app-import/src/main/java/eu/etaxonomy/cdm/io/iapt/IAPTExcelImport.java

   1 /**
   2  * Copyright (C) 2007 EDIT
   3  * European Distributed Institute of Taxonomy
   4  * http://www.e-taxonomy.eu
   5  *
   6  * The contents of this file are subject to the Mozilla Public License Version 1.1
   7  * See LICENSE.TXT at the top of this package for the full license terms.
   8  */
   9
  10 package eu.etaxonomy.cdm.io.iapt;
  11
  12 import eu.etaxonomy.cdm.api.facade.DerivedUnitFacade;
  13 import eu.etaxonomy.cdm.common.CdmUtils;
  14 import eu.etaxonomy.cdm.io.mexico.SimpleExcelTaxonImport;
  15 import eu.etaxonomy.cdm.io.mexico.SimpleExcelTaxonImportState;
  16 import eu.etaxonomy.cdm.model.agent.Institution;
  17 import eu.etaxonomy.cdm.model.agent.Person;
  18 import eu.etaxonomy.cdm.model.agent.TeamOrPersonBase;
  19 import eu.etaxonomy.cdm.model.common.*;
  20 import eu.etaxonomy.cdm.model.name.*;
  21 import eu.etaxonomy.cdm.model.occurrence.*;
  22 import eu.etaxonomy.cdm.model.occurrence.Collection;
  23 import eu.etaxonomy.cdm.model.reference.Reference;
  24 import eu.etaxonomy.cdm.model.reference.ReferenceFactory;
  25 import eu.etaxonomy.cdm.model.reference.ReferenceType;
  26 import eu.etaxonomy.cdm.model.taxon.*;
  27 import eu.etaxonomy.cdm.strategy.parser.NonViralNameParserImpl;
  28 import eu.etaxonomy.cdm.strategy.parser.ParserProblem;
  29 import org.apache.commons.lang.ArrayUtils;
  30 import org.apache.commons.lang.StringEscapeUtils;
  31 import org.apache.commons.lang.StringUtils;
  32 import org.apache.log4j.Level;
  33 import org.apache.log4j.Logger;
  34 import org.joda.time.DateTimeFieldType;
  35 import org.joda.time.Partial;
  36 import org.joda.time.format.DateTimeFormat;
  37 import org.joda.time.format.DateTimeFormatter;
  38 import org.springframework.stereotype.Component;
  39
  40 import java.util.*;
  41 import java.util.regex.Matcher;
  42 import java.util.regex.Pattern;
  43
  44 /**
  45  * @author a.mueller
  46  * @created 05.01.2016
  47  */
  48
  49 @Component("iAPTExcelImport")
  50 public class IAPTExcelImport<CONFIG extends IAPTImportConfigurator> extends SimpleExcelTaxonImport<CONFIG> {
  51     private static final long serialVersionUID = -747486709409732371L;
  52     private static final Logger logger = Logger.getLogger(IAPTExcelImport.class);
  53     public static final String ANNOTATION_MARKER_STRING = "[*]";
  54
  55
  56     private static UUID ROOT_UUID = UUID.fromString("4137fd2a-20f6-4e70-80b9-f296daf51d82");
  57
  58     private static NonViralNameParserImpl nameParser = NonViralNameParserImpl.NewInstance();
  59
  60     private final static String REGISTRATIONNO_PK= "RegistrationNo_Pk";
  61     private final static String HIGHERTAXON= "HigherTaxon";
  62     private final static String FULLNAME= "FullName";
  63     private final static String AUTHORSSPELLING= "AuthorsSpelling";
  64     private final static String LITSTRING= "LitString";
  65     private final static String REGISTRATION= "Registration";
  66     private final static String TYPE= "Type";
  67     private final static String CAVEATS= "Caveats";
  68     private final static String FULLBASIONYM= "FullBasionym";
  69     private final static String FULLSYNSUBST= "FullSynSubst";
  70     private final static String NOTESTXT= "NotesTxt";
  71     private final static String REGDATE= "RegDate";
  72     private final static String NAMESTRING= "NameString";
  73     private final static String BASIONYMSTRING= "BasionymString";
  74     private final static String SYNSUBSTSTR= "SynSubstStr";
  75     private final static String AUTHORSTRING= "AuthorString";
  76
  77     private  static List<String> expectedKeys= Arrays.asList(new String[]{
  78             REGISTRATIONNO_PK, HIGHERTAXON, FULLNAME, AUTHORSSPELLING, LITSTRING, REGISTRATION, TYPE, CAVEATS, FULLBASIONYM, FULLSYNSUBST, NOTESTXT, REGDATE, NAMESTRING, BASIONYMSTRING, SYNSUBSTSTR, AUTHORSTRING});
  79
  80     private static final Pattern nomRefTokenizeP = Pattern.compile("^(?<title>.*):\\s(?<detail>[^\\.:]+)\\.(?<date>.*?)(?:\\s\\((?<issue>[^\\)]*)\\)\\s*)?\\.?$");
  81     private static final Pattern[] datePatterns = new Pattern[]{
  82             // NOTE:
  83             // The order of the patterns is extremely important!!!
  84             //
  85             // all patterns cover the years 1700 - 1999
  86             Pattern.compile("^(?<year>1[7,8,9][0-9]{2})$"), // only year, like '1969'
  87             Pattern.compile("^(?<monthName>\\p{L}+\\.?)\\s(?<day>[0-9]{1,2})(?:st|rd|th)?\\.?,?\\s(?<year>(?:1[7,8,9])?[0-9]{2})$"), // full date like April 12, 1969 or april 12th 1999
  88             Pattern.compile("^(?<monthName>\\p{L}+\\.?),?\\s?(?<year>(?:1[7,8,9])?[0-9]{2})$"), // April 99 or April, 1999 or Apr. 12
  89             Pattern.compile("^(?<day>[0-9]{1,2})([\\.\\-/])(\\s?)(?<month>[0-1]?[0-9])\\2\\3(?<year>(?:1[7,8,9])?[0-9]{2})$"), // full date like 12.04.1969 or 12. 04. 1969 or 12/04/1969 or 12-04-1969
  90             Pattern.compile("^(?<day>[0-9]{1,2})([\\.\\-/])(?<month>[IVX]{1,2})\\2(?<year>(?:1[7,8,9])?[0-9]{2})$"), // full date like 12-VI-1969
  91             Pattern.compile("^(?:(?<day>[0-9]{1,2})(?:\\sde)\\s)(?<monthName>\\p{L}+)\\sde\\s(?<year>(?:1[7,8,9])?[0-9]{2})$"), // full and partial date like 12 de Enero de 1999 or Enero de 1999
  92             Pattern.compile("^(?<month>[0-1]?[0-9])([\\.\\-/])(?<year>(?:1[7,8,9])?[0-9]{2})$"), // partial date like 04.1969 or 04/1969 or 04-1969
  93             Pattern.compile("^(?<year>(?:1[7,8,9])?[0-9]{2})([\\.\\-/])(?<month>[0-1]?[0-9])$"),//  partial date like 1999-04
  94             Pattern.compile("^(?<month>[IVX]{1,2})([\\.\\-/])(?<year>(?:1[7,8,9])?[0-9]{2})$"), // partial date like VI-1969
  95             Pattern.compile("^(?<day>[0-9]{1,2})(?:[\\./]|th|rd|st)?\\s(?<monthName>\\p{L}+\\.?),?\\s?(?<year>(?:1[7,8,9])?[0-9]{2})$"), // full date like 12. April 1969 or april 1999 or 22 Dec.1999
  96         };
  97     private static final Pattern typeSpecimenSplitPattern =  Pattern.compile("^(?:\"*[Tt]ype: (?<fieldUnit>.*?))(?:[Hh]olotype:(?<holotype>.*?)\\.?)?(?:[Ii]sotype[^:]*:(?<isotype>.*)\\.?)?\\.?$");
  98
  99     private static final Pattern typeNameBasionymPattern =  Pattern.compile("\\([Bb]asionym\\s?\\:\\s?(?<basionymName>[^\\)]*).*$");
 100     private static final Pattern typeNameNotePattern =  Pattern.compile("\\[([^\\[]*)"); // matches the inner of '[...]'
