/* See LICENSE file for copyright and license details. */ #include <stdio.h>#include <stdlib.h>#include <string.h>#include <ctype.h> #include "parse.h"#include "util.h" ResourceList *reslist_new(void){    ResourceList *list = xmalloc(sizeof(ResourceList));    list->head = NULL;    list->tail = NULL;    list->count = 0;    return list;} voidreslist_free(ResourceList *list){    if (!list)        return;    Resource *r = list->head;    while (r) {        Resource *next = r->next;        free(r->url);        free(r);        r = next;    }    free(list);} intreslist_contains(ResourceList *list, const char *url){    for (Resource *r = list->head; r; r = r->next)        if (strcmp(r->url, url) == 0)            return 1;    return 0;} voidreslist_add(ResourceList *list, const char *url, ResourceType type){    if (!url || !*url || reslist_contains(list, url))        return;     /* Skip data: URLs */    if (str_starts_with(url, "data:"))        return;     Resource *r = xmalloc(sizeof(Resource));    r->url = xstrdup(url);    r->type = type;    r->next = NULL;     if (list->tail) {        list->tail->next = r;        list->tail = r;    } else {        list->head = r;        list->tail = r;    }    list->count++;} /* Extract attribute value from tag */static char *get_attr(const char *tag, const char *attr){    size_t attr_len = strlen(attr);    const char *p = tag;     while (*p) {        /* Skip whitespace */        while (*p && isspace((unsigned char)*p))            p++;         /* Check for attribute name */        if (strncasecmp(p, attr, attr_len) == 0) {            p += attr_len;            while (*p && isspace((unsigned char)*p))                p++;            if (*p == '=') {                p++;                while (*p && isspace((unsigned char)*p))                    p++;                 char quote = 0;                if (*p == '"' || *p == '\'') {                    quote = *p++;                }                 const char *start = p;                if (quote) {                    while (*p && *p != quote)                        p++;                } else {                    while (*p && !isspace((unsigned char)*p) && *p != '>')                        p++;                }                 size_t len = p - start;                char *value = xmalloc(len + 1);                memcpy(value, start, len);                value[len] = '\0';                return value;            }        }         /* Skip to next attribute */        while (*p && !isspace((unsigned char)*p) && *p != '>')            p++;    }     return NULL;} /* Determine resource type from URL/tag */static ResourceTypeguess_resource_type(const char *url, const char *tag_name){    if (!tag_name)        return RES_OTHER;     if (strcasecmp(tag_name, "img") == 0)        return RES_IMAGE;     if (strcasecmp(tag_name, "link") == 0) {        if (strstr(url, ".css") || strstr(url, "stylesheet"))            return RES_CSS;        if (strstr(url, ".woff") || strstr(url, ".ttf") || strstr(url, ".otf"))            return RES_FONT;        return RES_OTHER;    }     if (strcasecmp(tag_name, "a") == 0) {        /* <a> links: page by default, but assets (pdf, binaries, extra images) get RES_ASSET */        char *lower = xstrdup(url);        str_tolower(lower);        int is_asset = strstr(lower, ".pdf") || strstr(lower, ".ps") ||                       strstr(lower, ".zip") || strstr(lower, ".tar") ||                       strstr(lower, ".gz") || strstr(lower, ".doc") ||                       strstr(lower, ".rtf") || strstr(lower, ".txt") ||                       strstr(lower, ".csv") || strstr(lower, ".xls");        if (!is_asset) {            /* also treat linked images via <a> as assets so they get saved locally */            if (strstr(lower, ".jpg") || strstr(lower, ".jpeg") ||                strstr(lower, ".png") || strstr(lower, ".gif") ||                strstr(lower, ".webp") || strstr(lower, ".svg") ||                strstr(lower, ".ico"))                is_asset = 1;        }        free(lower);        return is_asset ? RES_ASSET : RES_PAGE;    }     if (strcasecmp(tag_name, "script") == 0)        return RES_OTHER;     /* Check by extension (for img/link etc that weren't caught above) */    char *lower = xstrdup(url);    str_tolower(lower);     ResourceType type = RES_OTHER;    if (strstr(lower, ".jpg") || strstr(lower, ".jpeg") ||        strstr(lower, ".png") || strstr(lower, ".gif") ||        strstr(lower, ".webp") || strstr(lower, ".svg") ||        strstr(lower, ".ico"))        type = RES_IMAGE;    else if (strstr(lower, ".css"))        type = RES_CSS;    else if (strstr(lower, ".woff") || strstr(lower, ".woff2") ||             strstr(lower, ".ttf") || strstr(lower, ".otf") ||             strstr(lower, ".eot"))        type = RES_FONT;    else if (strstr(lower, ".pdf"))        type = RES_ASSET;     free(lower);    return type;} ResourceList *parse_html(const char *html, const char *base_url){    ResourceList *list = reslist_new();    const char *p = html;     while (*p) {        /* Find tag start */        if (*p != '<') {            p++;            continue;        }        p++;         /* Skip comments */        if (str_starts_with(p, "!