/* See LICENSE file for copyright and license details. */ #include <stdio.h>#include <stdlib.h>#include <string.h> #include "detect.h"#include "util.h" /* CMS signature strings found in HTML */static const char *wp_sigs[] = {	"wp-content/",	"wp-includes/",	"wp-json/",	"/xmlrpc.php",	"name=\"generator\" content=\"WordPress",	"powered by WordPress",	NULL}; static const char *blogger_sigs[] = {	"blogger.com",	"blogspot.com",	"content=\"blogger\"",	"name=\"generator\" content=\"Blogger",	"b:skin",	"b:template",	NULL}; static const char *hugo_sigs[] = {	"name=\"generator\" content=\"Hugo",	"powered by Hugo",	"Hugo --",	NULL}; static const char *jekyll_sigs[] = {	"name=\"generator\" content=\"Jekyll",	"powered by Jekyll",	"jekyll-",	NULL}; static const char *ghost_sigs[] = {	"content=\"Ghost",	"ghost-",	"ghost/api/",	"class=\"gh-",	NULL}; static const char *drupal_sigs[] = {	"Drupal.settings",	"name=\"generator\" content=\"Drupal",	"/sites/default/files/",	"/modules/",	NULL}; static const char *mediawiki_sigs[] = {	"name=\"generator\" content=\"MediaWiki",	"wgArticleId",	"mw-content-text",	"/wiki/",	NULL}; /* Check if HTML contains any signature from a list */static intmatch_sigs(const char *html, const char **sigs){	int i, hits; 	hits = 0;	for (i = 0; sigs[i]; i++) {		if (strcasestr(html, sigs[i]))			hits++;	}	return hits;} static const char *sitetype_name(SiteType type){	switch (type) {	case SITE_WORDPRESS: return "WordPress";	case SITE_BLOGGER:   return "Blogger";	case SITE_HUGO:      return "Hugo";	case SITE_JEKYLL:    return "Jekyll";	case SITE_GHOST:     return "Ghost";	case SITE_DRUPAL:    return "Drupal";	case SITE_MEDIAWIKI: return "MediaWiki";	default:             return "Unknown";	}} /* Extract feed URL from <link> tags */static char *find_feed_url(const char *html, const char *base_url){	const char *p, *href_start, *href_end;	char *tag, *href;	size_t tag_len, href_len;	char quote; 	p = html;	while ((p = strcasestr(p, "<link")) != NULL) {		const char *end = strchr(p, '>'); 		if (!end)			break; 		tag_len = end - p;		tag = xmalloc(tag_len + 1);		memcpy(tag, p, tag_len);		tag[tag_len] = '\0'; 		/* Check for RSS/Atom type */		if (strcasestr(tag, "application/rss+xml") ||		    strcasestr(tag, "application/atom+xml")) {			href_start = strcasestr(tag, "href=");			if (href_start) {				href_start += 5;				quote = 0;				if (*href_start == '"' ||				    *href_start == '\'')					quote = *href_start++; 				href_end = href_start;				if (quote) {					while (*href_end &&					       *href_end != quote)						href_end++;				} else {					while (*href_end &&					       *href_end != ' ' &&					       *href_end != '>')						href_end++;				} 				href_len = href_end - href_start;				href = xmalloc(href_len + 1);				memcpy(href, href_start, href_len);				href[href_len] = '\0'; 				free(tag); 				/* Resolve relative URL */				if (str_starts_with(href, "http")) {					return href;				} else {					char *resolved; 					resolved = url_resolve(					    base_url, href);					free(href);					return resolved;				}			}		} 		free(tag);		p = end + 1;	} 	return NULL;} SiteInfo *detect_site(const char *html, const char *url){	SiteInfo *info;	int wp, bl, hu, jk, gh, dr, mw;	int best;	char *domain; 	info = xmalloc(sizeof(SiteInfo));	info->type = SITE_UNKNOWN;	info->name = "Unknown";	info->feed_url = NULL;	info->api_url = NULL;	info->sitemap_url = NULL;	info->has_json_api = 0; 	/* Count signature matches for each CMS */	wp = match_sigs(html, wp_sigs);	bl = match_sigs(html, blogger_sigs);	hu = match_sigs(html, hugo_sigs);	jk = match_sigs(html, jekyll_sigs);	gh = match_sigs(html, ghost_sigs);	dr = match_sigs(html, drupal_sigs);	mw = match_sigs(html, mediawiki_sigs); 	/* Pick the CMS with the most signature hits */	best = 0; 	if (wp > best) { info->type = SITE_WORDPRESS; best = wp; }	if (bl > best) { info->type = SITE_BLOGGER; best = bl; }	if (hu > best) { info->type = SITE_HUGO; best = hu; }	if (jk > best) { info->type = SITE_JEKYLL; best = jk; }	if (gh > best) { info->type = SITE_GHOST; best = gh; }	if (dr > best) { info->type = SITE_DRUPAL; best = dr; }	if (mw > best) { info->type = SITE_MEDIAWIKI; best = mw; } 	/* Require at least 1 hit */	if (best < 1) {		info->type = SITE_UNKNOWN;		info->name = "Unknown";		return info;	} 	info->name = sitetype_name(info->type);	domain = url_get_domain(url); 	/* Set CMS-specific hints */	switch (info->type) {	case SITE_WORDPRESS:		info->has_json_api = 1;		