 101     private static final Pattern typeNameSpecialSplitPattern =  Pattern.compile("(?<note>.*\\;.*?)\\:(?<agent>)\\;(<name>.*)");
 102
 103     private static final Pattern collectorPattern =  Pattern.compile(".*?(?<fullStr1>\\(leg\\.\\s+(?<data1>[^\\)]*)\\))|.*?(?<fullStr2>\\sleg\\.\\s+(?<data2>.*?)\\.?)$");
 104     private static final Pattern collectionDataPattern =  Pattern.compile("^(?<collector>[^,]*),\\s?(?<detail>.*?)\\.?$");
 105     private static final Pattern collectorsNumber =  Pattern.compile("^([nN]o\\.\\s.*)$");
 106
 107     // AccessionNumbers: , #.*, n°:?, 96/3293, No..*, -?\w{1,3}-[0-9\-/]*
 108     private static final Pattern accessionNumberOnlyPattern = Pattern.compile("^(?<accNumber>(?:n°\\:?\\s?|#|No\\.?\\s?)?[\\d\\w\\-/]*)$");
 109
 110     private static final Pattern[] specimenTypePatterns = new Pattern[]{
 111             Pattern.compile("^(?<colCode>[A-Z]+|CPC Micropaleontology Lab\\.?)\\s+(?:\\((?<institute>.*[^\\)])\\))(?<accNumber>.*)?$"), // like: GAUF (Gansu Agricultural University) No. 1207-1222
 112             Pattern.compile("^(?<colCode>[A-Z]+|CPC Micropaleontology Lab\\.?)\\s+(?:Coll\\.\\s(?<subCollection>[^\\.,;]*)(.))(?<accNumber>.*)?$"), // like KASSEL Coll. Krasske, Praep. DII 78
 113             Pattern.compile("^(?:Coll\\.\\s(?<subCollection>[^\\.,;]*)(.))(?<institute>.*?)(?<accNumber>Praep\\..*)?$"), // like Coll. Lange-Bertalot, Bot. Inst., Univ. Frankfurt/Main, Germany Praep. Neukaledonien OTL 62
 114             Pattern.compile("^(?<colCode>[A-Z]+)(?:\\s+(?<accNumber>.*))?$"), // identifies the Collection code and takes the rest as accessionNumber if any
 115     };
 116
 117     private static Map<String, Integer> monthFromNameMap = new HashMap<>();
 118
 119     static {
 120         String[] ck = new String[]{"leden", "únor", "březen", "duben", "květen", "červen", "červenec ", "srpen", "září", "říjen", "listopad", "prosinec"};
 121         String[] fr = new String[]{"janvier", "février", "mars", "avril", "mai", "juin", "juillet", "août", "septembre", "octobre", "novembre", "décembre"};
 122         String[] de = new String[]{"januar", "februar", "märz", "april", "mai", "juni", "juli", "august", "september", "oktober", "november", "dezember"};
 123         String[] en = new String[]{"january", "february", "march", "april", "may", "june", "july", "august", "september", "october", "november", "december"};
 124         String[] it = new String[]{"gennaio", "febbraio", "marzo", "aprile", "maggio", "giugno", "luglio", "agosto", "settembre", "ottobre", "novembre", "dicembre"};
 125         String[] sp = new String[]{"enero", "febrero", "marzo", "abril", "mayo", "junio", "julio", "agosto", "septiembre", "octubre", "noviembre", "diciembre"};
 126         String[] de_abbrev = new String[]{"jan.", "feb.", "märz", "apr.", "mai", "jun.", "jul.", "aug.", "sept.", "okt.", "nov.", "dez."};
 127         String[] en_abbrev = new String[]{"jan.", "feb.", "mar.", "apr.", "may", "jun.", "jul.", "aug.", "sep.", "oct.", "nov.", "dec."};
 128         String[] port = new String[]{"Janeiro", "Fevereiro", "Março", "Abril", "Maio", "Junho", "Julho", "Agosto", "Setembro", "Outubro", "Novembro", "Dezembro"};
 129         String[] rom_num = new String[]{"i", "ii", "iii", "iv", "v", "vi", "vii", "viii", "ix", "x", "xi", "xii"};
 130
 131         String[][] perLang =  new String[][]{ck, de, fr, en, it, sp, port, de_abbrev, en_abbrev, rom_num};
 132
 133         for (String[] months: perLang) {
 134             for(int m = 1; m < 13; m++){
 135                 monthFromNameMap.put(months[m - 1].toLowerCase(), m);
 136             }
 137         }
 138
 139         // special cases
 140         monthFromNameMap.put("mar", 3);
 141         monthFromNameMap.put("dec", 12);
 142         monthFromNameMap.put("Februari", 2);
 143     }
 144
 145
 146     DateTimeFormatter formatterYear = DateTimeFormat.forPattern("yyyy");
 147
 148     private Map<String, Collection> collectionMap = new HashMap<>();
 149
 150
 151     enum TypesName {
 152         fieldUnit, holotype, isotype;
 153
 154         public SpecimenTypeDesignationStatus status(){
 155             switch (this) {
 156                 case holotype:
 157                     return SpecimenTypeDesignationStatus.HOLOTYPE();
 158                 case isotype:
 159                     return SpecimenTypeDesignationStatus.ISOTYPE();
 160                 default:
 161                     return null;
 162             }
 163         }
 164     }
 165
 166     private MarkerType markerTypeFossil = null;
 167     private Rank rankUnrankedSupraGeneric = null;
 168     private Rank familyIncertisSedis = null;
 169     private AnnotationType annotationTypeCaveats = null;
 170
 171     private Reference bookVariedadesTradicionales = null;
 172
 173     private Taxon makeTaxon(HashMap<String, String> record, SimpleExcelTaxonImportState<CONFIG> state,
 174                             TaxonNode higherTaxonNode, boolean isFossil) {
 175
 176         String regNumber = getValue(record, REGISTRATIONNO_PK, false);
 177         String regStr = getValue(record, REGISTRATION, true);
 178         String titleCacheStr = getValue(record, FULLNAME, true);
 179         String nameStr = getValue(record, NAMESTRING, true);
 180         String authorStr = getValue(record, AUTHORSTRING, true);
 181         String nomRefStr = getValue(record, LITSTRING, true);
 182         String authorsSpelling = getValue(record, AUTHORSSPELLING, true);
 183         String notesTxt = getValue(record, NOTESTXT, true);
 184         String caveats = getValue(record, CAVEATS, true);
 185         String fullSynSubstStr = getValue(record, FULLSYNSUBST, true);
 186         String fullBasionymStr = getValue(record, FULLBASIONYM, true);
 187         String basionymNameStr = getValue(record, FULLBASIONYM, true);
 188         String synSubstStr = getValue(record, SYNSUBSTSTR, true);
 189         String typeStr = getValue(record, TYPE, true);
 190
 191
 192         String nomRefTitle = null;
 193         String nomRefDetail;
 194         String nomRefPupDate = null;
 195         String nomRefIssue = null;
 196         Partial pupDate = null;
 197
 198         boolean restoreOriginalReference = false;
 199         boolean nameIsValid = true;
 200
 201         // preprocess nomRef: separate citation, reference detail, publishing date
 202         if(!StringUtils.isEmpty(nomRefStr)){
 203             nomRefStr = nomRefStr.trim();
 204
 205             // handle the special case which is hard to parse:
 206             //
 207             // Las variedades tradicionales de frutales de la Cuenca del Río Segura. Catálogo Etnobotánico (1): Frutos secos, oleaginosos, frutales de hueso, almendros y frutales de pepita: 154. 1997.
