🐐 QVAERE

centvrion/ast_nodes.py

@@ -1277,6 +1277,19 @@ class BuiltIn(Node):
         with open(path, "a") as f:
           f.write(content)
         return vtable, ValNul()
+      case "QVAERE":
+        pattern = params[0]
+        text = params[1]
+        if not isinstance(pattern, ValStr) or not isinstance(text, ValStr):
+          raise CentvrionError("QVAERE requires two strings")
+        try:
+          matches = [
+            ValStr(m.group(0))
+            for m in re.finditer(pattern.value(), text.value())
+          ]
+        except re.error as e:
+          raise CentvrionError(f"Invalid regex: {e}")
+        return vtable, ValList(matches)
       case _:
         raise NotImplementedError(self.builtin)
 

centvrion/compiler/emit_expr.py

@@ -297,6 +297,9 @@ def _emit_builtin(node, ctx):
         lines.append(f"cent_adivnge({param_vars[0]}, {param_vars[1]});")
         lines.append(f"CentValue {tmp} = cent_null();")
 
+    case "QVAERE":
+      lines.append(f"CentValue {tmp} = cent_qvaere({param_vars[0]}, {param_vars[1]});")
+
     case _:
       raise NotImplementedError(node.builtin)
 

centvrion/compiler/runtime/cent_runtime.c

@@ -3,6 +3,7 @@
 #include <stdlib.h>
 #include <string.h>
 #include <time.h>
+#include <regex.h>
 
 /* ------------------------------------------------------------------ */
 /* Global arena                                                         */
@@ -870,6 +871,37 @@ CentValue cent_dict_keys(CentValue dict) {
   return result;
 }
 
+/* ------------------------------------------------------------------ */
+/* Regex                                                                */
+/* ------------------------------------------------------------------ */
+
+CentValue cent_qvaere(CentValue pattern, CentValue text) {
+  if (pattern.type != CENT_STR || text.type != CENT_STR)
+    cent_type_error("'QVAERE' requires two strings");
+  regex_t re;
+  int rc = regcomp(&re, pattern.sval, REG_EXTENDED);
+  if (rc != 0) {
+    char errbuf[256];
+    regerror(rc, &re, errbuf, sizeof(errbuf));
+    regfree(&re);
+    cent_runtime_error(errbuf);
+  }
+  CentValue result = cent_list_new(8);
+  const char *cursor = text.sval;
+  regmatch_t match;
+  while (*cursor && regexec(&re, cursor, 1, &match, 0) == 0) {
+    int len = match.rm_eo - match.rm_so;
+    char *buf = cent_arena_alloc(cent_arena, len + 1);
+    memcpy(buf, cursor + match.rm_so, len);
+    buf[len] = '\0';
+    cent_list_push(&result, cent_str(buf));
+    cursor += match.rm_eo;
+    if (len == 0) cursor++;  // avoid infinite loop on zero-length match
+  }
+  regfree(&re);
+  return result;
+}
+
 /* ------------------------------------------------------------------ */
 /* Initialisation                                                       */
 /* ------------------------------------------------------------------ */

centvrion/compiler/runtime/cent_runtime.h

@@ -232,6 +232,7 @@ CentValue cent_ordina(CentValue lst);                           /* ORDINA */
 CentValue cent_lege(CentValue path);                            /* LEGE */
 void      cent_scribe(CentValue path, CentValue content);       /* SCRIBE */
 void      cent_adivnge(CentValue path, CentValue content);      /* ADIVNGE */
+CentValue cent_qvaere(CentValue pattern, CentValue text);       /* QVAERE */
 
 /* ------------------------------------------------------------------ */
 /* Array helpers                                                        */

centvrion/lexer.py

@@ -57,7 +57,8 @@ builtin_tokens = [("BUILTIN", i) for i in [
   "TYPVS",
   "LEGE",
   "SCRIBE",
-  "ADIVNGE"
+  "ADIVNGE",
+  "QVAERE"
 ]]
 
 data_tokens = [