharacters(quotedUnescaped);
+        }
+        return quotedUnescaped;
     }
 
     TextPatternTerm simplePattern(String str) {

query-parser/src/main/java/nl/inl/blacklab/queryParser/corpusql/GeneratedCorpusQueryLanguageParser.java

@@ -82,10 +82,10 @@ public class GeneratedCorpusQueryLanguageParser implements GeneratedCorpusQueryL
       }
     case DEFAULT_VALUE:
     case ROOT_DEP_OP:
-    case NAME:
-    case NUMBER:
     case QUOTED_STRING:
     case SINGLE_QUOTED_STRING:
+    case NAME:
+    case NUMBER:
     case 25:
     case 26:
     case 28:
@@ -652,9 +652,9 @@ public class GeneratedCorpusQueryLanguageParser implements GeneratedCorpusQueryL
     Integer[] rep = null;
     switch ((jj_ntk==-1)?jj_ntk_f():jj_ntk) {
     case DEFAULT_VALUE:
-    case NAME:
     case QUOTED_STRING:
     case SINGLE_QUOTED_STRING:
+    case NAME:
     case 25:
     case 26:
     case 42:{
@@ -724,10 +724,10 @@ public class GeneratedCorpusQueryLanguageParser implements GeneratedCorpusQueryL
     switch ((jj_ntk==-1)?jj_ntk_f():jj_ntk) {
     case DEFAULT_VALUE:
     case ROOT_DEP_OP:
-    case NAME:
-    case NUMBER:
     case QUOTED_STRING:
     case SINGLE_QUOTED_STRING:
+    case NAME:
+    case NUMBER:
     case 25:
     case 26:
     case 28:
@@ -1103,9 +1103,9 @@ private boolean jj_3R_40()
  {
     Token xsp;
     xsp = jj_scanpos;
-    if (jj_scan_token(17)) {
+    if (jj_scan_token(13)) {
     jj_scanpos = xsp;
-    if (jj_scan_token(18)) return true;
+    if (jj_scan_token(14)) return true;
     }
     return false;
   }
@@ -1191,9 +1191,9 @@ private boolean jj_3R_4()
  {
     Token xsp;
     xsp = jj_scanpos;
-    if (jj_scan_token(13)) {
+    if (jj_scan_token(15)) {
     jj_scanpos = xsp;
-    if (jj_scan_token(15)) return true;
+    if (jj_scan_token(17)) return true;
     }
     return false;
   }
@@ -1600,7 +1600,7 @@ private boolean jj_3R_26()
     if (jj_scan_token(38)) return true;
     Token xsp;
     xsp = jj_scanpos;
-    if (jj_scan_token(15)) jj_scanpos = xsp;
+    if (jj_scan_token(17)) jj_scanpos = xsp;
     if (jj_scan_token(37)) return true;
     return false;
   }
@@ -1706,7 +1706,7 @@ private boolean jj_3R_46()
       jj_la1_init_1();
    }
    private static void jj_la1_init_0() {
-      jj_la1_0 = new int[] {0x1607a600,0x80000,0x0,0x3f00000,0x3f00000,0x20000000,0x14060000,0x180,0x180,0xa000,0x1800,0xa000,0x0,0x2000,0x0,0x60000,0x8000,0x0,0x0,0x0,0x0,0xa000,0x6062200,0x16062200,0x1606a600,0x0,0xa000,0x4000,0x14002000,0x60200,0x0,0x14000000,0x4000,0x4000,0x2000,0x0,};
+      jj_la1_0 = new int[] {0x1606e600,0x80000,0x0,0x3f00000,0x3f00000,0x20000000,0x14006000,0x180,0x180,0x28000,0x1800,0x28000,0x0,0x8000,0x0,0x6000,0x20000,0x0,0x0,0x0,0x0,0x28000,0x600e200,0x1600e200,0x1602e600,0x0,0x28000,0x10000,0x14008000,0x6200,0x0,0x14000000,0x10000,0x10000,0x8000,0x0,};
    }
    private static void jj_la1_init_1() {
       jj_la1_1 = new int[] {0x400,0x0,0x380,0x0,0x0,0x0,0x0,0x0,0x0,0x0,0x0,0x0,0x1,0x0,0x1,0x0,0x0,0xe,0x10,0x380,0x380,0x0,0x400,0x400,0x400,0x40,0x0,0x0,0x0,0x400,0x380,0x0,0x0,0x0,0x0,0x1,};

query-parser/src/main/java/nl/inl/blacklab/queryParser/corpusql/GeneratedCorpusQueryLanguageParserConstants.java

@@ -27,17 +27,17 @@ public interface GeneratedCorpusQueryLanguageParserConstants {
   /** RegularExpression Id. */
   int ALIGNMENT_OP = 12;
   /** RegularExpression Id. */
-  int NAME = 13;
+  int QUOTED_STRING = 13;
   /** RegularExpression Id. */
-  int FLAGS = 14;
+  int SINGLE_QUOTED_STRING = 14;
   /** RegularExpression Id. */
-  int NUMBER = 15;
+  int NAME = 15;
   /** RegularExpression Id. */
-  int SETTINGS_OP = 16;
+  int FLAGS = 16;
   /** RegularExpression Id. */
-  int QUOTED_STRING = 17;
+  int NUMBER = 17;
   /** RegularExpression Id. */
-  int SINGLE_QUOTED_STRING = 18;
+  int SETTINGS_OP = 18;
 
   /** Lexical state. */
   int DEFAULT = 0;
@@ -57,12 +57,12 @@ public interface GeneratedCorpusQueryLanguageParserConstants {
     "<ROOT_DEP_OP>",
     "<DEP_OP>",
     "<ALIGNMENT_OP>",
+    "<QUOTED_STRING>",
+    "<SINGLE_QUOTED_STRING>",
     "<NAME>",
     "<FLAGS>",
     "<NUMBER>",
     "<SETTINGS_OP>",
-    "<QUOTED_STRING>",
-    "<SINGLE_QUOTED_STRING>",
     "\"::\"",
     "\"=\"",
     "\"!=\"",