remove invalid XML characters while building document

This commit is contained in:
2014-02-12 23:34:43 +01:00
parent c0b4635865
commit 4f01f25a72
2 changed files with 33 additions and 1 deletions
+24 -1
View File
@@ -22,12 +22,25 @@ using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
using System.Xml.Linq;
namespace ReadSharp.Ports.NReadability
{
public static class DomExtensions
{
// filters control characters but allows only properly-formed surrogate sequences
private static Regex _invalidXMLChars = new Regex(@"(?<![\uD800-\uDBFF])[\uDC00-\uDFFF]|[\uD800-\uDBFF](?![\uDC00-\uDFFF])|[\x00-\x08\x0B\x0C\x0E-\x1F\x7F-\x9F\uFEFF\uFFFE\uFFFF]", RegexOptions.CultureInvariant);
/// <summary>
/// removes any unusual unicode characters that can't be encoded into XML
/// </summary>
public static string RemoveInvalidXMLChars(string text)
{
if (String.IsNullOrEmpty(text)) return "";
return _invalidXMLChars.Replace(text, "");
}
#region XDocument extensions
public static XElement GetBody(this XDocument document)
@@ -229,7 +242,17 @@ namespace ReadSharp.Ports.NReadability
foreach (var childNode in container.Nodes())
{
resultSb.Append(childNode.ToString(SaveOptions.DisableFormatting));
try
{
resultSb.Append(childNode.ToString(SaveOptions.DisableFormatting));
}
catch (ArgumentException)
{
if (childNode is XElement)
{
resultSb.Append(RemoveInvalidXMLChars((childNode as XElement).Value));
}
}
}
return resultSb.ToString();
+9
View File
@@ -184,5 +184,14 @@ namespace ReadSharp.Tests
result = await reader.Read(new Uri("http://arstechnica.com/apple/2014/01/two-steps-forward-a-review-of-the-2013-mac-pro/"));
Assert.NotNull(result.NextPage);
}
[Fact]
public async Task DebugArticle()
{
string uri = "http://habrahabr.ru/post/211905/";
Article result = await reader.Read(new Uri(uri));
Assert.NotEmpty(result.Content);
}
}
}