 208             if(nomRefStr.startsWith("Las variedades tradicionales de frutales ")){
 209
 210                 if(bookVariedadesTradicionales == null){
 211                     bookVariedadesTradicionales = ReferenceFactory.newBook();
 212                     bookVariedadesTradicionales.setTitle("Las variedades tradicionales de frutales de la Cuenca del Río Segura. Catálogo Etnobotánico (1): Frutos secos, oleaginosos, frutales de hueso, almendros y frutales de pepita");
 213                     bookVariedadesTradicionales.setDatePublished(TimePeriod.NewInstance(1997));
 214                     getReferenceService().save(bookVariedadesTradicionales);
 215                 }
 216                 nomRefStr = nomRefStr.replaceAll("^.*?\\:.*?\\:", "Las variedades tradicionales:");
 217                 restoreOriginalReference = true;
 218             }
 219
 220             Matcher m = nomRefTokenizeP.matcher(nomRefStr);
 221             if(m.matches()){
 222                 nomRefTitle = m.group("title");
 223                 nomRefDetail = m.group("detail");
 224                 nomRefPupDate = m.group("date").trim();
 225                 nomRefIssue = m.group("issue");
 226
 227                 pupDate = parseDate(regNumber, nomRefPupDate);
 228                 if (pupDate != null) {
 229                     nomRefTitle = nomRefTitle + ": " + nomRefDetail + ". " + pupDate.toString(formatterYear) + ".";
 230                 } else {
 231                     logger.warn(csvReportLine(regNumber, "Pub date", nomRefPupDate, "in", nomRefStr, "not parsable"));
 232                 }
 233             } else {
 234                 nomRefTitle = nomRefStr;
 235             }
 236         }
 237
 238         BotanicalName taxonName = makeBotanicalName(state, regNumber, titleCacheStr, nameStr, authorStr, nomRefTitle);
 239
 240         // always add the original strings of parsed data as annotation
 241         taxonName.addAnnotation(Annotation.NewInstance("imported and parsed data strings:" +
 242                         "\n -  '" + LITSTRING + "': "+ nomRefStr +
 243                         "\n -  '" + TYPE + "': " + typeStr +
 244                         "\n -  '" + REGISTRATION  + "': " + regStr
 245                 , AnnotationType.TECHNICAL(), Language.DEFAULT()));
 246
 247         if(restoreOriginalReference){
 248             taxonName.setNomenclaturalReference(bookVariedadesTradicionales);
 249         }
 250         if(pupDate != null) {
 251             taxonName.getNomenclaturalReference().setDatePublished(TimePeriod.NewInstance(pupDate));
 252         }
 253         if(nomRefIssue != null) {
 254             ((Reference)taxonName.getNomenclaturalReference()).setVolume(nomRefIssue);
 255         }
 256
 257
 258         if(!StringUtils.isEmpty(notesTxt)){
 259             notesTxt = notesTxt.replace("Notes: ", "").trim();
 260             taxonName.addAnnotation(Annotation.NewInstance(notesTxt, AnnotationType.EDITORIAL(), Language.DEFAULT()));
 261             nameIsValid = false;
 262
 263         }
 264         if(!StringUtils.isEmpty(caveats)){
 265             caveats = caveats.replace("Caveats: ", "").trim();
 266             taxonName.addAnnotation(Annotation.NewInstance(caveats, annotationTypeCaveats(), Language.DEFAULT()));
 267             nameIsValid = false;
 268         }
 269
 270         if(nameIsValid){
 271             // Status is always considered valid if no notes and cavets are set
 272             taxonName.addStatus(NomenclaturalStatus.NewInstance(NomenclaturalStatusType.VALID()));
 273         }
 274
 275         getNameService().save(taxonName);
 276
 277         // Namerelations
 278         if(!StringUtils.isEmpty(authorsSpelling)){
 279             authorsSpelling = authorsSpelling.replaceFirst("Author's spelling:", "").replaceAll("\"", "").trim();
 280
 281             String[] authorSpellingTokens = StringUtils.split(authorsSpelling, " ");
 282             String[] nameStrTokens = StringUtils.split(nameStr, " ");
 283
 284             ArrayUtils.reverse(authorSpellingTokens);
 285             ArrayUtils.reverse(nameStrTokens);
 286
 287             for (int i = 0; i < nameStrTokens.length; i++){
 288                 if(i < authorSpellingTokens.length){
 289                     nameStrTokens[i] = authorSpellingTokens[i];
 290                 }
 291             }
 292             ArrayUtils.reverse(nameStrTokens);
 293
 294             String misspelledNameStr = StringUtils.join (nameStrTokens, ' ');
 295             // build the fullnameString of the misspelled name
 296             misspelledNameStr = taxonName.getTitleCache().replace(nameStr, misspelledNameStr);
 297
 298             TaxonNameBase misspelledName = (BotanicalName) nameParser.parseReferencedName(misspelledNameStr, NomenclaturalCode.ICNAFP, null);
 299             misspelledName.addRelationshipToName(taxonName, NameRelationshipType.MISSPELLING(), null);
 300             getNameService().save(misspelledName);
 301         }
 302
 303         // Replaced Synonyms
 304         if(!StringUtils.isEmpty(fullSynSubstStr)){
 305             fullSynSubstStr = fullSynSubstStr.replace("Syn. subst.: ", "");
 306             BotanicalName replacedSynonymName = makeBotanicalName(state, regNumber, fullSynSubstStr, synSubstStr, null, null);
 307             replacedSynonymName.addReplacedSynonym(taxonName, null, null, null);
 308             getNameService().save(replacedSynonymName);
 309         }
 310
 311         Reference sec = state.getConfig().getSecReference();
 312         Taxon taxon = Taxon.NewInstance(taxonName, sec);
 313
 314         // Basionym
 315         if(fullBasionymStr != null){
 316             fullBasionymStr = fullBasionymStr.replaceAll("^\\w*:\\s", ""); // Strip off the leading 'Basionym: "
 317             basionymNameStr = basionymNameStr.replaceAll("^\\w*:\\s", ""); // Strip off the leading 'Basionym: "
 318             BotanicalName basionym = makeBotanicalName(state, regNumber, fullBasionymStr, basionymNameStr, null, null);
 319             getNameService().save(basionym);
 320             taxonName.addBasionym(basionym);
 321
 322             Synonym syn = Synonym.NewInstance(basionym, sec);
 323             taxon.addSynonym(syn, SynonymRelationshipType.HOMOTYPIC_SYNONYM_OF());
 324             getTaxonService().save(syn);
 325         }
 326
 327         // Markers
 328         if(isFossil){
 329             taxon.addMarker(Marker.NewInstance(markerTypeFossil(), true));
 330         }
 331
 332         // Types
 333         if(!StringUtils.isEmpty(typeStr)){
 334
 335             if(taxonName.getRank().isSpecies() || taxonName.getRank().isLower(Rank.SPECIES())) {
 336                 makeSpecimenTypeData(typeStr, taxonName, regNumber, state);
 337             } else {
 338                 makeNameTypeData(typeStr, taxonName, regNumber, state);
 339             }
 340         }
 341
 342         getTaxonService().save(taxon);
 343         if(higherTaxonNode != null){
 344             higherTaxonNode.addChildTaxon(taxon, null, null);
 345             getTaxonNodeService().save(higherTaxonNode);
 346         }
 347
 348         return taxon;
 349     }
 350
 351     private void makeSpecimenTypeData(String typeStr, BotanicalName taxonName, String regNumber, SimpleExcelTaxonImportState<CONFIG> state) {
 352
 353         Matcher m = typeSpecimenSplitPattern.matcher(typeStr);
 354
 355         if(m.matches()){
 356             String fieldUnitStr = m.group(TypesName.fieldUnit.name());
 357             // boolean isFieldUnit = typeStr.matches(".*([°']|\\d+\\s?m\\s|\\d+\\s?km\\s).*"); // check for location or unit m, km // makes no sense!!!!
