1
|
/**
|
2
|
* Copyright (C) 2013 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
|
package eu.etaxonomy.cdm.ext.geo;
|
10
|
|
11
|
import java.io.File;
|
12
|
import java.io.IOException;
|
13
|
import java.io.Reader;
|
14
|
import java.util.HashMap;
|
15
|
import java.util.List;
|
16
|
import java.util.Map;
|
17
|
import java.util.Set;
|
18
|
|
19
|
import org.apache.commons.lang.ArrayUtils;
|
20
|
import org.apache.log4j.Level;
|
21
|
import org.apache.log4j.Logger;
|
22
|
|
23
|
import au.com.bytecode.opencsv.CSVReader;
|
24
|
import eu.etaxonomy.cdm.model.common.TermVocabulary;
|
25
|
import eu.etaxonomy.cdm.model.location.NamedArea;
|
26
|
|
27
|
/**
|
28
|
* Reads csv data containing the attributes from a shape file and adds the
|
29
|
* shapefile data to each area in the given set of {@link NamedAreas}. The way
|
30
|
* this data it attached to the areas is specific to the
|
31
|
* {@link IGeoServiceAreaMapping} implementation. It is recommended to create
|
32
|
* csv file directly from the original shape file by making use of the
|
33
|
* {@code org2ogr} command which is contained in the <a
|
34
|
* href="http://www.gdal.org/ogr2ogr.html">gdal</a> tools:
|
35
|
*
|
36
|
* <pre>
|
37
|
* ogr2ogr -f csv out.csv input_shape_file.shp
|
38
|
* </pre>
|
39
|
*
|
40
|
*
|
41
|
* @author a.kohlbecker
|
42
|
\* @since Oct 18, 2013
|
43
|
*
|
44
|
*/
|
45
|
public class ShpAttributesToNamedAreaMapper {
|
46
|
|
47
|
public static final Logger logger = Logger.getLogger(ShpAttributesToNamedAreaMapper.class);
|
48
|
|
49
|
private static final char COMMA = ',';
|
50
|
|
51
|
private final Set<NamedArea> areas;
|
52
|
|
53
|
private final IGeoServiceAreaMapping areaMapping;
|
54
|
|
55
|
/**
|
56
|
* maps the idSearchFields supplied to the {@link #readCsv(File, List)} method to the
|
57
|
* column index
|
58
|
*/
|
59
|
private Map<String,Integer> searchColumnMap = null;
|
60
|
|
61
|
public ShpAttributesToNamedAreaMapper(Set<NamedArea> areas, IGeoServiceAreaMapping areaMapping) {
|
62
|
this.areas = areas;
|
63
|
this.areaMapping = areaMapping;
|
64
|
}
|
65
|
|
66
|
public ShpAttributesToNamedAreaMapper(TermVocabulary<NamedArea> areaVocabulary, IGeoServiceAreaMapping areaMapping) {
|
67
|
this.areas = areaVocabulary.getTerms();
|
68
|
this.areaMapping = areaMapping;
|
69
|
}
|
70
|
|
71
|
/**
|
72
|
* @param reader A reader of the csv data
|
73
|
* @param idSearchFields
|
74
|
* An ordered list column names in the the csv file to be
|
75
|
* imported. These columns will be used to search for the
|
76
|
* {@link NamedArea#getIdInVocabulary() IdInVocabulary} of each
|
77
|
* area
|
78
|
* @param wmsLayerName
|
79
|
* @return the resulting table of the import, also together with diagnostic
|
80
|
* messages per NamedArea (id not found)
|
81
|
* @throws IOException
|
82
|
*/
|
83
|
public Map<NamedArea, String> readCsv(Reader reader, List<String> idSearchFields, String wmsLayerName) throws IOException {
|
84
|
|
85
|
logger.setLevel(Level.DEBUG);
|
86
|
|
87
|
Map<NamedArea, String> resultMap = new HashMap<>(areas.size());
|
88
|
|
89
|
CSVReader csvReader = new CSVReader(reader, COMMA);
|
90
|
|
91
|
// read header row and prepare the searchColumnMap
|
92
|
String[] headerRow = csvReader.readNext();
|
93
|
searchColumnMap = new HashMap<>();
|
94
|
for(String colName : idSearchFields){
|
95
|
int idx = ArrayUtils.indexOf(headerRow, colName);
|
96
|
if(idx > -1){
|
97
|
searchColumnMap.put(colName, idx);
|
98
|
} else {
|
99
|
logger.debug("no header row found for " + colName);
|
100
|
}
|
101
|
}
|
102
|
|
103
|
// read the rest of the file
|
104
|
List<String[]> data = csvReader.readAll();
|
105
|
csvReader.close();
|
106
|
csvReader = null; // release memory
|
107
|
|
108
|
String matchIdCode;
|
109
|
String matchColName;
|
110
|
for(NamedArea a : areas){
|
111
|
|
112
|
if(a.getIdInVocabulary() == null){
|
113
|
String message = "has no IdInVocabulary";
|
114
|
resultMap.put(a, message);
|
115
|
logger.warn(a.getTitleCache() + " " + message);
|
116
|
continue;
|
117
|
}
|
118
|
matchIdCode = null;
|
119
|
matchColName = null;
|
120
|
// search in each of the columns until found
|
121
|
for(String colName : searchColumnMap.keySet()){
|
122
|
|
123
|
int idx = searchColumnMap.get(colName);
|
124
|
|
125
|
for(String[] row : data){
|
126
|
String fieldData = row[idx].trim();
|
127
|
if(a.getIdInVocabulary().equals(fieldData)){
|
128
|
// FOUND!
|
129
|
matchIdCode = fieldData;
|
130
|
break;
|
131
|
}
|
132
|
}
|
133
|
|
134
|
if(matchIdCode != null){
|
135
|
// no need to search in next column
|
136
|
matchColName = colName;
|
137
|
break;
|
138
|
}
|
139
|
|
140
|
} // END loop columns
|
141
|
|
142
|
if(matchIdCode != null){
|
143
|
//TODO need to clear the area mapping since the mapping impl can not distinguish multiple layers
|
144
|
// see http://dev.e-taxonomy.eu/trac/ticket/4263
|
145
|
areaMapping.clear(a);
|
146
|
|
147
|
GeoServiceArea geoServiceArea;
|
148
|
geoServiceArea = areaMapping.valueOf(a);
|
149
|
if(geoServiceArea == null){
|
150
|
geoServiceArea = new GeoServiceArea();
|
151
|
}
|
152
|
geoServiceArea.add(wmsLayerName, matchColName, matchIdCode);
|
153
|
areaMapping.set(a, geoServiceArea );
|
154
|
|
155
|
String message = matchColName + ": " + matchIdCode;
|
156
|
resultMap.put(a, message);
|
157
|
if(logger.isDebugEnabled()){
|
158
|
logger.debug(a.getIdInVocabulary() + ": " + message);
|
159
|
}
|
160
|
} else {
|
161
|
String message = "no match for " + a.getIdInVocabulary();
|
162
|
resultMap.put(a, message);
|
163
|
logger.warn(message);
|
164
|
}
|
165
|
|
166
|
|
167
|
} // END of areas loop
|
168
|
return resultMap;
|
169
|
}
|
170
|
|
171
|
}
|