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                 protected string base_language;
  45
  46                 protected Stack entries_stack = new Stack ();
  47
  48                 protected class DocbookEntry {
  49                         public string Id = null;
  50                         public string Title = null;
  51                         public string Language = null;
  52                         public int Depth = -1;
  53                         public StringBuilder Content = new StringBuilder ();
  54                 }
  55
  56                 //////////////////////////////////////////////////
  57
  58                 public FilterDocbook ()
  59                 {
  60                         SnippetMode = false;
  61                         SetVersion (4);
  62
  63                         AddSupportedFlavor (FilterFlavor.NewFromMimeType ("application/docbook+xml"));
  64                         AddSupportedFlavor (FilterFlavor.NewFromExtension (".docbook"));
  65
  66                         // FIXME: Uri/Extension mapping?
  67                         AddSupportedFlavor (new FilterFlavor ("file:///usr/share/doc/*", ".xml", null, 0));
  68                 }
  69
  70                 ///////////////////////////////////////////////////
  71
  72                 override protected void DoOpen (FileInfo info)
  73                 {
  74                         base_path = info.FullName;
  75                         reader = new XmlTextReader (Stream);
  76                         reader.XmlResolver = null;
  77                 }
  78
  79                 override protected void DoPullProperties ()
  80                 {
  81                         Stopwatch watch = new Stopwatch ();
  82
  83                         watch.Start ();
  84
  85                         while (reader.Read ()) {
  86                                 switch (reader.NodeType) {
  87                                 case XmlNodeType.Element:
  88                                         if (reader.Name.StartsWith ("sect") || reader.Name.StartsWith ("chapter")) {
  89                                                 string id = reader.GetAttribute ("id");
  90
  91                                                 if (id != null && id != String.Empty) {
  92                                                         DocbookEntry entry = new DocbookEntry ();
  93                                                         entry.Id = id;
  94                                                         entry.Depth = reader.Depth;
  95
  96                                                         string language = reader.GetAttribute ("lang");
  97
  98                                                         if (language != null && language != String.Empty)
  99                                                                 entry.Language = language;
 100
 101                                                         entries_stack.Push (entry);
 102                                                 }
 103                                         } else if (reader.Name == "article" || reader.Name == "book") {
 104                                                 string language = reader.GetAttribute ("lang");
 105
 106                                                 if (language != null && language != String.Empty)
 107                                                         base_language = language;
 108                                         } else if (reader.Name == "title") {
 109                                                 reader.Read (); // Go to the text node
 110
 111                                                 if (entries_stack.Count == 0 && base_title == null) {
 112                                                         // This is probably the book title
 113                                                         base_title = reader.Value;
 114                                                 } else if (entries_stack.Count > 0) {
 115                                                         DocbookEntry entry = (DocbookEntry) entries_stack.Peek ();
 116
 117                                                         if (entry.Title == null)
 118                                                                 entry.Title = reader.Value;
 119                                                 }
 120                                         }
 121                                         break;
 122
 123                                 case XmlNodeType.Text:
 124                                         // Append text to the child indexable
 125                                         if (entries_stack.Count > 0)
 126                                                 ((DocbookEntry) entries_stack.Peek ()).Content.Append (reader.Value);
 127
 128                                         // Append text to the main indexable
 129                                         AppendText (reader.Value);
 130                                         break;
 131
 132                                 case XmlNodeType.EndElement:
 133                                         if (entries_stack.Count > 0 &&
 134                                             ((DocbookEntry) entries_stack.Peek ()).Depth == reader.Depth) {
 135                                                 DocbookEntry entry, parent_entry = null;
 136
 137                                                 entry = (DocbookEntry) entries_stack.Pop ();
 138
 139                                                 if (entries_stack.Count > 0)
 140                                                         parent_entry = (DocbookEntry) entries_stack.Peek ();
 141
 142                                                 Indexable indexable = new Indexable (UriFu.PathToFileUri (String.Format ("{0}#{1}", base_path, entry.Id)));
 143                                                 indexable.HitType = "DocbookEntry";
 144                                                 indexable.MimeType = "text/x-docbook-entry";
 145                                                 indexable.Filtering = IndexableFiltering.AlreadyFiltered;
 146
 147                                                 indexable.AddProperty (Property.NewUnsearched ("fixme:id", entry.Id));
 148                                                 indexable.AddProperty (Property.New ("dc:title", entry.Title));
 149
 150                                                 // Add the docbook book title
 151                                                 indexable.AddProperty (Property.NewUnsearched ("fixme:base_title", base_title));
 152
 153                                                 // Add the child language (or docbook language if none is specified)
 154                                                 if (entry.Language != null)
 155                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:language", entry.Language));
 156                                                 else if (base_language != null)
 157                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:language", base_language));
 158
 159                                                 // Add any parent (as in docbook parent entry, not beagle) data if we have it
 160                                                 if (parent_entry != null) {
 161                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:parent_id", parent_entry.Id));
 162                                                         indexable.AddProperty (Property.NewUnsearched ("fixme:parent_title", parent_entry.Title));
 163                                                 }
 164
 165
 166                                                 StringReader content_reader = new StringReader (entry.Content.ToString ());
 167                                                 indexable.SetTextReader (content_reader);
 168
 169                                                 AddChildIndexable (indexable);
 170                                         }
 171                                         break;
 172                                 }
 173                         }
 174
 175                         // Add the common properties to the top-level
 176                         // file item such as Title, Language etc.
 177
 178                         AddProperty (Property.New ("dc:title", base_title));
 179                         AddProperty (Property.NewUnsearched ("fixme:language", base_language));
 180
 181                         watch.Stop ();
 182
 183                         // If we've successfully crawled the file but haven't
 184                         // found any indexables, we shouldn't consider it
 185                         // successfull at all (unless we have a title, which
 186                         // means that it's actually a docbook file, just without
 187                         // sections.
 188                         if (ChildIndexables.Count == 0 && base_title == null) {
 189                                 Error ();
 190                                 return;
 191                         }
 192
 193                         Logger.Log.Debug ("Parsed docbook file in {0}", watch);
 194
 195                         Finished ();
 196                 }
 197         }
 198 }