 358             FieldUnit fieldUnit = parseFieldUnit(fieldUnitStr, regNumber, state);
 359             if(fieldUnit == null) {
 360                 // create a field unit with only a titleCache using the fieldUnitStr substring
 361                 logger.warn(csvReportLine(regNumber, "Type: fieldUnitStr can not be parsed", fieldUnitStr));
 362                 fieldUnit = FieldUnit.NewInstance();
 363                 fieldUnit.setTitleCache(fieldUnitStr, true);
 364                 getOccurrenceService().save(fieldUnit);
 365             }
 366             getOccurrenceService().save(fieldUnit);
 367
 368             // all others ..
 369             addSpecimenTypes(taxonName, fieldUnit, m.group(TypesName.holotype.name()), TypesName.holotype, false, regNumber);
 370             addSpecimenTypes(taxonName, fieldUnit, m.group(TypesName.isotype.name()), TypesName.isotype, true, regNumber);
 371
 372         } else {
 373             // create a field unit with only a titleCache using the full typeStr
 374             FieldUnit fieldUnit = FieldUnit.NewInstance();
 375             fieldUnit.setTitleCache(typeStr, true);
 376             getOccurrenceService().save(fieldUnit);
 377             logger.warn(csvReportLine(regNumber, "Type: field 'Type' can not be parsed", typeStr));
 378         }
 379         getNameService().save(taxonName);
 380     }
 381
 382     private void makeNameTypeData(String typeStr, BotanicalName taxonName, String regNumber, SimpleExcelTaxonImportState<CONFIG> state) {
 383
 384         String nameStr = typeStr.replaceAll("^Type\\s?\\:\\s?", "");
 385         if(nameStr.isEmpty()) {
 386             return;
 387         }
 388
 389         String basionymNameStr = null;
 390         String noteStr = null;
 391         String agentStr = null;
 392
 393         Matcher m;
 394
 395         if(typeStr.startsWith("not to be indicated")){
 396             // Special case:
 397             // Type: not to be indicated (Art. H.9.1. Tokyo Code); stated parent genera: Hechtia Klotzsch; Deuterocohnia Mez
 398             // FIXME
 399             m = typeNameSpecialSplitPattern.matcher(nameStr);
 400             if(m.matches()){
 401                 nameStr = m.group("name");
 402                 noteStr = m.group("note");
 403                 agentStr = m.group("agent");
 404                 // TODO better import of agent?
 405                 if(agentStr != null){
 406                     noteStr = noteStr + ": " + agentStr;
 407                 }
 408             }
 409         } else {
 410             // Generic case
 411             m = typeNameBasionymPattern.matcher(nameStr);
 412             if (m.find()) {
 413                 basionymNameStr = m.group("basionymName");
 414                 if (basionymNameStr != null) {
 415                     nameStr = nameStr.replace(m.group(0), "");
 416                 }
 417             }
 418
 419             m = typeNameNotePattern.matcher(nameStr);
 420             if (m.find()) {
 421                 noteStr = m.group(1);
 422                 if (noteStr != null) {
 423                     nameStr = nameStr.replace(m.group(0), "");
 424                 }
 425             }
 426         }
 427
 428         BotanicalName typeName = (BotanicalName) nameParser.parseFullName(nameStr, NomenclaturalCode.ICNAFP, null);
 429
 430         if(typeName.isProtectedTitleCache() || typeName.getNomenclaturalReference() != null && typeName.getNomenclaturalReference().isProtectedTitleCache()) {
 431             logger.warn(csvReportLine(regNumber, "NameType not parsable", typeStr, nameStr));
 432         }
 433
 434         if(basionymNameStr != null){
 435             BotanicalName basionymName = (BotanicalName) nameParser.parseFullName(nameStr, NomenclaturalCode.ICNAFP, null);
 436             getNameService().save(basionymName);
 437             typeName.addBasionym(basionymName);
 438         }
 439
 440
 441         NameTypeDesignation nameTypeDesignation = NameTypeDesignation.NewInstance();
 442         nameTypeDesignation.setTypeName(typeName);
 443         getNameService().save(typeName);
 444
 445         if(noteStr != null){
 446             nameTypeDesignation.addAnnotation(Annotation.NewInstance(noteStr, AnnotationType.EDITORIAL(), Language.UNKNOWN_LANGUAGE()));
 447         }
 448         taxonName.addNameTypeDesignation(typeName, null, null, null, null, false);
 449
 450     }
 451
 452     /**
 453      * Currently only parses the collector, fieldNumber and the collection date.