info->api_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/wp-json/wp/v2/") + 1);		sprintf(info->api_url, "https://%s/wp-json/wp/v2/",		        domain);		info->sitemap_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/wp-sitemap.xml") + 1);		sprintf(info->sitemap_url,		        "https://%s/wp-sitemap.xml", domain);		break;	case SITE_BLOGGER:		info->has_json_api = 1;		/* Blogger Atom feed */		info->feed_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/feeds/posts/default") + 1);		sprintf(info->feed_url,		        "https://%s/feeds/posts/default", domain);		break;	case SITE_HUGO:		info->sitemap_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/sitemap.xml") + 1);		sprintf(info->sitemap_url,		        "https://%s/sitemap.xml", domain);		break;	case SITE_JEKYLL:		info->sitemap_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/sitemap.xml") + 1);		sprintf(info->sitemap_url,		        "https://%s/sitemap.xml", domain);		break;	case SITE_GHOST:		info->has_json_api = 1;		info->api_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/ghost/api/content/") + 1);		sprintf(info->api_url,		        "https://%s/ghost/api/content/", domain);		info->sitemap_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/sitemap.xml") + 1);		sprintf(info->sitemap_url,		        "https://%s/sitemap.xml", domain);		break;	case SITE_DRUPAL:		info->sitemap_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/sitemap.xml") + 1);		sprintf(info->sitemap_url,		        "https://%s/sitemap.xml", domain);		break;	case SITE_MEDIAWIKI:		info->has_json_api = 1;		info->api_url = xmalloc(		    strlen("https://") + strlen(domain) +		    strlen("/w/api.php") + 1);		sprintf(info->api_url,		        "https://%s/w/api.php", domain);		break;	default:		break;	} 	/* Try to find feed URL from HTML if not set */	if (!info->feed_url)		info->feed_url = find_feed_url(html, url); 	free(domain);	return info;} voidsiteinfo_free(SiteInfo *info){	if (!info)		return;	free(info->feed_url);	free(info->api_url);	free(info->sitemap_url);	free(info);} /* * Parse a simple sitemap.xml to extract <loc> URLs. * Returns array of URL strings, sets *count. * Caller frees the array and each string. */char **detect_sitemap_urls(SiteInfo *info, const char *domain, int *count){	char **urls;	int capacity, n; 	(void)info;	(void)domain; 	capacity = 64;	n = 0;	urls = xmalloc(capacity * sizeof(char *)); 	*count = n;	return urls;} /* * Get additional seed URLs based on CMS type. * For WordPress: /feed/, /wp-sitemap.xml * For Hugo/Jekyll: /sitemap.xml, /index.xml * For Blogger: /feeds/posts/default */char **detect_seed_urls(SiteInfo *info, const char *domain, int *count){	char **urls;	int n;	size_t len; 	n = 0;	urls = xmalloc(8 * sizeof(char *)); 	switch (info->type) {	case SITE_WORDPRESS:		len = strlen("https://") + strlen(domain) +		      strlen("/feed/") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/feed/", domain);		n++;		len = strlen("https://") + strlen(domain) +		      strlen("/wp-sitemap.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/wp-sitemap.xml", domain);		n++;		break;	case SITE_BLOGGER:		len = strlen("https://") + strlen(domain) +		      strlen("/feeds/posts/default") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/feeds/posts/default",		        domain);		n++;		len = strlen("https://") + strlen(domain) +		      strlen("/sitemap.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/sitemap.xml", domain);		n++;		break;	case SITE_HUGO:		/* fallthrough */	case SITE_JEKYLL:		len = strlen("https://") + strlen(domain) +		      strlen("/sitemap.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/sitemap.xml", domain);		n++;		len = strlen("https://") + strlen(domain) +		      strlen("/index.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/index.xml", domain);		n++;		break;	case SITE_GHOST:		len = strlen("https://") + strlen(domain) +		      strlen("/sitemap.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/sitemap.xml", domain);		n++;		break;	case SITE_DRUPAL:		len = strlen("https://") + strlen(domain) +		      strlen("/sitemap.xml") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/sitemap.xml", domain);		n++;		break;	case SITE_MEDIAWIKI:		len = strlen("https://") + strlen(domain) +		      strlen("/wiki/Special:AllPages") + 1;		urls[n] = xmalloc(len);		sprintf(urls[n], "https://%s/wiki/Special:AllPages",		        domain);		n++;		break;	default:		break;	} 	*count = n;	return urls;}