Better format for html-to-text

SubZero0 · Jan 29, 2019 · cf178aa · cf178aa
1 parent 18d7488
commit cf178aa
Show file tree

Hide file tree

Showing 2 changed files with 34 additions and 4 deletions.
diff --git a/src/DiscordNet/Query/Extensions/BaseDisplay.cs b/src/DiscordNet/Query/Extensions/BaseDisplay.cs
@@ -26,7 +26,7 @@ private async Task<DocsHttpResult> GetWebDocsAsync(string url, object o)
                 anchor = anchor.Substring(0, anchor.IndexOf('"'));
                 summary = block.Substring(block.IndexOf("summary\">") + 9);
                 summary = summary.Substring(0, summary.IndexOf("</div>"));
-                summary = WebUtility.HtmlDecode(StripTags(summary));
+                summary = Utils.ResolveHtml(summary);
                 /*string example = block.Substring(block.IndexOf("example\">")); //TODO: Find this
                 summary = summary.Substring(0, summary.IndexOf("</div>"));*/
                 if (!(o is TypeInfoWrapper) && !IsInherited(o))

diff --git a/src/DiscordNet/Utils.cs b/src/DiscordNet/Utils.cs
@@ -1,16 +1,46 @@
 using System;
 using System.Collections.Generic;
 using System.Linq;
+using System.Net;
 
 namespace DiscordNet
 {
     public static class Utils
     {
         public static string ResolveHtml(string html)
         {
-            //TODO: Pass summary, replace \n's with spaces, replace </p>'s with \n's, strip tags, decode html
-            //ISSUE: <p></p> could be empty, check for \n on first char
-            return "";
+            string res = WebUtility.HtmlDecode(StripTags(html.Replace('\n', ' ').Replace("</p>", "\n")));
+            if (res[0] == '\n')
+                res = res.Substring(1);
+            return res;
+        }
+
+        public static string StripTags(string source)
+        {
+            char[] array = new char[source.Length];
+            int arrayIndex = 0;
+            bool inside = false;
+
+            for (int i = 0; i < source.Length; i++)
+            {
+                char let = source[i];
+                if (let == '<')
+                {
+                    inside = true;
+                    continue;
+                }
+                if (let == '>')
+                {
+                    inside = false;
+                    continue;
+                }
+                if (!inside)
+                {
+                    array[arrayIndex] = let;
+                    arrayIndex++;
+                }
+            }
+            return new string(array, 0, arrayIndex);
         }
 
         public static IEnumerable<T> RandomShuffle<T>(this IEnumerable<T> source)