Program to count identifiers.

Makefile

CFLAGS += $(shell pkg-config --cflags glib-2.0)
LDFLAGS += $(shell pkg-config --libs glib-2.0)

all: task

task.c

#include <stdio.h>
#include <stdbool.h>
#include <string.h>
#include <glib.h>
#include <assert.h>

// read a single line from input.
GString *get_line() {
  // read line
  char *line = NULL;
  size_t line_length;
  getline(&line, &line_length, stdin);

  // truncate newline
  g_strstrip(line);

  return g_string_new(line);
}

// check if the next word on the input is a valid identifier.
bool is_identifier(const char *word) {
  // compile the regular expression
  const GRegex *regex = g_regex_new("^[a-zA-Z_][a-zA-Z0-9_]*$", G_REGEX_OPTIMIZE, 0, NULL);
  return g_regex_match(regex, word, 0, NULL);
}

// counts the unique identifiers in a list of words.
int count_identifiers(gchar **words) {
  // find identifiers. by using a hash table, we know that repeated identifiers
  // are only counted once.
  GHashTable *identifiers = g_hash_table_new(g_str_hash, g_str_equal);
  for(size_t pos = 0; words[pos]; ++pos) {
    if(is_identifier(words[pos])) {
      g_hash_table_insert(identifiers, words[pos], NULL);
      printf("identifier: '%s'\n", words[pos]);
    }
  }

  // find out how many unique identifiers there are
  int count = g_hash_table_size(identifiers);

  g_hash_table_destroy(identifiers);
  return count;
}

// read a single line from the input, parse it, and return the number of
// identifiers found on it.
int parse_line() {
  // read a single line from input and split into words.
  GString *line = get_line();
  gchar **words = g_strsplit(line->str, " ", 0);

  int count = count_identifiers(words);

  free(g_string_free(line, false));
  g_strfreev(words);
  return count;
}

// tests to make sure everything works as intended
void test() {
  assert(is_identifier("Identifier9_"));
  assert(is_identifier("ALLCAPS"));
  assert(is_identifier("allsmall"));
  assert(is_identifier("________"));
  assert(is_identifier("i1431"));
  assert(!is_identifier("------"));
  assert(!is_identifier("9gag"));
  
  // you can add more tests here to make sure the functions work.
}

int main(int argc, char *argv[]) {
  test();

  // parse data until it's empty.
  while(!feof(stdin)) {
    printf("found %i identifiers.\n", parse_line());
  }

  return 0;
}