 454      *
 455      * @param fieldUnitStr
 456      * @param regNumber
 457      * @param state
 458      * @return null if the fieldUnitStr could not be parsed
 459      */
 460     private FieldUnit parseFieldUnit(String fieldUnitStr, String regNumber, SimpleExcelTaxonImportState<CONFIG> state) {
 461
 462         FieldUnit fieldUnit = null;
 463
 464         Matcher m1 = collectorPattern.matcher(fieldUnitStr);
 465         if(m1.matches()){
 466
 467             String collectorData = m1.group(2); // like (leg. Metzeltin, 30. 9. 1996)
 468             String removal = m1.group(1);
 469             if(collectorData == null){
 470                 collectorData = m1.group(4); // like leg. Metzeltin, 30. 9. 1996
 471                 removal = m1.group(3);
 472             }
 473             if(collectorData == null){
 474                 return null;
 475             }
 476
 477             // the fieldUnitStr is parsable
 478             // remove all collectorData from the fieldUnitStr and use the rest as locality
 479             String locality = fieldUnitStr.replace(removal, "");
 480
 481             String collectorStr = null;
 482             String detailStr = null;
 483             Partial date = null;
 484             String fieldNumber = null;
 485
 486             Matcher m2 = collectionDataPattern.matcher(collectorData);
 487             if(m2.matches()){
 488                 collectorStr = m2.group("collector");
 489                 detailStr = m2.group("detail");
 490
 491                 // Try to make sense of the detailStr
 492                 if(detailStr != null){
 493                     detailStr = detailStr.trim();
 494                     // 1. try to parse as date
 495                     date = parseDate(regNumber, detailStr);
 496                     if(date == null){
 497                         // 2. try to parse as number
 498                         if(collectorsNumber.matcher(detailStr).matches()){
 499                             fieldNumber = detailStr;
 500                         }
 501                     }
 502                 }
 503                 if(date == null && fieldNumber == null){
 504                     // detailed parsing not possible, so need fo fallback
 505                     collectorStr = collectorData;
 506                 }
 507             }
 508
 509             if(collectorStr == null) {
 510                 collectorStr = collectorData;
 511             }
 512
 513             fieldUnit = FieldUnit.NewInstance();
 514             GatheringEvent ge = GatheringEvent.NewInstance();
 515             ge.setLocality(LanguageString.NewInstance(locality, Language.UNKNOWN_LANGUAGE()));
 516
 517             TeamOrPersonBase agent =  state.getAgentBase(collectorStr);
 518             if(agent == null) {
 519                 agent = Person.NewTitledInstance(collectorStr);
 520                 getAgentService().save(agent);
 521                 state.putAgentBase(collectorStr, agent);
 522             }
 523             ge.setCollector(agent);
 524
 525             if(date != null){
 526                 ge.setGatheringDate(date);
 527             }
 528
 529             getEventBaseService().save(ge);
 530             fieldUnit.setGatheringEvent(ge);
 531
 532             if(fieldNumber != null) {
 533                 fieldUnit.setFieldNumber(fieldNumber);
 534             }
 535             getOccurrenceService().save(fieldUnit);
 536
 537         }
 538
 539         return fieldUnit;
 540     }
 541
 542     private Partial parseDate(String regNumber, String dateStr) {
 543
 544         Partial pupDate = null;
 545         boolean parseError = false;
 546
 547         String day = null;
 548         String month = null;
 549         String monthName = null;
 550         String year = null;
 551
 552         for(Pattern p : datePatterns){
 553             Matcher m2 = p.matcher(dateStr);
 554             if(m2.matches()){
 555                 try {
 556                     year = m2.group("year");
 557                 } catch (IllegalArgumentException e){
 558                     // named capture group not found
 559                 }
 560                 try {
 561                     month = m2.group("month");
 562                 } catch (IllegalArgumentException e){
 563                     // named capture group not found
 564                 }
 565
 566                 try {
 567                     monthName = m2.group("monthName");
 568                     month = monthFromName(monthName, regNumber);
 569                     if(month == null){
 570                         parseError = true;
 571                     }
 572                 } catch (IllegalArgumentException e){
 573                     // named capture group not found
 574                 }
 575                 try {
 576                     day = m2.group("day");
 577                 } catch (IllegalArgumentException e){
 578                     // named capture group not found
 579                 }
 580
 581                 if(year != null){
 582                     if (year.length() == 2) {
 583                         // it is an abbreviated year from the 19** years
 584                         year = "19" + year;
 585                     }
 586                     break;
 587                 } else {
 588                     parseError = true;
 589                 }
 590             }
 591         }
 592         if(year == null){
 593             parseError = true;
 594         }
 595         List<DateTimeFieldType> types = new ArrayList<>();
 596         List<Integer> values = new ArrayList<>();
 597         if(!parseError) {
 598             types.add(DateTimeFieldType.year());
 599             values.add(Integer.parseInt(year));
 600             if (month != null) {
 601                 types.add(DateTimeFieldType.monthOfYear());
 602                 values.add(Integer.parseInt(month));
 603             }
 604             if (day != null) {
 605                 types.add(DateTimeFieldType.dayOfMonth());
 606                 values.add(Integer.parseInt(day));
 607             }
 608             pupDate = new Partial(types.toArray(new DateTimeFieldType[types.size()]), ArrayUtils.toPrimitive(values.toArray(new Integer[values.size()])));
 609         }
 610         return pupDate;
 611     }
 612
 613     private String monthFromName(String monthName, String regNumber) {
 614
 615         Integer month = monthFromNameMap.get(monthName.toLowerCase());
 616         if(month == null){
 617             logger.warn(csvReportLine(regNumber, "Unknown month name", monthName));
 618             return null;
 619         } else {
 620             return month.toString();
 621         }
 622     }
 623
 624
 625     private void addSpecimenTypes(BotanicalName taxonName, FieldUnit fieldUnit, String typeStr, TypesName typeName, boolean multiple, String regNumber){
 626
 627         if(StringUtils.isEmpty(typeStr)){
 628             return;
 629         }
 630         typeStr = typeStr.trim().replaceAll("\\.$", "");
 631
 632         Collection collection = null;
 633         DerivedUnit specimen = null;
 634
 635         List<DerivedUnit> specimens = new ArrayList<>();
 636         if(multiple){
 637             String[] tokens = typeStr.split("\\s?,\\s?");
 638             for (String t : tokens) {
 639                 // command to  list all complex parsabel types:
 640                 // csvcut -t -c RegistrationNo_Pk,Type iapt.csv | csvgrep -c Type -m "Holotype" | egrep -o 'Holotype:\s([A-Z]*\s)[^.]*?'
 641                 // csvcut -t -c RegistrationNo_Pk,Type iapt.csv | csvgrep -c Type -m "Holotype" | egrep -o 'Isotype[^:]*:\s([A-Z]*\s)[^.]*?'
 642
 643                 if(!t.isEmpty()){
 644                     // trying to parse the string
 645                     specimen = parseSpecimenType(fieldUnit, typeName, collection, t, regNumber);
 646                     if(specimen != null){
 647                         specimens.add(specimen);
 648                     } else {
 649                         // parsing was not successful make simple specimen
 650                         specimens.add(makeSpecimenType(fieldUnit, t));
 651                     }
 652                 }
 653             }
 654         } else {
 655             specimen = parseSpecimenType(fieldUnit, typeName, collection, typeStr, regNumber);
 656             if(specimen != null) {
 657                 specimens.add(specimen);
 658                 // remember current collection
 659                 collection = specimen.getCollection();
 660             } else {
 661                 // parsing was not successful make simple specimen
 662                 specimens.add(makeSpecimenType(fieldUnit, typeStr));
 663             }
 664         }
 665
 666         for(DerivedUnit s : specimens){
 667             taxonName.addSpecimenTypeDesignation(s, typeName.status(), null, null, null, false, true);
 668        }
 669     }
 670
 671     private DerivedUnit makeSpecimenType(FieldUnit fieldUnit, String titleCache) {
 672         DerivedUnit specimen;DerivedUnitFacade facade = DerivedUnitFacade.NewInstance(SpecimenOrObservationType.PreservedSpecimen, fieldUnit);
 673         facade.setTitleCache(titleCache.trim(), true);
 674         specimen = facade.innerDerivedUnit();
 675         return specimen;
 676     }
 677
 678     /**
 679      *
 680      * @param fieldUnit
 681      * @param typeName
 682      * @param collection
 683      * @param text
 684      * @param regNumber
 685      * @return
 686      */
 687     private DerivedUnit parseSpecimenType(FieldUnit fieldUnit, TypesName typeName, Collection collection, String text, String regNumber) {
 688
 689         DerivedUnit specimen = null;
 690
 691         String collectionCode = null;
 692         String subCollectionStr = null;
 693         String instituteStr = null;
 694         String accessionNumber = null;
 695
 696         boolean unusualAccessionNumber = false;
 697
 698         text = text.trim();
 699
 700         // 1.  For Isotypes often the accession number is noted alone if the
 701         //     preceeding entry has a collection code.
