Initial Commit

tests/regex.c

@@ -0,0 +1,87 @@
+#define PCRE2_CODE_UNIT_WIDTH 8
+
+#include <pcre2.h>
+#include <stdio.h>
+#include <string.h>
+
+typedef struct Pattern {
+  const char *name;
+  pcre2_code *re;
+} Pattern;
+
+char *read_file(const char *filename) {
+  FILE *fp = fopen(filename, "rb");
+  if (!fp) {
+    perror("fopen");
+    return NULL;
+  }
+  fseek(fp, 0, SEEK_END);
+  long size = ftell(fp);
+  rewind(fp);
+  char *buffer = malloc(size + 1);
+  if (!buffer) {
+    perror("malloc");
+    fclose(fp);
+    return NULL;
+  }
+  size_t read = fread(buffer, 1, size, fp);
+  buffer[read] = '\0';
+  fclose(fp);
+  return buffer;
+}
+
+int main() {
+  const char *pattern = "\\b(class)\\s+(([a-zA-Z0-9_]+)((::[a-zA-Z0-9_]+)*))"
+                        "\\s*((<)\\s*(([a-zA-Z0-9_]+)((::[a-zA-Z0-9_]+)*))?)?";
+  const char *input = read_file("/home/syed/main/cubit/tests/test.rb");
+
+  int errorcode;
+  PCRE2_SIZE erroroffset;
+
+  // Compile the regex
+  pcre2_code *re = pcre2_compile((PCRE2_SPTR)pattern,   // pattern
+                                 PCRE2_ZERO_TERMINATED, // pattern length
+                                 0,                     // options
+                                 &errorcode,            // for error code
+                                 &erroroffset,          // for error offset
+                                 NULL                   // compile context
+  );
+
+  if (!re) {
+    PCRE2_UCHAR buffer[256];
+    pcre2_get_error_message(errorcode, buffer, sizeof(buffer));
+    fprintf(stderr, "PCRE2 compilation failed at offset %d: %s\n",
+            (int)erroroffset, buffer);
+    return 1;
+  }
+
+  // Prepare match data
+  pcre2_match_data *match_data = pcre2_match_data_create_from_pattern(re, NULL);
+
+  int rc = pcre2_match(re,                // the compiled pattern
+                       (PCRE2_SPTR)input, // subject string
+                       strlen(input),     // length of subject
+                       0,                 // start offset
+                       0,                 // options
+                       match_data,        // match data
+                       NULL               // match context
+  );
+
+  if (rc < 0) {
+    printf("No match found.\n");
+  } else {
+    PCRE2_SIZE *ovector = pcre2_get_ovector_pointer(match_data);
+    for (int i = 0; i < rc; i++) {
+      PCRE2_SIZE start = ovector[2 * i];
+      PCRE2_SIZE end = ovector[2 * i + 1];
+      printf("Group %d: start=%lu end=%lu text=\"%.*s\"\n", i, start, end,
+             (int)(end - start), input + start);
+    }
+  }
+
+  // Cleanup
+  pcre2_match_data_free(match_data);
+  pcre2_code_free(re);
+
+  return 0;
+}