zynk/cmd_yt.h

#ifndef CMD_YT_H
#define CMD_YT_H

#include "zynk.h"

/*
 * yt_hex_digit - Convert a hexadecimal character to its numeric value.
 *
 * Parameters:
 *   c - A character ('0'-'9', 'a'-'f', or 'A'-'F').
 *
 * Returns: The numeric value (0-15), or -1 if not a valid hex digit.
 */
static int yt_hex_digit(char c) {
  if (c >= '0' && c <= '9') return c - '0';
  if (c >= 'a' && c <= 'f') return c - 'a' + 10;
  if (c >= 'A' && c <= 'F') return c - 'A' + 10;
  return -1;
}

/*
 * yt_codepoint_to_utf8 - Encode a Unicode code point as UTF-8.
 *
 * Parameters:
 *   cp  - The Unicode code point (U+0000 to U+10FFFF).
 *   out - Output buffer for the UTF-8 bytes (at least 4 bytes needed).
 *
 * Returns: The number of UTF-8 bytes written (1-4).
 */
static size_t yt_codepoint_to_utf8(unsigned int cp, char *out) {
  if (cp < 0x80) { out[0] = (char)cp; return 1; }
  if (cp < 0x800) { out[0] = (char)(0xC0 | (cp >> 6)); out[1] = (char)(0x80 | (cp & 0x3F)); return 2; }
  if (cp < 0x10000) { out[0] = (char)(0xE0 | (cp >> 12)); out[1] = (char)(0x80 | ((cp >> 6) & 0x3F)); out[2] = (char)(0x80 | (cp & 0x3F)); return 3; }
  out[0] = (char)(0xF0 | (cp >> 18)); out[1] = (char)(0x80 | ((cp >> 12) & 0x3F));
  out[2] = (char)(0x80 | ((cp >> 6) & 0x3F)); out[3] = (char)(0x80 | (cp & 0x3F)); return 4;
}

/*
 * yt_decode_unicode_escapes - Decode \\uXXXX JSON Unicode escapes in-place.
 *
 * Scans a string for JSON Unicode escape sequences (e.g. \\u0027 for single
 * quote) and replaces each with the corresponding UTF-8 bytes. Safe to use
 * in-place because UTF-8 encoding of any code point is never longer than the
 * 6-byte \\uXXXX escape.
 *
 * Parameters:
 *   s - The string to decode (modified in-place, must be writable).
 *
 * Returns: void.
 */
static void yt_decode_unicode_escapes(char *s) {
  char *r = s, *w = s;
  while (*r) {
    if (r[0] == '\\' && r[1] == 'u' && yt_hex_digit(r[2]) >= 0 && yt_hex_digit(r[3]) >= 0 &&
        yt_hex_digit(r[4]) >= 0 && yt_hex_digit(r[5]) >= 0) {
      unsigned int cp = ((unsigned int)yt_hex_digit(r[2]) << 12) | ((unsigned int)yt_hex_digit(r[3]) << 8) |
                        ((unsigned int)yt_hex_digit(r[4]) << 4)  | (unsigned int)yt_hex_digit(r[5]);
      char utf8[5];
      size_t n = yt_codepoint_to_utf8(cp, utf8);
      memcpy(w, utf8, n);
      w += n;
      r += 6;
    } else {
      *w++ = *r++;
    }
  }
  *w = '\0';
}

/*
 * yt_url_encode - Percent-encode a string for safe use in a URL query parameter.
 *
 * Encodes all characters except unreserved characters (alphanumerics, '-', '_',
 * '.', '~') using RFC 3986 percent-encoding. Used to encode the YouTube video
 * URL before embedding it in the oEmbed API request.
 *
 * Parameters:
 *   src    - The input string to encode.
 *   dst    - Output buffer for the encoded string.
 *   dstsz  - Size of the output buffer.
 *
 * Returns: void (output is written to dst, NUL-terminated).
 */
static void yt_url_encode(const char *src, char *dst, size_t dstsz) {
  static const char hex[] = "0123456789ABCDEF";
  size_t di = 0;
  for (; *src && di + 4 < dstsz; src++) {
    unsigned char c = (unsigned char)*src;
    if (isalnum(c) || c == '-' || c == '_' || c == '.' || c == '~') {
      dst[di++] = c;
    } else {
      dst[di++] = '%';
      dst[di++] = hex[c >> 4];
      dst[di++] = hex[c & 0x0F];
    }
  }
  dst[di] = '\0';
}