 702         if(typeName .equals(TypesName.isotype) && collection != null){
 703             Matcher m = accessionNumberOnlyPattern.matcher(text);
 704             if(m.matches()){
 705                 try {
 706                     accessionNumber = m.group("accNumber");
 707                     specimen = makeSpecimenType(fieldUnit, collection, accessionNumber);
 708                 } catch (IllegalArgumentException e){
 709                     // match group acc_number not found
 710                 }
 711             }
 712         }
 713
 714         //2. try it the 'normal' way
 715         if(specimen == null) {
 716             for (Pattern p : specimenTypePatterns) {
 717                 Matcher m = p.matcher(text);
 718                 if (m.matches()) {
 719                     // collection code is mandatory
 720                     try {
 721                         collectionCode = m.group("colCode");
 722                     } catch (IllegalArgumentException e){
 723                         // match group colCode not found
 724                     }
 725                     try {
 726                         subCollectionStr = m.group("subCollection");
 727                     } catch (IllegalArgumentException e){
 728                         // match group subCollection not found
 729                     }
 730                     try {
 731                         instituteStr = m.group("institute");
 732                     } catch (IllegalArgumentException e){
 733                         // match group col_name not found
 734                     }
 735                     try {
 736                         accessionNumber = m.group("accNumber");
 737
 738                         // try to improve the accessionNumber
 739                         if(accessionNumber!= null) {
 740                             accessionNumber = accessionNumber.trim();
 741                             Matcher m2 = accessionNumberOnlyPattern.matcher(accessionNumber);
 742                             String betterAccessionNumber = null;
 743                             if (m2.matches()) {
 744                                 try {
 745                                     betterAccessionNumber = m.group("accNumber");
 746                                 } catch (IllegalArgumentException e) {
 747                                     // match group acc_number not found
 748                                 }
 749                             }
 750                             if (betterAccessionNumber != null) {
 751                                 accessionNumber = betterAccessionNumber;
 752                             } else {
 753                                 unusualAccessionNumber = true;
 754                             }
 755                         }
 756
 757                     } catch (IllegalArgumentException e){
 758                         // match group acc_number not found
 759                     }
 760
 761                     if(collectionCode == null && instituteStr == null){
 762                         logger.warn(csvReportLine(regNumber, "Type: neither 'collectionCode' nor 'institute' found in ", text));
 763                         continue;
 764                     }
 765                     collection = getCollection(collectionCode, instituteStr, subCollectionStr);
 766                     specimen = makeSpecimenType(fieldUnit, collection, accessionNumber);
 767                     break;
 768                 }
 769             }
 770         }
 771         if(specimen == null) {
 772             logger.warn(csvReportLine(regNumber, "Type: Could not parse specimen", typeName.name().toString(), text));
 773         }
 774         if(unusualAccessionNumber){
 775             logger.warn(csvReportLine(regNumber, "Type: Unusual accession number", typeName.name().toString(), text, accessionNumber));
 776         }
 777         return specimen;
 778     }
 779
 780     private DerivedUnit makeSpecimenType(FieldUnit fieldUnit, Collection collection, String accessionNumber) {
 781
 782         DerivedUnitFacade facade = DerivedUnitFacade.NewInstance(SpecimenOrObservationType.PreservedSpecimen, fieldUnit);
 783         facade.setCollection(collection);
 784         if(accessionNumber != null){
 785             facade.setAccessionNumber(accessionNumber);
 786         }
 787         return facade.innerDerivedUnit();
 788     }
 789
 790     private BotanicalName makeBotanicalName(SimpleExcelTaxonImportState<CONFIG> state, String regNumber, String titleCacheStr, String nameStr,
 791                                             String authorStr, String nomRefTitle) {
 792
 793         BotanicalName taxonName;// cache field for the taxonName.titleCache
 794         String taxonNameTitleCache = null;
 795         Map<String, AnnotationType> nameAnnotations = new HashMap<>();
 796
 797         // TitleCache preprocessing
 798         if(titleCacheStr.endsWith(ANNOTATION_MARKER_STRING) || (authorStr != null && authorStr.endsWith(ANNOTATION_MARKER_STRING))){
 799             nameAnnotations.put("Author abbreviation not checked.", AnnotationType.EDITORIAL());
 800             titleCacheStr = titleCacheStr.replace(ANNOTATION_MARKER_STRING, "").trim();
 801             if(authorStr != null) {
 802                 authorStr = authorStr.replace(ANNOTATION_MARKER_STRING, "").trim();
 803             }
 804         }
 805
 806         // parse the full taxon name
 807         if(!StringUtils.isEmpty(nomRefTitle)){
 808             String referenceSeparator = nomRefTitle.startsWith("in ") ? " " : ", ";
 809             String taxonFullNameStr = titleCacheStr + referenceSeparator + nomRefTitle;
 810             logger.debug(":::::" + taxonFullNameStr);
 811             taxonName = (BotanicalName) nameParser.parseReferencedName(taxonFullNameStr, NomenclaturalCode.ICNAFP, null);
 812         } else {
 813             taxonName = (BotanicalName) nameParser.parseFullName(titleCacheStr, NomenclaturalCode.ICNAFP, null);
 814         }
 815
 816         taxonNameTitleCache = taxonName.getTitleCache().trim();
 817         if (taxonName.isProtectedTitleCache()) {
 818             logger.warn(csvReportLine(regNumber, "Name could not be parsed", titleCacheStr));
 819         } else {
 820
 821             boolean doRestoreTitleCacheStr = false;
 822
 823             // Check if titleCache and nameCache are plausible
 824             String titleCacheCompareStr = titleCacheStr;
 825             String nameCache = taxonName.getNameCache();
 826             String nameCompareStr = nameStr;
 827             if(taxonName.isBinomHybrid()){
 828                 titleCacheCompareStr = titleCacheCompareStr.replace(" x ", " ×");
 829                 nameCompareStr = nameCompareStr.replace(" x ", " ×");
 830             }
 831             if(taxonName.isMonomHybrid()){
 832                 titleCacheCompareStr = titleCacheCompareStr.replaceAll("^X ", "× ");
 833                 nameCompareStr = nameCompareStr.replace("^X ", "× ");
 834             }
 835             if(authorStr != null && authorStr.contains(" et ")){
 836                 titleCacheCompareStr = titleCacheCompareStr.replaceAll(" et ", " & ");
 837             }
 838             if (!taxonNameTitleCache.equals(titleCacheCompareStr)) {
 839                 logger.warn(csvReportLine(regNumber, "The generated titleCache differs from the imported string", taxonNameTitleCache, " != ", titleCacheStr, " ==> original titleCacheStr has been restored"));
 840                 doRestoreTitleCacheStr = true;
 841             }
 842             if (!nameCache.trim().equals(nameCompareStr)) {
