Filters/FilterSvg.cs

   1 //
   2 // FilterSVG.cs
   3 //
   4 // Copyright (C) 2006 Alexander Macdonald <alex@alexmac.cc>
   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
  32 using Beagle.Util;
  33 using Beagle.Daemon;
  34
  35 namespace Beagle.Filters {
  36         public class FilterSvg : Beagle.Daemon.Filter {
  37                 private StringBuilder sb = new StringBuilder ();
  38
  39                 private enum RdfGrabModes {
  40                         TitleMode,
  41                         DescriptionMode,
  42                         DateMode,
  43                         Num
  44                 };
  45
  46                 static private string [] rdf_grab_strings = { "title", "description", "date" };
  47                 static private string [] rdf_nongrab_strings = { "creator", "contributor", "publisher", "rights" };
  48
  49                 // List of keys that should be ignored when adding to content.
  50                 // For example, dc:format is the mime type, so it's not interesting text.
  51                 static private string [] ignore_strings = { "format" };
  52
  53                 public FilterSvg ()
  54                 {
  55                         AddSupportedFlavor (FilterFlavor.NewFromMimeType ("image/svg+xml"));
  56                         AddSupportedFlavor (FilterFlavor.NewFromExtension (".svg"));
  57                 }
  58
  59                 override protected void DoPullProperties ()
  60                 {
  61                         XmlTextReader reader = new XmlTextReader (Stream);
  62                         reader.XmlResolver = null;
  63
  64                         int depth = 0;
  65                         bool grab_text = false, ignore_text = false;;
  66                         string text = "";
  67
  68                         try {
  69                                 while (reader.Read ()) {
  70                                         switch (reader.NodeType) {
  71                                         case XmlNodeType.Element:
  72                                                 if (grab_text)
  73                                                         break;
  74
  75                                                 if (ArrayFu.IndexOfString (ignore_strings, reader.LocalName) != -1)
  76                                                         ignore_text = true;
  77                                                 else if (reader.LocalName == "title" || reader.LocalName == "desc") {
  78                                                         grab_text = true;
  79                                                         depth = reader.Depth;
  80                                                 } else if (reader.LocalName == "RDF")
  81                                                         PullRdfProperties (reader, reader.Depth);
  82
  83                                                 break;
  84
  85                                         case XmlNodeType.Text:
  86                                                 text = reader.Value.Trim ();
  87                                                 if (text.Length == 0 || ignore_text)
  88                                                         break;
  89
  90                                                 if (grab_text) {
  91                                                         sb.Append (text);
  92                                                 } else {
  93                                                         AppendText (text);
  94                                                         AppendStructuralBreak ();
  95                                                 }
  96                                                 break;
  97
  98                                         case XmlNodeType.Comment:
  99                                                 AppendText (reader.Value.Trim ());
 100                                                 AppendStructuralBreak ();
 101                                                 break;
 102
 103                                         case XmlNodeType.EndElement:
 104                                                 if (! (grab_text && depth == reader.Depth))
 105                                                         break;
 106
 107                                                 grab_text = false;
 108                                                 ignore_text = false;
 109
 110                                                 if (reader.LocalName == "title") {
 111                                                         AddProperty (Property.New ("dc:title", sb.ToString ()));
 112                                                         sb.Length = 0;
 113                                                 } else if (reader.LocalName == "desc") {
 114                                                         AddProperty (Property.New ("dc:description", sb.ToString ()));
 115                                                         sb.Length = 0;
 116                                                 }
 117                                                 break;
 118                                         }
 119                                 }
 120
 121                                 Finished ();
 122                         } catch (System.Xml.XmlException e) {
 123                                 Logger.Log.Error ("error parsing xml file {0}", FileInfo.FullName);
 124                                 Logger.Log.Debug (e);
 125                                 Error ();
 126                         }
 127                 }
 128
 129                 protected void PullRdfProperties (XmlTextReader reader, int depth)
 130                 {
 131                         int grab_mode = -1, nongrab_mode = -1;
 132                         bool grab_text = false, ignore_text = false, agent_mode = false;
 133
 134                         string text = "";
 135
 136                         try {
 137                                 while (reader.Read ()) {
 138                                         if (depth == reader.Depth)
 139                                                 return;
 140
 141                                         switch (reader.NodeType) {
 142                                         case XmlNodeType.Element:
 143                                                 if (grab_text)
 144                                                         break;
 145
 146                                                 if (ArrayFu.IndexOfString (ignore_strings, reader.LocalName) != -1)
 147                                                         ignore_text = true;
 148                                                 else if (reader.LocalName == "Agent")
 149                                                         grab_text = agent_mode = true;
 150                                                 else {
 151                                                         for (int i = 0; i < (int) RdfGrabModes.Num; i++) {
 152                                                                 if (reader.LocalName == rdf_grab_strings [i]) {
 153                                                                         grab_text = true;
 154                                                                         grab_mode = i;
 155                                                                         break;
 156                                                                 }
 157                                                         }
 158
 159                                                         for (int i = 0; i < (int) RdfGrabModes.Num; i++) {
 160                                                                 if (reader.LocalName == rdf_nongrab_strings [i]) {
 161                                                                         grab_text = false;
 162                                                                         nongrab_mode = i;
 163                                                                         break;
 164                                                                 }
 165                                                         }
 166                                                 }
 167                                                 break;
 168
 169                                         case XmlNodeType.Text:
 170                                                 text = reader.Value.Trim ();
 171                                                 if (text.Length == 0 || ignore_text)
 172                                                         break;
 173
 174                                                 if (grab_text) {
 175                                                         sb.Append(text);
 176                                                 } else {
 177                                                         AppendText (text);
 178                                                         AppendStructuralBreak ();
 179                                                 }
 180                                                 break;
 181
 182                                         case XmlNodeType.EndElement:
 183                                                 ignore_text = false;
 184
 185                                                 if (agent_mode) {
 186                                                         if (reader.LocalName ==  "Agent") {
 187                                                                 agent_mode = grab_text = false;
 188                                                                 AddProperty (Property.New ("dc:" + rdf_nongrab_strings [nongrab_mode], sb.ToString ()));
 189                                                                 sb.Length = 0;
 190                                                         }
 191                                                 } else if (grab_mode >= 0 && reader.LocalName == rdf_grab_strings [grab_mode]) {
 192                                                         if (grab_mode == (int) RdfGrabModes.DateMode) {
 193                                                                 try {
 194                                                                         AddProperty (Property.NewDate ("dc:date", System.Convert.ToDateTime (sb.ToString ())));
 195                                                                 } catch (FormatException) {
 196                                                                         AddProperty (Property.New ("dc:date", sb.ToString ()));
 197                                                                 }
 198                                                         } else {
 199                                                                 AddProperty (Property.New ("dc:" + rdf_grab_strings [grab_mode], sb.ToString ()));
 200                                                         }
 201                                                         sb.Length = 0;
 202                                                         grab_mode = -1;
 203                                                         grab_text = false;
 204                                                 } else if (nongrab_mode >= 0 && reader.LocalName == rdf_nongrab_strings [nongrab_mode]) {
 205                                                         nongrab_mode = -1;
 206                                                         grab_text = false;
 207                                                 }
 208                                                 break;
 209                                         }
 210                                 }
 211                         } catch (System.Xml.XmlException e) {
 212                                 Logger.Log.Error ("error parsing embedded RDF {0}", FileInfo.FullName);
 213                                 Logger.Log.Debug (e);
 214                                 Error ();
 215                         }
 216                 }
 217         }
 218 }