Function get

tools/tokenizer/ntokenize.c:263–299

Phase 1. Forms logical lines by resolving escaped newlines (line continuations). */

   Forms logical lines by resolving escaped newlines (line continuations).
*/
int get(void)
{
  int cc;
 restart:
  // Read a fresh char:
  cc = normalize_newline(); // cc != '\r'
  if (cc == EOF) return EOF;
  char_count++;
  if (utf8_start(cc)) utf8_count++;

  if (cc == '\n') { // a normalized end-of-line (\r|\r?\n)
    linenr++;
    column = 0;
    return cc; // \n here signals a logical end-of-line
  }

  // Deal with \ line continuations!
  if (cc == '\\') {
    // Must look ahead (never maintained across get calls!):
    int nc = normalize_newline();
    if (nc == '\n') { // 1 logical line: discard \\n combo:
      char_count++; // counts the newline
      utf8_count++;
      linenr++;     // on next physical line
      column = 0;

      // Still need to get a character.
      // Could again start a line continuation!
      goto restart;
    }
    // Mind nc not \n. ungetc(EOF) is Okay.
    ungetc(nc, stdin);
    // cc == '\\' a regular backslash
  }
  column++;
  return cc;
}

/* Phase 2.
   Removes line and block comments and condenses white-space up to a newline.