forester/java/src/org/forester/io/parsers/nexus/NexusPhylogeniesParser.java

   1 // $Id:
   2 // FORESTER -- software libraries and applications
   3 // for evolutionary biology research and applications.
   4 //
   5 // Copyright (C) 2008-2009 Christian M. Zmasek
   6 // Copyright (C) 2008-2009 Burnham Institute for Medical Research
   7 // All rights reserved
   8 //
   9 // This library is free software; you can redistribute it and/or
  10 // modify it under the terms of the GNU Lesser General Public
  11 // License as published by the Free Software Foundation; either
  12 // version 2.1 of the License, or (at your option) any later version.
  13 //
  14 // This library is distributed in the hope that it will be useful,
  15 // but WITHOUT ANY WARRANTY; without even the implied warranty of
  16 // MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  17 // Lesser General Public License for more details.
  18 //
  19 // You should have received a copy of the GNU Lesser General Public
  20 // License along with this library; if not, write to the Free Software
  21 // Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
  22 //
  23 // Contact: phylosoft @ gmail . com
  24 // WWW: https://sites.google.com/site/cmzmasek/home/software/forester
  25
  26 package org.forester.io.parsers.nexus;
  27
  28 import java.io.BufferedReader;
  29 import java.io.FileNotFoundException;
  30 import java.io.IOException;
  31 import java.util.ArrayList;
  32 import java.util.HashMap;
  33 import java.util.List;
  34 import java.util.Map;
  35 import java.util.regex.Matcher;
  36 import java.util.regex.Pattern;
  37
  38 import org.forester.archaeopteryx.Constants;
  39 import org.forester.io.parsers.IteratingPhylogenyParser;
  40 import org.forester.io.parsers.PhylogenyParser;
  41 import org.forester.io.parsers.nhx.NHXFormatException;
  42 import org.forester.io.parsers.nhx.NHXParser;
  43 import org.forester.io.parsers.nhx.NHXParser.TAXONOMY_EXTRACTION;
  44 import org.forester.io.parsers.util.ParserUtils;
  45 import org.forester.io.parsers.util.PhylogenyParserException;
  46 import org.forester.phylogeny.Phylogeny;
  47 import org.forester.phylogeny.PhylogenyNode;
  48 import org.forester.phylogeny.iterators.PhylogenyNodeIterator;
  49 import org.forester.util.ForesterUtil;
  50
  51 public final class NexusPhylogeniesParser implements IteratingPhylogenyParser, PhylogenyParser {
  52
  53     final private static String  begin_trees               = NexusConstants.BEGIN_TREES.toLowerCase();
  54     final private static String  taxlabels                 = NexusConstants.TAXLABELS.toLowerCase();
  55     final private static String  translate                 = NexusConstants.TRANSLATE.toLowerCase();
  56     final private static String  tree                      = NexusConstants.TREE.toLowerCase();
  57     final private static String  utree                     = NexusConstants.UTREE.toLowerCase();
  58     final private static String  end                       = NexusConstants.END.toLowerCase();
  59     final private static String  endblock                  = "endblock";
  60     final private static Pattern TREE_NAME_PATTERN         = Pattern.compile( "\\s*.?Tree\\s+(.+?)\\s*=.+",
  61                                                                               Pattern.CASE_INSENSITIVE );
  62     final private static Pattern ROOTEDNESS_PATTERN        = Pattern.compile( ".+=\\s*\\[&([R|U])\\].*" );
  63     private Object               _nexus_source;
  64     private List<String>         _taxlabels;
  65     private Map<String, String>  _translate_map;
  66     private boolean              _replace_underscores      = NHXParser.REPLACE_UNDERSCORES_DEFAULT;
  67     private boolean              _ignore_quotes_in_nh_data = Constants.NH_PARSING_IGNORE_QUOTES_DEFAULT;
  68     private TAXONOMY_EXTRACTION  _taxonomy_extraction      = TAXONOMY_EXTRACTION.NO;
  69     private Phylogeny            _next;
  70     private BufferedReader       _br;
  71     private boolean              _in_trees_block;
  72     private StringBuilder        _nh;
  73     private String               _name;
  74     private StringBuilder        _translate_sb;
  75     private boolean              _in_taxalabels;
  76     private boolean              _in_translate;
