From a1ca82847eb356c6b85ada2ac11f38d98f6e085e Mon Sep 17 00:00:00 2001
From: dec05eba <dec05eba@protonmail.com>
Date: Mon, 13 Jul 2020 15:57:10 +0200
Subject: Start on add_rss, add rss parser

---
 rss.c | 176 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 176 insertions(+)
 create mode 100644 rss.c

(limited to 'rss.c')

diff --git a/rss.c b/rss.c
new file mode 100644
index 0000000..fdb932c
--- /dev/null
+++ b/rss.c
@@ -0,0 +1,176 @@
+#include "rss.h"
+#include "download.h"
+#include "buffer.h"
+#include <string.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+static int is_alpha_lowercase(char c) {
+    return c >= 'a' && c <= 'z';
+}
+
+static int is_digit(char c) {
+    return c >= '0' && c <= '9';
+}
+
+static char* get_amp_end(char *str) {
+    for(;;) {
+        char c = *str;
+        if(is_alpha_lowercase(c) || is_digit(c) || c == '#')
+            ++str;
+        else if(c == ';' || c == '\0')
+            break;
+    }
+    return str;
+}
+
+static void xml_unescape(char *str, char *result, int result_length) {
+    int index = 0;
+    for(;;) {
+        char c = *str;
+        if(c == '&') {
+            char *amp_end = get_amp_end(str + 1);
+            char prev_char = *amp_end;
+            *amp_end = '\0';
+
+            if(str[1] == '#') {
+                result[index++] = atoi(str + 2);
+            } else {
+                if(strcmp(str + 1, "amp") == 0)
+                    result[index++] = '&';
+                else if(strcmp(str + 1, "lt") == 0)
+                    result[index++] = '<';
+                else if(strcmp(str + 1, "gt") == 0)
+                    result[index++] = '>';
+                else if(strcmp(str + 1, "apos") == 0)
+                    result[index++] = '\'';
+            }
+
+            *amp_end = prev_char;
+            str = amp_end;
+            if(prev_char != '\0')
+                ++str;
+        } else if(c == '\0') {
+            result[index] = '\0';
+            break;
+        } else {
+            result[index++] = c;
+            ++str;
+        }
+
+        if(index == result_length - 1) {
+            result[index] = '\0';
+            break;
+        }
+    }
+}
+
+static char* string_substr_before_tag_end(char *str, const char *tag) {
+    char *tag_p = strstr(str, tag);
+    if(tag_p)
+        *tag_p = '\0';
+    return tag_p;
+}
+
+typedef void (*RssParseCallback)(const char *title, const char *link, void *userdata);
+
+static int parse_rss(char *str, char *rss_title_str, int rss_title_str_size, RssParseCallback parse_callback, void *userdata) {
+    char *channel_start = strstr(str, "<channel>");
+    if(!channel_start)
+        return 1;
+
+    char *after_channel = channel_start + 9;
+
+    char *rss_title = strstr(after_channel, "<title>");
+    char *first_item = strstr(after_channel, "<item>");
+    if(!first_item) {
+        rss_title += 7;
+        string_substr_before_tag_end(rss_title, "</title>");
+        xml_unescape(rss_title, rss_title_str, rss_title_str_size);
+        return 0;
+    }
+
+    if(rss_title < first_item) {
+        rss_title += 7;
+        string_substr_before_tag_end(rss_title, "</title>");
+        xml_unescape(rss_title, rss_title_str, rss_title_str_size);
+    } else {
+        rss_title_str[0] = '\0';
+    }
+
+    char title_str[256];
+    char link_str[2084];
+
+    char *item = first_item;
+    for(;;) {
+        char *after_first_item = item + 6;
+        char *item_end = strstr(after_first_item, "</item>");
+        if(!item_end)
+            return 1;
+
+        char *item_title = strstr(after_first_item, "<title>");
+        if(!item_title)
+            return 1;
+
+        if(item_title >= item_end)
+            return 1;
+        
+        item_title += 7;
+        char *after_title = string_substr_before_tag_end(item_title, "</title>");
+        if(!after_title)
+            return 1;
+
+        after_title += 8;
+        char *item_link = strstr(after_title, "<link>");
+        if(!item_link)
+            return 1;
+
+        if(item_link >= item_end)
+            return 1;
+
+        item_link += 6;
+        string_substr_before_tag_end(item_link, "</link>");
+
+        xml_unescape(item_title, title_str, sizeof(title_str));
+        xml_unescape(item_link, link_str, sizeof(link_str));
+        parse_callback(title_str, link_str, userdata);
+
+        item = strstr(item_end + 7, "<item>");
+        if(!item)
+            return 0;
+    }
+}
+
+static void rss_parse_callback(const char *title, const char *link, void *userdata) {
+    (void)userdata;
+    fprintf(stderr, "title: |%s|, link: |%s|\n", title, link);
+}
+
+int add_rss(const char *name, const char *url, const char *rss_config_dir, const char *start_after) {
+    (void)name;
+    (void)rss_config_dir;
+    (void)start_after;
+    int result = 0;
+
+    Buffer buffer;
+    buffer_init(&buffer);
+    int res = download_to_buffer(url, &buffer);
+    if(res != 0) {
+        fprintf(stderr, "Failed to download rss: %s\n", url);
+        result = res;
+        goto cleanup;
+    }
+
+    char rss_title[256];
+    res = parse_rss(buffer.data, rss_title, sizeof(rss_title), rss_parse_callback, NULL);
+    if(res != 0) {
+        fprintf(stderr, "Failed to parse rss for url: %s\n", url);
+        result = res;
+        goto cleanup;
+    }
+    fprintf(stderr, "rss title: |%s|\n", rss_title);
+
+    cleanup:
+    buffer_deinit(&buffer);
+    return result;
+}
-- 
cgit v1.2.3