C

int cmp_str(const void *a, const void *b) { char *const *sa = a; char *const *sb = b; return strcmp(*sa, *sb); }

int main() {

const char *pattern = "<[:blank:]+?[/]?[A-Aa-z0-9]+[:blank:]+?";  // only digits
const char *pattern_post = "[A-Za-z0-9]+";
char buffer[MAX_INPUT];
regex_t regex, regex_post;
int result;
char error_msg[128];

// --- Get input from stdin---
{
    size_t len = fread(buffer, 1, sizeof(buffer) - 1, stdin);
    buffer[len] = '\0';  // null-terminate
}

// --- Compile regex ---
result = regcomp(&regex, pattern, REG_EXTENDED);
if (result != 0) {
    regerror(result, &regex, error_msg, sizeof(error_msg));
    printf("Regex compile error: %s\n", error_msg);
    return 1;
}

result = regcomp(&regex_post, pattern_post, REG_EXTENDED);
if (result != 0) {
    regerror(result, &regex, error_msg, sizeof(error_msg));
    printf("Regex compile error: %s\n", error_msg);
    return 1;
}

// --- Execute regex ---
// Loop until no more matches
const char *ptr = buffer;
regmatch_t match[1];  // store one match at a time
regmatch_t match_post[1];  // store one match at a time
char hit[MAX_OUTPUT];
char match_out[MAX_OUTPUT];
memset(match_out,0,sizeof(match_out));
memset(hit,0,sizeof(hit));

while (regexec(&regex, ptr, 1, match, 0) == 0) {

    int start = match[0].rm_so;
    int end   = match[0].rm_eo;
    int len   = end - start;

    memset(hit,0,sizeof(hit));
    memcpy(hit,ptr+start,len);

    if(regexec(&regex_post, hit, 1, match_post, 0) == 0){
        int start_post = match_post[0].rm_so;
        int end_post   = match_post[0].rm_eo;
        int len_post   = end - start;
        strncat(match_out,hit+start_post,len_post);
        strcat(match_out,";");
    }
    // Move the pointer forward to search next part
    ptr += end;
}

//sort
{
    char *tokens[100];  // store pointers to words
    int count = 0;

    // Split string by ';'
    char *token = strtok(match_out, ";");
    while (token != NULL) {
        tokens[count++] = token;
        token = strtok(NULL, ";");
    }

    // Sort words
    qsort(tokens, count, sizeof(char *), cmp_str);

    // Print back in sorted order, separated by ';'
    for (int i = 0; i < count; i++) {
        if (i == 0 || strcmp(tokens[i], tokens[i - 1]) != 0) {
            if(i!= 0){printf(";");}
            printf("%s", tokens[i]);
        }
    }
    printf("\n");
}

// --- Free compiled regex ---
regfree(&regex);
return 0;

}

For Python 3:

import re
tag_pattern = re.compile(r'<\s*(?!/|!)([A-Za-z][\w\-]*)',re.ASCII)

number_of_lines = int(input())
unique_tag_names = set()
for _ in range(number_of_lines):
    html_line = input()
    unique_tag_names.update(tag_pattern.findall(html_line))
    
print(";".join(sorted(unique_tag_names)))

import re T = int(input())

pattern = r'\<([A-Za-z0-9]{1,})|\<\s+([A-Za-z0-9]{1,})' tag_set = set()

while T>0: line = input() pattern_found = re.findall(pattern, line) for p in pattern_found: tag_set.add(p[0]) tag_set.add(p[1]) T-=1

print(";".join(sorted([tag for tag in tag_set if len(tag) > 0 ])))

Python:

import re; html = "\n".join([input() for _ in range(int(input()))]); print(";".join(sorted(set(re.findall(r"<\s?([a-zA-Z]+[0-9A-Za-z])[^>]>", html)))));