  77     private boolean              _is_rooted;
  78     private boolean              _rooted_info_present;
  79     private boolean              _in_tree;
  80
  81     @Override
  82     public final boolean hasNext() {
  83         return _next != null;
  84     }
  85
  86     @Override
  87     public final Phylogeny next() throws NHXFormatException, IOException {
  88         final Phylogeny phy = _next;
  89         getNext();
  90         return phy;
  91     }
  92
  93     @Override
  94     public final Phylogeny[] parse() throws IOException {
  95         final List<Phylogeny> l = new ArrayList<Phylogeny>();
  96         while ( hasNext() ) {
  97             l.add( next() );
  98         }
  99         final Phylogeny[] p = new Phylogeny[ l.size() ];
 100         for( int i = 0; i < l.size(); ++i ) {
 101             p[ i ] = l.get( i );
 102         }
 103         reset();
 104         return p;
 105     }
 106
 107     @Override
 108     public final void reset() throws FileNotFoundException, IOException {
 109         _taxlabels = new ArrayList<String>();
 110         _translate_map = new HashMap<String, String>();
 111         _nh = new StringBuilder();
 112         _name = "";
 113         _translate_sb = new StringBuilder();
 114         _next = null;
 115         _in_trees_block = false;
 116         _in_taxalabels = false;
 117         _in_translate = false;
 118         _in_tree = false;
 119         _rooted_info_present = false;
 120         _is_rooted = false;
 121         _br = ParserUtils.createReader( _nexus_source );
 122         getNext();
 123     }
 124
 125     public final void setIgnoreQuotes( final boolean ignore_quotes_in_nh_data ) {
 126         _ignore_quotes_in_nh_data = ignore_quotes_in_nh_data;
 127     }
 128
 129     public final void setReplaceUnderscores( final boolean replace_underscores ) {
 130         _replace_underscores = replace_underscores;
 131     }
 132
 133     @Override
 134     public final void setSource( final Object nexus_source ) throws PhylogenyParserException, IOException {
 135         if ( nexus_source == null ) {
 136             throw new PhylogenyParserException( "attempt to parse null object" );
 137         }
 138         _nexus_source = nexus_source;
 139         reset();
 140     }
 141
 142     public final void setTaxonomyExtraction( final TAXONOMY_EXTRACTION taxonomy_extraction ) {
 143         _taxonomy_extraction = taxonomy_extraction;
 144     }
 145
 146     private final void createPhylogeny( final String name,
 147                                         final StringBuilder nhx,
 148                                         final boolean rooted_info_present,
 149                                         final boolean is_rooted ) throws IOException {
 150         _next = null;
 151         final NHXParser pars = new NHXParser();
 152         pars.setTaxonomyExtraction( _taxonomy_extraction );
 153         pars.setReplaceUnderscores( _replace_underscores );
 154         pars.setIgnoreQuotes( _ignore_quotes_in_nh_data );
 155         if ( rooted_info_present ) {
 156             pars.setGuessRootedness( false );
 157         }
 158         pars.setSource( nhx );
 159         final Phylogeny p = pars.next();
 160         if ( p == null ) {
 161             throw new PhylogenyParserException( "failed to create phylogeny" );
 162         }
 163         p.setName( name );
 164         if ( rooted_info_present ) {
 165             p.setRooted( is_rooted );
 166         }
 167         if ( ( _taxlabels.size() > 0 ) || ( _translate_map.size() > 0 ) ) {
 168             final PhylogenyNodeIterator it = p.iteratorExternalForward();
 169             while ( it.hasNext() ) {
 170                 final PhylogenyNode node = it.next();
 171                 if ( ( _translate_map.size() > 0 ) && _translate_map.containsKey( node.getName() ) ) {
 172                     node.setName( _translate_map.get( node.getName() ).replaceAll( "['\"]+", "" ) );
 173                 }
 174                 else if ( _taxlabels.size() > 0 ) {
 175                     int i = -1;
 176                     try {
 177                         i = Integer.parseInt( node.getName() );
 178                     }
 179                     catch ( final NumberFormatException e ) {
 180                         // Ignore.
