Filters/FilterDocbook.cs

   1 //
   2 // FilterDocbook.cs
   3 //
   4 // Copyright (C) 2005 Novell, Inc.
   5 //
   6
   7 //
   8 // Permission is hereby granted, free of charge, to any person obtaining a
   9 // copy of this software and associated documentation files (the "Software"),
  10 // to deal in the Software without restriction, including without limitation
  11 // the rights to use, copy, modify, merge, publish, distribute, sublicense,
  12 // and/or sell copies of the Software, and to permit persons to whom the
  13 // Software is furnished to do so, subject to the following conditions:
  14 //
  15 // The above copyright notice and this permission notice shall be included in
  16 // all copies or substantial portions of the Software.
  17 //
  18 // THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  19 // IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  20 // FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  21 // AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  22 // LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  23 // FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
  24 // DEALINGS IN THE SOFTWARE.
  25 //
  26
  27 using System;
  28 using System.IO;
  29 using System.Xml;
  30 using System.Text;
  31 using System.Collections;
  32
  33 using Beagle.Util;
  34 using Beagle.Daemon;
  35
  36 namespace Beagle.Filters
  37 {
  38         public class FilterDocbook : Filter
  39         {
  40                 protected XmlTextReader reader;
  41
  42                 protected string base_path;
  43                 protected string base_title;
  44
  45                 protected Stack entries_stack = new Stack ();
  46
  47                 protected class DocbookEntry {
  48                         public string Id = null;
  49                         public string Title = null;
  50                         public int Depth = -1;
  51                         public StringBuilder Content = new StringBuilder ();
  52                 }
  53
  54                 //////////////////////////////////////////////////
  55
  56                 public FilterDocbook ()
  57                 {
  58                         SnippetMode = false;
  59                         SetVersion (3);
  60
  61                         AddSupportedFlavor (FilterFlavor.NewFromMimeType ("application/docbook+xml"));
  62                         AddSupportedFlavor (FilterFlavor.NewFromExtension (".docbook"));
  63
  64                         // FIXME: Uri/Extension mapping?
  65                         AddSupportedFlavor (new FilterFlavor ("file:///usr/share/doc/*", ".xml", null, 0));
  66                 }
  67
  68                 ///////////////////////////////////////////////////
  69
  70                 override protected void DoOpen (FileInfo info)
  71                 {
  72                         base_path = info.FullName;
  73                         reader = new XmlTextReader (Stream);
  74                         reader.XmlResolver = null;
  75                 }
  76
  77                 override protected void DoPullProperties ()
  78                 {
  79                         Stopwatch watch = new Stopwatch ();
  80
  81                         watch.Start ();
  82
  83                         while (reader.Read ()) {
  84                                 switch (reader.NodeType) {
  85                                 case XmlNodeType.Element:
  86                                         if (reader.Name.StartsWith ("sect") || reader.Name.StartsWith ("chapter")) {
  87                                                 string id = reader.GetAttribute ("id");
  88
  89                                                 if (id != null && id != "") {
  90                                                         DocbookEntry entry = new DocbookEntry ();
  91                                                         entry.Id = id;
  92                                                         entry.Depth = reader.Depth;
  93                                                         entries_stack.Push (entry);
  94                                                 }
  95                                         } else if (reader.Name == "title") {
  96                                                 reader.Read (); // Go to the text node
  97
  98                                                 if (entries_stack.Count == 0 && base_title == null) {
  99                                                         // This is probably the book title
 100                                                         base_title = reader.Value;
 101                                                 } else if (entries_stack.Count > 0) {
 102                                                         DocbookEntry entry = (DocbookEntry) entries_stack.Peek ();
 103
 104                                                         if (entry.Title == null)
 105                                                                 entry.Title = reader.Value;
 106                                                 }
 107                                         }
 108                                         break;
 109
 110                                 case XmlNodeType.Text:
 111                                         // Append text to the child indexable
 112                                         if (entries_stack.Count > 0)
 113                                                 ((DocbookEntry) entries_stack.Peek ()).Content.Append (reader.Value);
 114
 115                                         // Append text to the main indexable
 116                                         AppendText (reader.Value);
 117                                         break;
 118
 119                                 case XmlNodeType.EndElement:
 120                                         if (entries_stack.Count > 0 && ((DocbookEntry) entries_stack.Peek ()).Depth == reader.Depth) {
 121                                                 DocbookEntry entry, parent_entry = null;
 122
 123                                                 entry = (DocbookEntry) entries_stack.Pop ();
 124
 125                                                 if (entries_stack.Count > 0)
 126                                                         parent_entry = (DocbookEntry) entries_stack.Peek ();
 127
 128                                                 Indexable indexable = new Indexable (UriFu.PathToFileUri (String.Format ("{0}#{1}", base_path, entry.Id)));
 129                                                 indexable.Type = "DocbookEntry";
 130                                                 indexable.MimeType = "text/plain";
 131                                                 indexable.AddProperty (Property.NewUnsearched ("fixme:id", entry.Id));
 132
 133                                                 if (entry.Title != null)
 134                                                         indexable.AddProperty (Property.New ("dc:title", entry.Title));
 135
 136                                                 // Add the docbook book title
 137                                                 if (base_title != null)
 138                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:base_title", base_title));
 139
 140                                                 // Add any parent (as in docbook parent entry, not beagle) data if we have it
 141                                                 if (parent_entry != null) {
 142                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:parent_id", parent_entry.Id));
 143
 144                                                         if (parent_entry.Title != null)
 145                                                                 indexable.AddProperty (Property.NewUnsearched ("fixme:parent_title", parent_entry.Title));
 146                                                 }
 147
 148
 149                                                 StringReader content_reader = new StringReader (entry.Content.ToString ());
 150                                                 indexable.SetTextReader (content_reader);
 151
 152                                                 AddChildIndexable (indexable);
 153                                         }
 154                                         break;
 155                                 }
 156                         }
 157
 158                         watch.Stop ();
 159
 160                         // If we've successfully crawled the file but haven't
 161                         // found any indexables, we shouldn't consider it
 162                         // successfull at all.
 163                         if (ChildIndexables.Count == 0) {
 164                                 Error ();
 165                                 return;
 166                         }
 167
 168                         Logger.Log.Debug ("Parsed docbook file in {0}", watch);
 169
 170                         Finished ();
 171                 }
 172         }
 173 }