--")) {            p = strstr(p, "-->");            if (p)                p += 3;            else                break;            continue;        }         /* Get tag name */        const char *tag_start = p;        while (*p && !isspace((unsigned char)*p) && *p != '>' && *p != '/')            p++;         size_t tag_len = p - tag_start;        if (tag_len == 0 || tag_len > 20)            continue;         char tag_name[21];        memcpy(tag_name, tag_start, tag_len);        tag_name[tag_len] = '\0';         /* Find tag end */        const char *tag_end = strchr(p, '>');        if (!tag_end)            break;         /* Extract tag content for attribute parsing */        size_t content_len = tag_end - tag_start;        char *tag_content = xmalloc(content_len + 1);        memcpy(tag_content, tag_start, content_len);        tag_content[content_len] = '\0';         /* Check for relevant attributes based on tag */        char *url = NULL;         if (strcasecmp(tag_name, "img") == 0) {            url = get_attr(tag_content, "src");            if (!url)                url = get_attr(tag_content, "data-src");        } else if (strcasecmp(tag_name, "link") == 0) {            url = get_attr(tag_content, "href");        } else if (strcasecmp(tag_name, "script") == 0) {            url = get_attr(tag_content, "src");        } else if (strcasecmp(tag_name, "a") == 0) {            url = get_attr(tag_content, "href");        } else if (strcasecmp(tag_name, "source") == 0) {            url = get_attr(tag_content, "srcset");            if (!url)                url = get_attr(tag_content, "src");        }         if (url && *url) {            char *resolved = url_resolve(base_url, url);            ResourceType type = guess_resource_type(resolved, tag_name);            reslist_add(list, resolved, type);            free(resolved);        }         free(url);        free(tag_content);        p = tag_end + 1;    }     return list;} char *parse_title(const char *html){    const char *start = strcasestr(html, "<title");    if (!start)        return xstrdup("Untitled");     start = strchr(start, '>');    if (!start)        return xstrdup("Untitled");    start++;     const char *end = strcasestr(start, "</title>");    if (!end)        return xstrdup("Untitled");     size_t len = end - start;    char *title = xmalloc(len + 1);    memcpy(title, start, len);    title[len] = '\0';     return str_trim(title);} /* Helper to find and replace in string, returns new allocated string */static char *str_replace_first(const char *str, const char *old, size_t old_len, const char *new, size_t new_len){    const char *pos = strstr(str, old);    if (!pos)        return xstrdup(str);     size_t before_len = pos - str;    size_t after_len = strlen(pos + old_len);    size_t result_len = before_len + new_len + after_len;     char *result = xmalloc(result_len + 1);    memcpy(result, str, before_len);    memcpy(result + before_len, new, new_len);    memcpy(result + before_len + new_len, pos + old_len, after_len + 1);     return result;} char *inline_resources(const char *html, const char *base_url,                 char *(*fetch_and_encode)(const char *url, const char *base_url)){    char *result = xstrdup(html);    size_t search_offset = 0;     /* Process img tags */    while (1) {        const char *p = strcasestr(result + search_offset, "<img");        if (!p)            break;         const char *tag_end = strchr(p, '>');        if (!tag_end)            break;         /* Calculate offset for this tag */        size_t tag_offset = p - result;         /* Find src attribute */        const char *src_start = strcasestr(p, "src=");        if (!src_start || src_start > tag_end) {            /* No src, skip this img */            search_offset = (tag_end - result) + 1;            continue;        }         src_start += 4;        char quote = 0;        if (*src_start == '"' || *src_start == '\'')            quote = *src_start++;         const char *src_end = src_start;        if (quote) {            while (*src_end && *src_end != quote)                src_end++;        } else {            while (*src_end && !isspace((unsigned char)*src_end) && *src_end != '>')                src_end++;        }         /* Extract URL */        size_t url_len = src_end - src_start;        char *url = xmalloc(url_len + 1);        memcpy(url, src_start, url_len);        url[url_len] = '\0';         /* Skip if already data URI */        if (str_starts_with(url, "data:")) {            free(url);            search_offset = (tag_end - result) + 1;            continue;        }         /* Fetch and encode */        char *data_uri = fetch_and_encode(url, base_url);        if (data_uri) {            /* Build old and new strings for replacement */            char *old_attr = xmalloc(url_len + 8);            snprintf(old_attr, url_len + 8, "src=%c%s%c",                     quote ? quote : '"', url, quote ? quote : '"');             size_t new_attr_len = 5 + strlen(data_uri) + 2;            char *new_attr = xmalloc(new_attr_len + 1);            snprintf(new_attr, new_attr_len + 1, "src=\"%s\"", data_uri);             char *new_result = str_replace_first(result, old_attr, strlen(old_attr),                                                  new_attr, strlen(new_attr));            /* Continue searching after the new data URI */            search_offset = tag_offset + strlen(new_attr);             free(result);            result = new_result;             free(old_attr);            free(new_attr);            free(data_uri);        } else {            /* Fetch failed, skip this img tag */            search_offset = (tag_end - result) + 1;        }         free(url);    }     return result;}