 181                     }
 182                     if ( i > 0 ) {
 183                         node.setName( _taxlabels.get( i - 1 ).replaceAll( "['\"]+", "" ) );
 184                     }
 185                 }
 186                 if ( !_replace_underscores && ( ( _taxonomy_extraction != TAXONOMY_EXTRACTION.NO ) ) ) {
 187                     ParserUtils.extractTaxonomyDataFromNodeName( node, _taxonomy_extraction );
 188                 }
 189             }
 190         }
 191         _next = p;
 192     }
 193
 194     private final void getNext() throws IOException, NHXFormatException {
 195         _next = null;
 196         String line;
 197         while ( ( line = _br.readLine() ) != null ) {
 198             line = line.trim();
 199             if ( ( line.length() > 0 ) && !line.startsWith( "#" ) && !line.startsWith( ">" ) ) {
 200                 line = ForesterUtil.collapseWhiteSpace( line );
 201                 line = removeWhiteSpaceBeforeSemicolon( line );
 202                 final String line_lc = line.toLowerCase();
 203                 if ( line_lc.startsWith( begin_trees ) ) {
 204                     _in_trees_block = true;
 205                     _in_taxalabels = false;
 206                     _in_translate = false;
 207                 }
 208                 else if ( line_lc.startsWith( taxlabels ) ) {
 209                     _in_trees_block = false;
 210                     _in_taxalabels = true;
 211                     _in_translate = false;
 212                 }
 213                 else if ( line_lc.startsWith( translate ) ) {
 214                     _in_taxalabels = false;
 215                     _in_translate = true;
 216                 }
 217                 else if ( _in_trees_block ) {
 218                     //FIXME TODO need to work on this "title" and "link"
 219                     if ( line_lc.startsWith( "title" ) || line_lc.startsWith( "link" ) ) {
 220                         // Do nothing.
 221                     }
 222                     else if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 223                         _in_trees_block = false;
 224                         _in_tree = false;
 225                         _in_translate = false;
 226                         if ( _nh.length() > 0 ) {
 227                             createPhylogeny( _name, _nh, _rooted_info_present, _is_rooted );
 228                             _nh = new StringBuilder();
 229                             _name = "";
 230                             _rooted_info_present = false;
 231                             _is_rooted = false;
 232                             if ( _next != null ) {
 233                                 return;
 234                             }
 235                         }
 236                     }
 237                     else if ( line_lc.startsWith( tree ) || ( line_lc.startsWith( utree ) ) ) {
 238                         boolean might = false;
 239                         if ( _nh.length() > 0 ) {
 240                             might = true;
 241                             createPhylogeny( _name, _nh, _rooted_info_present, _is_rooted );
 242                             _nh = new StringBuilder();
 243                             _name = "";
 244                             _rooted_info_present = false;
 245                             _is_rooted = false;
 246                         }
 247                         _in_tree = true;
 248                         _nh.append( line.substring( line.indexOf( '=' ) ) );
 249                         final Matcher name_matcher = TREE_NAME_PATTERN.matcher( line );
 250                         if ( name_matcher.matches() ) {
 251                             _name = name_matcher.group( 1 );
 252                             _name = _name.replaceAll( "['\"]+", "" );
 253                         }
 254                         final Matcher rootedness_matcher = ROOTEDNESS_PATTERN.matcher( line );
 255                         if ( rootedness_matcher.matches() ) {
 256                             final String s = rootedness_matcher.group( 1 );
 257                             line = line.replaceAll( "\\[\\&.\\]", "" );
 258                             _rooted_info_present = true;
 259                             if ( s.toUpperCase().equals( "R" ) ) {
 260                                 _is_rooted = true;
 261                             }
 262                         }
 263                         if ( might && ( _next != null ) ) {
 264                             return;
