« back to all changes in this revision
Viewing changes to ParseURLs.cs
- browse files at revision 11
- compare with another revision
- download diff
- download tarball
- view history from revision 11
- Committer: thindil
- Date: 2009-10-07 09:37:05 UTC
- Revision ID: thindil2@gmail.com-20091007093705-kp67qhweaxd1prmw
added process rename
- files removed:
- 7Zip
- misc
- files modified:
- AssemblyInfo.cs
added
removed
ParseURLs.cs
1
//
2
// Copyright (C) 2009,2010,2011 Bartek thindil Jasicki
3
//
4
// This file is part of Grubng
5
//
6
// Grubng is free software: you can redistribute it and/or modify
7
// it under the terms of the GNU General Public License as published by
8
// the Free Software Foundation, either version 3 of the License, or
9
// (at your option) any later version.
10
//
11
// This program is distributed in the hope that it will be useful,
12
// but WITHOUT ANY WARRANTY; without even the implied warranty of
13
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14
// GNU General Public License for more details.
15
//
16
// You should have received a copy of the GNU General Public License
17
// along with this program. If not, see <http://www.gnu.org/licenses/>.
18
//
19
20
using System;
21
using System.Text;
22
using System.Text.RegularExpressions;
23
24
namespace urlsdb
25
{
26
27
/// <summary>
28
/// Class provide functions for manipulate URL's
29
/// </summary>
30
internal sealed class ParseURLs : IDisposable
31
{
32
/// <summary>
33
/// Used to generate SHA256 hash of URL
34
/// </summary>
35
System.Security.Cryptography.SHA1Managed SHhash;
36
/// <summary>
37
/// SHA256 hash of URL
38
/// </summary>
39
StringBuilder key;
40
/// <summary>
41
/// Used to generate SHA256 hash of URL
42
/// </summary>
43
byte[] HashValue;
44
/// <summary>
45
/// Used to generate SHA256 hash of URL
46
/// </summary>
47
byte[] PureHash;
48
/// <summary>
49
/// If true, all disposable fields are disposed.
50
/// </summary>
51
bool disposed;
52
53
/// <summary>
54
/// Standard class constructor.
55
/// </summary>
56
public ParseURLs()
57
{
58
this.SHhash = new System.Security.Cryptography.SHA1Managed();
59
this.key = new StringBuilder();
60
}
61
62
/// <summary>
63
/// Function manipulate URL - check it correctness, drop to lower case host.
64
/// </summary>
65
/// <param name="url">
66
/// A <see cref="System.String"/> URL to parse.
67
/// </param>
68
/// <returns>
69
/// A <see cref="System.String"/> parsed URL or String.Empty if URL is invalid.
70
/// </returns>
71
public static string ParseURL(string url)
72
{
73
//Remove http(s):// from URL
74
int index = url.IndexOf("//");
75
if (index > 0)
76
{
77
url = url.Remove(0, (index + 2));
78
}
79
//Check correctness of URL
80
if (!Regex.IsMatch(url, @"^[a-z0-9]+([a-z0-9\=\-\.\?\,\'\/\\\+&%\$#_~]*)?$", RegexOptions.IgnoreCase))
81
{
82
return String.Empty;
83
}
84
//Conver to lower case host name
85
index = url.IndexOf('/');
86
string host, path;
87
if (index > 0)
88
{
89
host = url.Substring(0, index);
90
path = url.Substring(index);
91
}
92
else
93
{
94
host = url;
95
path = "/";
96
}
97
host = host.ToLower();
98
//Validate hostname
99
if (!Regex.IsMatch(host, @"^[a-z0-9]+([a-z0-9\-\.]*)?$", RegexOptions.IgnoreCase))
100
{
101
return String.Empty;
102
}
103
//Remove sessions id's from path
104
if (path.Length > 1)
105
{
106
path = Regex.Replace(path, @"(PHPSESSID|s|ses)=[a-z0-9]{32}(&|&)*", String.Empty, RegexOptions.IgnoreCase);
107
path = path.TrimEnd(new char[] {'&', '?'});
108
if ((path.LastIndexOf("&") > -1) && (path.LastIndexOf("&") == (path.Length - 5)))
109
{
110
path = path.Remove(path.LastIndexOf("&"));
111
}
112
}
113
url = host + path;
114
url = System.Web.HttpUtility.UrlEncode(host + path);
115
url = url.Replace("%2f", "/");
116
url = url.Replace("%2c", ",");
117
url = url.Replace("%3d", "=");
118
url = url.Replace("%3f", "?");
119
url = url.Replace("%26", "&");
120
url = url.Replace("%3b", ";");
121
url = url.Replace("%3a", ":");
122
url = url.Replace("%40", "@");
123
url = url.Replace("%23", "#");
124
url = url.Replace("%7e", "~");
125
url = url.Replace("%252f", "/");
126
return url;
127
}
128
129
/// <summary>
130
/// Function return SHA1 hash of URL
131
/// </summary>
132
/// <param name="url">
133
/// A <see cref="System.String"/> URL to hash
134
/// </param>
135
/// <returns>
136
/// A <see cref="System.String"/> SHA1 hash of URL
137
/// </returns>
138
public string GetHash(string url)
139
{
140
if (this.disposed)
141
{
142
throw new ObjectDisposedException(GetType().Name);
143
}
144
if (this.key.Length > 0)
145
{
146
this.key.Remove(0, key.Length);
147
}
148
this.PureHash = Encoding.UTF8.GetBytes(url);
149
this.HashValue = SHhash.ComputeHash(this.PureHash);
150
foreach(byte b in this.HashValue)
151
{
152
this.key.Append(String.Format("{0:x2}", b));
153
}
154
return this.key.ToString();
155
}
156
157
/// <summary>
158
/// Function dispose unmanaged resources.
159
/// </summary>
160
public void Dispose()
161
{
162
this.Dispose(true);
163
}
164
165
/// <summary>
166
/// Function dispose unmanaged resources. Private function.
167
/// </summary>
168
/// <param name="disposing">
169
/// A <see cref="System.Boolean"/> if true, dispose all unmanaged resources.
170
/// </param>
171
void Dispose(bool disposing)
172
{
173
if (!this.disposed)
174
{
175
this.SHhash = null;
176
this.HashValue = null;
177
this.PureHash = null;
178
this.disposed = true;
179
disposing = true;
180
}
181
}
182
}
183
}
Loggerhead is a web-based interface for Breezy
Version: 2.0.1