 843                 logger.warn(csvReportLine(regNumber, "The parsed nameCache differs from field '" + NAMESTRING + "'", nameCache, " != ", nameCompareStr));
 844             }
 845
 846             //  Author
 847             //nameParser.handleAuthors(taxonName, titleCacheStr, authorStr);
 848             //if (!titleCacheStr.equals(taxonName.getTitleCache())) {
 849             //    logger.warn(regNumber + ": titleCache has changed after setting authors, will restore original titleCacheStr");
 850             //    doRestoreTitleCacheStr = true;
 851             //}
 852
 853             if(doRestoreTitleCacheStr){
 854                 taxonName.setTitleCache(titleCacheStr, true);
 855             }
 856
 857             // deduplicate
 858             replaceAuthorNamesAndNomRef(state, taxonName);
 859         }
 860
 861         // Annotations
 862         if(!nameAnnotations.isEmpty()){
 863             for(String text : nameAnnotations.keySet()){
 864                 taxonName.addAnnotation(Annotation.NewInstance(text, nameAnnotations.get(text), Language.DEFAULT()));
 865             }
 866             getNameService().save(taxonName);
 867         }
 868         return taxonName;
 869     }
 870
 871     /**
 872      * @param state
 873      * @return
 874      */
 875     private TaxonNode getClassificationRootNode(IAPTImportState state) {
 876
 877      //   Classification classification = state.getClassification();
 878      //   if (classification == null){
 879      //       IAPTImportConfigurator config = state.getConfig();
 880      //       classification = Classification.NewInstance(state.getConfig().getClassificationName());
 881      //       classification.setUuid(config.getClassificationUuid());
 882      //       classification.setReference(config.getSecReference());
 883      //       classification = getClassificationService().find(state.getConfig().getClassificationUuid());
 884      //   }
 885         TaxonNode rootNode = state.getRootNode();
 886         if (rootNode == null){
 887             rootNode = getTaxonNodeService().find(ROOT_UUID);
 888         }
 889         if (rootNode == null){
 890             Classification classification = state.getClassification();
 891             if (classification == null){
 892                 Reference sec = state.getSecReference();
 893                 String classificationName = state.getConfig().getClassificationName();
 894                 Language language = Language.DEFAULT();
 895                 classification = Classification.NewInstance(classificationName, sec, language);
 896                 state.setClassification(classification);
 897                 classification.setUuid(state.getConfig().getClassificationUuid());
 898                 classification.getRootNode().setUuid(ROOT_UUID);
 899                 getClassificationService().save(classification);
 900             }
 901             rootNode = classification.getRootNode();
 902             state.setRootNode(rootNode);
 903         }
 904         return rootNode;
 905     }
 906
 907     private Collection getCollection(String collectionCode, String instituteStr, String subCollectionStr){
 908
 909         Collection superCollection = null;
 910         if(subCollectionStr != null){
 911             superCollection = getCollection(collectionCode, instituteStr, null);
 912             collectionCode = subCollectionStr;
 913             instituteStr = null;
 914         }
 915
 916         final String key = collectionCode + "-#i:" + StringUtils.defaultString(instituteStr);
 917
 918         Collection collection = collectionMap.get(key);
 919
 920         if(collection == null) {
 921             collection = Collection.NewInstance();
 922             collection.setCode(collectionCode);
 923             if(instituteStr != null){
 924                 collection.setInstitute(Institution.NewNamedInstance(instituteStr));
 925             }
 926             if(superCollection != null){
 927                 collection.setSuperCollection(superCollection);
 928             }
 929             collectionMap.put(key, collection);
 930             getCollectionService().save(collection);
 931         }
 932
 933         return collection;
 934     }
 935
 936
 937     /**
 938      * @param record
 939      * @param originalKey
 940      * @param doUnescapeHtmlEntities
 941      * @return
 942      */
 943     private String getValue(HashMap<String, String> record, String originalKey, boolean doUnescapeHtmlEntities) {
 944         String value = record.get(originalKey);
 945
 946         value = fixCharacters(value);
 947
 948         if (! StringUtils.isBlank(value)) {
 949                 if (logger.isDebugEnabled()) {
 950                     logger.debug(originalKey + ": " + value);
 951                 }
 952                 value = CdmUtils.removeDuplicateWhitespace(value.trim()).toString();
 953             if(doUnescapeHtmlEntities){
 954                 value = StringEscapeUtils.unescapeHtml(value);
 955             }
 956                 return value.trim();
 957         }else{
 958                 return null;
 959         }
 960     }
 961
 962     /**
 963      * Fixes broken characters.
 964      * For details see
 965      * http://dev.e-taxonomy.eu/redmine/issues/6035
 966      *
 967      * @param value
 968      * @return
 969      */
 970     private String fixCharacters(String value) {
 971
 972         value = StringUtils.replace(value, "s$K", "š");
 973         value = StringUtils.replace(value, "n$K", "ň");
 974         value = StringUtils.replace(value, "e$K", "ě");
 975         value = StringUtils.replace(value, "r$K", "ř");
 976         value = StringUtils.replace(value, "c$K", "č");
 977         value = StringUtils.replace(value, "z$K", "ž");
 978         value = StringUtils.replace(value, "S>U$K", "Š");
 979         value = StringUtils.replace(value, "C>U$K", "Č");
 980         value = StringUtils.replace(value, "R>U$K", "Ř");
 981         value = StringUtils.replace(value, "Z>U$K", "Ž");
 982         value = StringUtils.replace(value, "g$K", "ǧ");
 983         value = StringUtils.replace(value, "s$A", "ś");
 984         value = StringUtils.replace(value, "n$A", "ń");
 985         value = StringUtils.replace(value, "c$A", "ć");
 986         value = StringUtils.replace(value, "e$E", "ę");
 987         value = StringUtils.replace(value, "o$H", "õ");
 988         value = StringUtils.replace(value, "s$C", "ş");
 989         value = StringUtils.replace(value, "t$C", "ț");
 990         value = StringUtils.replace(value, "S>U$C", "Ş");
 991         value = StringUtils.replace(value, "a$O", "å");
 992         value = StringUtils.replace(value, "A>U$O", "Å");
 993         value = StringUtils.replace(value, "u$O", "ů");
 994         value = StringUtils.replace(value, "g$B", "ğ");
 995         value = StringUtils.replace(value, "g$B", "ĕ");
 996         value = StringUtils.replace(value, "a$B", "ă");
 997         value = StringUtils.replace(value, "l$/", "ł");
 998         value = StringUtils.replace(value, ">i", "ı");
 999         value = StringUtils.replace(value, "i$U", "ï");
1000         // Special-cases
1001         value = StringUtils.replace(value, "&yacute", "ý");
1002         value = StringUtils.replace(value, ">L", "Ł"); // corrected rule
1003         value = StringUtils.replace(value, "E>U$D", "З");
1004         value = StringUtils.replace(value, "S>U$E", "Ş");
1005         value = StringUtils.replace(value, "s$E", "ş");
1006
1007         value = StringUtils.replace(value, "c$k", "č");