 265                         }
 266                     }
 267                     else if ( _in_tree && !_in_translate ) {
 268                         _nh.append( line );
 269                     }
 270                     if ( !line_lc.startsWith( "title" ) && !line_lc.startsWith( "link" ) && !_in_translate
 271                             && !line_lc.startsWith( end ) && !line_lc.startsWith( endblock ) && line_lc.endsWith( ";" ) ) {
 272                         _in_tree = false;
 273                         _in_translate = false;
 274                         createPhylogeny( _name, _nh, _rooted_info_present, _is_rooted );
 275                         _nh = new StringBuilder();
 276                         _name = "";
 277                         _rooted_info_present = false;
 278                         _is_rooted = false;
 279                         if ( _next != null ) {
 280                             return;
 281                         }
 282                     }
 283                 }
 284                 if ( _in_taxalabels ) {
 285                     if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 286                         _in_taxalabels = false;
 287                     }
 288                     else {
 289                         final String[] labels = line.split( "\\s+" );
 290                         for( String label : labels ) {
 291                             if ( !label.toLowerCase().equals( taxlabels ) ) {
 292                                 if ( label.endsWith( ";" ) ) {
 293                                     _in_taxalabels = false;
 294                                     label = label.substring( 0, label.length() - 1 );
 295                                 }
 296                                 if ( label.length() > 0 ) {
 297                                     _taxlabels.add( label );
 298                                 }
 299                             }
 300                         }
 301                     }
 302                 }
 303                 if ( _in_translate ) {
 304                     if ( line_lc.startsWith( end ) || line_lc.startsWith( endblock ) ) {
 305                         _in_translate = false;
 306                     }
 307                     else {
 308                         _translate_sb.append( " " );
 309                         _translate_sb.append( line.trim() );
 310                         if ( line.endsWith( ";" ) ) {
 311                             _in_translate = false;
 312                             setTranslateKeyValuePairs( _translate_sb );
 313                         }
 314                     }
 315                 }
 316             }
 317         }
 318         if ( _nh.length() > 0 ) {
 319             createPhylogeny( _name, _nh, _rooted_info_present, _is_rooted );
 320             if ( _next != null ) {
 321                 return;
 322             }
 323         }
 324     }
 325
 326     private final void setTranslateKeyValuePairs( final StringBuilder translate_sb ) throws IOException {
 327         String s = translate_sb.toString().trim();
 328         if ( s.endsWith( ";" ) ) {
 329             s = s.substring( 0, s.length() - 1 ).trim();
 330         }
 331         for( final String pair : s.split( "," ) ) {
 332             final String[] kv = pair.trim().split( "\\s+" );
 333             if ( ( kv.length < 2 ) || ( kv.length > 3 ) ) {
 334                 throw new IOException( "ill-formatted translate values: " + translate_sb );
 335             }
 336             if ( ( kv.length == 3 ) && !kv[ 0 ].toLowerCase().trim().equals( translate ) ) {
 337                 throw new IOException( "ill-formatted translate values: " + translate_sb );
 338             }
 339             String key = "";
 340             String value = "";
 341             if ( kv.length == 3 ) {
 342                 key = kv[ 1 ];
 343                 value = kv[ 2 ];
 344             }
 345             else {
 346                 key = kv[ 0 ];
 347                 value = kv[ 1 ];
 348             }
 349             if ( value.endsWith( ";" ) ) {
 350                 value = value.substring( 0, value.length() - 1 );
 351             }
 352             _translate_map.put( key, value );
 353         }
 354     }
 355
 356     private final static String removeWhiteSpaceBeforeSemicolon( final String s ) {
 357         return s.replaceAll( "\\s+;", ";" );
 358     }
 359
 360     @Override
 361     public String getName() {
 362         return "Nexus Phylogenies Parser";
 363     }
 364 }