Filters/FilterChm.cs

   1 //
   2 // FilterChm.cs : Trivial implementation of a CHM filter.
   3 //
   4 // Author :
   5 //      Miguel Cabrera <mfcabrer@unalmed.edu.co>
   6 //
   7 // Copyright (C) 2005 Miguel Cabrera
   8 //
   9 //
  10 // Permission is hereby granted, free of charge, to any person obtaining a
  11 // copy of this software and associated documentation files (the "Software"),
  12 // to deal in the Software without restriction, including without limitation
  13 // the rights to use, copy, modify, merge, publish, distribute, sublicense,
  14 // and/or sell copies of the Software, and to permit persons to whom the
  15 // Software is furnished to do so, subject to the following conditions:
  16 //
  17 // The above copyright notice and this permission notice shall be included in
  18 // all copies or substantial portions of the Software.
  19 //
  20 // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  21 // IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  22 // FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  23 // AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  24 // LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  25 // FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
  26 // DEALINGS IN THE SOFTWARE.
  27 //
  28
  29 using System;
  30 using System.Collections;
  31 using System.IO;
  32 using System.Text;
  33 using HtmlAgilityPack;
  34
  35 using Beagle.Util;
  36 using Beagle.Daemon;
  37
  38 namespace Beagle.Filters {
  39
  40         public class FilterChm : FilterHtml {
  41
  42                 ChmFile chmFile;
  43
  44                 public FilterChm ()
  45                 {
  46                         RegisterSupportedTypes();
  47                         SnippetMode= true;
  48
  49                 }
  50
  51
  52                 new protected  void WalkHeadNodes (HtmlNode node)
  53                 {
  54                         foreach (HtmlNode subnode in node.ChildNodes) {
  55                                 if (subnode.NodeType == HtmlNodeType.Element
  56                                     && subnode.Name == "title") {
  57                                         String title = WalkChildNodesForText (subnode);
  58                                         title = HtmlEntity.DeEntitize (title);
  59                                         //AddProperty (Beagle.Property.New ("dc:title", title));
  60                                         AppendText (title);
  61                                 }
  62                                 if (subnode.NodeType == HtmlNodeType.Element
  63                                     && subnode.Name == "meta") {
  64                                         string name = subnode.GetAttributeValue ("name", "");
  65                                         string content = subnode.GetAttributeValue ("content", "");
  66                                         if (name != "" && content != "")
  67                                                 AddProperty (Beagle.Property.New (name, content));
  68                                 }
  69                         }
  70                 }
  71
  72
  73                 public void WalkTocFile(HtmlNode node)
  74                 {
  75
  76
  77
  78                         foreach (HtmlNode subnode in node.ChildNodes) {
  79                                 if (subnode.NodeType == HtmlNodeType.Element) {
  80                                         switch (subnode.Name) {
  81                                         case "html":
  82                                 case "head":
  83                                         WalkTocFile (subnode);
  84                                         break;
  85                                 case "body":
  86                                         default:
  87                                                 WalkToc (subnode);
  88                                                 break;
  89                                         }
  90                                 }
  91                         }
  92
  93                 }
  94
  95
  96
  97                 public void WalkToc(HtmlNode node)
  98                 {
  99
 100                         switch (node.NodeType) {
 101
 102                         case HtmlNodeType.Document:
 103                         case HtmlNodeType.Element:
 104
 105                                 if(node.Name == "li")
 106                                         foreach(HtmlNode subnode in node.ChildNodes)
 107                                                 HandleTocEntry(subnode);
 108
 109                                 foreach(HtmlNode subnode in node.ChildNodes)
 110                                         WalkToc(subnode);
 111                                 break;
 112
 113
 114
 115                         }
 116
 117                 }
 118
 119
 120                 public void HandleTocEntry(HtmlNode node)
 121                 {
 122
 123                         if(node.Name == "object") {
 124
 125                         string attr = node.GetAttributeValue ("type", "");
 126
 127                         if(String.Compare(attr,"text/sitemap",true) == 0)
 128                                 foreach(HtmlNode subnode in node.ChildNodes)
 129                                         if(String.Compare(subnode.Name,"param",true) == 0 &&
 130                                            subnode.GetAttributeValue("name","") == "Name" ){
 131                                                 HotUp();
 132                                                 AppendText(subnode.GetAttributeValue("value",""));
 133                                                 HotDown();
 134
 135                                         }
 136
 137
 138
 139                         }
 140
 141                 }
 142
 143
 144                 void ReadHtml(TextReader reader)
 145                 {
 146
 147                         HtmlDocument doc = new HtmlDocument ();
 148
 149                         try {
 150                                 doc.Load (reader);
 151                         } catch (ArgumentNullException e) {
 152                                 /*Weird should not happend*/
 153                                 //¿What should do here?
 154                                 Logger.Log.Warn (e.Message);
 155                                 return;
 156
 157                         }
 158
 159                         if (doc != null)
 160                                 WalkNodes (doc.DocumentNode);
 161
 162
 163
 164                 }
 165
 166                 override protected void DoOpen (FileInfo info)
 167                 {
 168
 169                         chmFile = new ChmFile();
 170
 171                         try {
 172
 173                                 chmFile.Load(info.FullName);
 174
 175                         }
 176                         catch (Exception e) {
 177
 178                                 Logger.Log.Warn ("Could not parse {0}: {1}",info.Name,e.Message);
 179                                 Finished ();
 180                                 return;
 181
 182                         }
 183
 184
 185
 186
 187                 }
 188
 189
 190
 191                 override protected void DoPullProperties()
 192                 {
 193
 194                         if(chmFile.Title != "")
 195                                 AddProperty (Beagle.Property.New ("dc:title", chmFile.Title));
 196
 197
 198
 199                 }
 200
 201                 override protected void DoPull()
 202                 {
 203                         //Logger.Log.Debug("FilterCHM: Parsing:" + chmFile.Title);
 204                         //chmFile.ParseContents(ReadHtml);
 205
 206
 207                         /*
 208                           We only read the default file and the topic file
 209                         **/
 210                         ReadHtml(chmFile.GetDefaultFile());
 211
 212                         HtmlDocument doc = new HtmlDocument();
 213
 214                         doc.Load(chmFile.GetTopicsFile());
 215
 216                         WalkTocFile(doc.DocumentNode);
 217
 218                         Finished();
 219
 220
 221                 }
 222
 223                 override protected void  DoClose()
 224                 {
 225                         chmFile.Dispose();
 226
 227                 }
 228
 229                 override protected  void  RegisterSupportedTypes()
 230                 {
 231                         AddSupportedFlavor (FilterFlavor.NewFromMimeType ("application/x-chm"));
 232                 }
 233
 234         }
 235 }