1008         value = StringUtils.replace(value, " U$K", " Š");
1009
1010         value = StringUtils.replace(value, "B.O>U>!", "Ø");
1011         value = StringUtils.replace(value, "S$K", "Ŝ");
1012         value = StringUtils.replace(value, "§B>i", "ğ");
1013
1014
1015         return value;
1016     }
1017
1018
1019     /**
1020          *  Stores taxa records in DB
1021          */
1022         @Override
1023     protected void firstPass(SimpleExcelTaxonImportState<CONFIG> state) {
1024
1025         String lineNumber = "L#" + state.getCurrentLine() + ": ";
1026         logger.setLevel(Level.DEBUG);
1027         HashMap<String, String> record = state.getOriginalRecord();
1028         logger.debug(lineNumber + record.toString());
1029
1030         Set<String> keys = record.keySet();
1031         for (String key: keys) {
1032             if (! expectedKeys.contains(key)){
1033                 logger.warn(lineNumber + "Unexpected Key: " + key);
1034             }
1035         }
1036
1037         String reg_id = record.get(REGISTRATIONNO_PK);
1038
1039         //higherTaxon
1040         String higherTaxaString = record.get(HIGHERTAXON);
1041         boolean isFossil = false;
1042         if(higherTaxaString.startsWith("FOSSIL ")){
1043             higherTaxaString = higherTaxaString.replace("FOSSIL ", "");
1044             isFossil = true;
1045         }
1046         TaxonNode higherTaxon = getHigherTaxon(higherTaxaString, (IAPTImportState)state);
1047
1048        //Taxon
1049         Taxon taxon = makeTaxon(record, state, higherTaxon, isFossil);
1050         if (taxon == null){
1051             logger.warn(lineNumber + "taxon could not be created and is null");
1052             return;
1053         }
1054         ((IAPTImportState)state).setCurrentTaxon(taxon);
1055
1056
1057                 return;
1058     }
1059
1060     private TaxonNode getHigherTaxon(String higherTaxaString, IAPTImportState state) {
1061         String[] higherTaxaNames = higherTaxaString.toLowerCase().replaceAll("[\\[\\]]", "").split(":");
1062         TaxonNode higherTaxonNode = null;
1063
1064         ITaxonTreeNode rootNode = getClassificationRootNode(state);
1065         for (String htn :  higherTaxaNames) {
1066             htn = StringUtils.capitalize(htn.trim());
1067             Taxon higherTaxon = state.getHigherTaxon(htn);
1068             if (higherTaxon != null){
1069                 higherTaxonNode = higherTaxon.getTaxonNodes().iterator().next();
1070             }else{
1071                 BotanicalName name = makeHigherTaxonName(state, htn);
1072                 Reference sec = state.getSecReference();
1073                 higherTaxon = Taxon.NewInstance(name, sec);
1074                 getTaxonService().save(higherTaxon);
1075                 higherTaxonNode = rootNode.addChildTaxon(higherTaxon, sec, null);
1076                 state.putHigherTaxon(htn, higherTaxon);
1077                 getClassificationService().saveTreeNode(higherTaxonNode);
1078             }
1079             rootNode = higherTaxonNode;
1080         }
1081         return higherTaxonNode;
1082     }
1083
1084     private BotanicalName makeHigherTaxonName(IAPTImportState state, String name) {
1085
1086         Rank rank = guessRank(name);
1087
1088         BotanicalName taxonName = BotanicalName.NewInstance(rank);
1089         taxonName.addSource(makeOriginalSource(state));
1090         taxonName.setGenusOrUninomial(StringUtils.capitalize(name));
1091         return taxonName;
1092     }
1093
1094     private Rank guessRank(String name) {
1095
1096         // normalize
1097         name = name.replaceAll("\\(.*\\)", "").trim();
1098
1099         if(name.matches("^Plantae$|^Fungi$")){
1100            return Rank.KINGDOM();
1101         } else if(name.matches("^Incertae sedis$|^No group assigned$")){
1102            return rankFamilyIncertisSedis();
1103         } else if(name.matches(".*phyta$|.*mycota$")){
1104            return Rank.PHYLUM();
1105         } else if(name.matches(".*phytina$|.*mycotina$")){
1106            return Rank.SUBPHYLUM();
1107         } else if(name.matches("Gymnospermae$|.*ones$")){ // Monocotyledones, Dicotyledones
1108             return rankUnrankedSupraGeneric();
1109         } else if(name.matches(".*opsida$|.*phyceae$|.*mycetes$|.*ones$|^Musci$|^Hepaticae$")){
1110            return Rank.CLASS();
1111         } else if(name.matches(".*idae$|.*phycidae$|.*mycetidae$")){
1112            return Rank.SUBCLASS();
1113         } else if(name.matches(".*ales$")){
1114            return Rank.ORDER();
1115         } else if(name.matches(".*ineae$")){
1116            return Rank.SUBORDER();
1117         } else if(name.matches(".*aceae$")){
1118             return Rank.FAMILY();
1119         } else if(name.matches(".*oideae$")){
1120            return Rank.SUBFAMILY();
1121         } else
1122         //    if(name.matches(".*eae$")){
1123         //    return Rank.TRIBE();
1124         // } else
1125             if(name.matches(".*inae$")){
1126            return Rank.SUBTRIBE();
1127         } else if(name.matches(".*ae$")){
1128            return Rank.FAMILY();
1129         }
1130         return Rank.UNKNOWN_RANK();
1131     }
1132
1133     private Rank rankUnrankedSupraGeneric() {
1134
1135         if(rankUnrankedSupraGeneric == null){
1136             rankUnrankedSupraGeneric = Rank.NewInstance(RankClass.Suprageneric, "Unranked supra generic", " ", " ");
1137             getTermService().save(rankUnrankedSupraGeneric);
1138         }
1139         return rankUnrankedSupraGeneric;
1140     }
1141
1142     private Rank rankFamilyIncertisSedis() {
1143
1144         if(familyIncertisSedis == null){
1145             familyIncertisSedis = Rank.NewInstance(RankClass.Suprageneric, "Family incertis sedis", " ", " ");
1146             getTermService().save(familyIncertisSedis);
1147         }
1148         return familyIncertisSedis;
1149     }
1150
1151     private AnnotationType annotationTypeCaveats(){
1152         if(annotationTypeCaveats == null){
1153             annotationTypeCaveats = AnnotationType.NewInstance("Caveats", "Caveats", "");
1154             getTermService().save(annotationTypeCaveats);
1155         }
1156         return annotationTypeCaveats;
1157     }
1158
1159
1160     /**
1161      * @param state
1162      * @return
1163      */
1164     private IdentifiableSource makeOriginalSource(IAPTImportState state) {
1165         return IdentifiableSource.NewDataImportInstance("line: " + state.getCurrentLine(), null, state.getConfig().getSourceReference());
1166     }
1167
1168
1169     private Reference makeReference(IAPTImportState state, UUID uuidRef) {
1170         Reference ref = state.getReference(uuidRef);
1171         if (ref == null){
1172             ref = getReferenceService().find(uuidRef);
1173             state.putReference(uuidRef, ref);
1174         }
1175         return ref;
1176     }
1177
1178     private MarkerType markerTypeFossil(){
1179         if(this.markerTypeFossil == null){
1180             markerTypeFossil = MarkerType.NewInstance("isFossilTaxon", "isFossil", null);
1181             getTermService().save(this.markerTypeFossil);
1182         }
1183         return markerTypeFossil;
1184     }
1185
1186     private String csvReportLine(String regId, String message, String ... fields){
1187         StringBuilder out = new StringBuilder("regID#");
1188         out.append(regId).append(",\"").append(message).append('"');
1189
1190         for(String f : fields){
1191             out.append(",\"").append(f).append('"');
1192         }
1193         return out.toString();
1194     }
1195
1196
1197 }