/*
 * cmd_yt - Handle the "!yt <url>" command.
 *
 * Fetches the title and author of a YouTube video using the YouTube oEmbed
 * API (https://www.youtube.com/oembed). The video URL is percent-encoded
 * and passed as a query parameter. The JSON response is parsed for the
 * "title" and "author_name" fields, and the result is formatted as
 * "Title — Author" and sent to IRC. Rate-limited.
 *
 * Parameters:
 *   s            - The IRC session.
 *   msg          - The raw message text.
 *   reply_target - Channel or nick to reply to.
 *   src_nick     - The nick of the user who sent the command.
 *
 * Returns: 1 if matched, 0 otherwise.
 */
int cmd_yt(Session *s, const char *msg, const char *reply_target, const char *src_nick) {
  if (strncmp(msg, "!yt", 3) != 0) return 0;
  if (msg[3] != ' ' && msg[3] != '\0') return 0;
  if (rate_limit_check(src_nick, reply_target, current_userhost) < 0) return 1;
  const char *rest = msg[3] == ' ' ? msg + 4 : "";
  while (*rest == ' ') rest++;
  if (!*rest) {
    irc_reply(s, reply_target, src_nick, "Usage: !yt <youtube-url>");
    return 1;
  }
  char url_buf[512];
  int i = 0;
  while (rest[i] && rest[i] != ' ' && i < 500) { url_buf[i] = rest[i]; i++; }
  url_buf[i] = '\0';
  if (strncmp(url_buf, "http://", 7) != 0 && strncmp(url_buf, "https://", 8) != 0) {
    irc_reply(s, reply_target, src_nick, "Please provide a valid URL");
    return 1;
  }
  char encoded[1024];
  yt_url_encode(url_buf, encoded, sizeof encoded);
  char api_url[1536];
  snprintf(api_url, sizeof api_url, "https://www.youtube.com/oembed?url=%s&format=json", encoded);
  char *json = fetch_url(api_url);
  if (!json) {
    irc_reply(s, reply_target, src_nick, "YouTube: could not fetch video info");
    return 1;
  }
  const char *title_s = json_find(json, "\"title\":");
  char title[256];
  if (title_s && title_s[0] == '"') {
    title_s++;
    const char *tq = strchr(title_s, '"');
    if (tq) {
      size_t tlen = tq - title_s;
      if (tlen > 255) tlen = 255;
      memcpy(title, title_s, tlen);
      title[tlen] = '\0';
      yt_decode_unicode_escapes(title);
    } else {
      snprintf(title, sizeof title, "(parse error)");
    }
  } else {
    snprintf(title, sizeof title, "(no title found)");
  }
  const char *author_s = json_find(json, "\"author_name\":");
  char resp[512];
  if (author_s && author_s[0] == '"') {
    author_s++;
    const char *aq = strchr(author_s, '"');
    if (aq) {
      size_t alen = aq - author_s;
      if (alen > 127) alen = 127;
      char author[128];
      memcpy(author, author_s, alen);
      author[alen] = '\0';
      yt_decode_unicode_escapes(author);
      snprintf(resp, sizeof resp, "%s — %s", title, author);
    } else {
      snprintf(resp, sizeof resp, "%s", title);
    }
  } else {
    snprintf(resp, sizeof resp, "%s", title);
  }
  free(json);
  irc_reply(s, reply_target, src_nick, resp);
  log_stamp(); fprintf(stderr, CLR_MAGENTA "YT %s from %s:" CLR_RESET " %s\n", reply_target, src_nick, resp);
  return 1;
}

#endif