jalviewX
File PfamFile.java
Coverage histogram
12% of files have more coverage
Classes
| Class | Line # | Actions | ||||
|---|---|---|---|---|---|---|
| PfamFile | 32 | 61 | 25 | 17 |
Contributing tests
This file is covered by 8 tests.
.
Source view
| 1 | /* | |
| 2 | * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$) | |
| 3 | * Copyright (C) $$Year-Rel$$ The Jalview Authors | |
| 4 | * | |
| 5 | * This file is part of Jalview. | |
| 6 | * | |
| 7 | * Jalview is free software: you can redistribute it and/or | |
| 8 | * modify it under the terms of the GNU General Public License | |
| 9 | * as published by the Free Software Foundation, either version 3 | |
| 10 | * of the License, or (at your option) any later version. | |
| 11 | * | |
| 12 | * Jalview is distributed in the hope that it will be useful, but | |
| 13 | * WITHOUT ANY WARRANTY; without even the implied warranty | |
| 14 | * of MERCHANTABILITY or FITNESS FOR A PARTICULAR | |
| 15 | * PURPOSE. See the GNU General Public License for more details. | |
| 16 | * | |
| 17 | * You should have received a copy of the GNU General Public License | |
| 18 | * along with Jalview. If not, see <http://www.gnu.org/licenses/>. | |
| 19 | * The Jalview Authors are detailed in the 'AUTHORS' file. | |
| 20 | */ | |
| 21 | package jalview.io; | |
| 22 | ||
| 23 | import jalview.datamodel.Sequence; | |
| 24 | import jalview.datamodel.SequenceI; | |
| 25 | import jalview.util.Format; | |
| 26 | import jalview.util.MessageManager; | |
| 27 | ||
| 28 | import java.io.IOException; | |
| 29 | import java.util.ArrayList; | |
| 30 | import java.util.HashMap; | |
| 31 | ||
| 32 | public class PfamFile extends AlignFile | |
| 33 | { | |
| 34 | ||
| 35 | 1 |  public PfamFile() |
| 36 | { | |
| 37 | } | |
| 38 | ||
| 39 | 0 | public PfamFile(String inFile, DataSourceType sourceType) |
| 40 | throws IOException | |
| 41 | { | |
| 42 | 0 | super(inFile, sourceType); |
| 43 | } | |
| 44 | ||
| 45 | 9 | public PfamFile(FileParse source) throws IOException |
| 46 | { | |
| 47 | 9 | super(source); |
| 48 | } | |
| 49 | ||
| 50 | 10 | @Override |
| 51 | public void initData() | |
| 52 | { | |
| 53 | 10 | super.initData(); |
| 54 | } | |
| 55 | ||
| 56 | 9 | @Override |
| 57 | public void parse() throws IOException | |
| 58 | { | |
| 59 | 9 | int i = 0; |
| 60 | 9 | String line; |
| 61 | ||
| 62 | 9 | HashMap<String, StringBuffer> seqhash = new HashMap<String, StringBuffer>(); |
| 63 | 9 | ArrayList<String> headers = new ArrayList<String>(); |
| 64 | 9 | boolean useTabs = false; |
| 65 | 9 | int spces; |
| 66 | ? | while ((line = nextLine()) != null) |
| 67 | { | |
| 68 | 41 | if (line.indexOf("#") == 0) |
| 69 | { | |
| 70 | // skip comment lines | |
| 71 | 0 | continue; |
| 72 | } | |
| 73 | // locate first space or (if already checked), tab | |
| 74 | 41 | if (useTabs) |
| 75 | { | |
| 76 | 0 | spces = line.indexOf("\t"); |
| 77 | } | |
| 78 | else | |
| 79 | { | |
| 80 | 41 | spces = line.indexOf(" "); |
| 81 | // check to see if we ought to split on tabs instead. | |
| 82 | 41 | if (!useTabs && spces == -1) |
| 83 | { | |
| 84 | 3 | useTabs = true; |
| 85 | 3 | spces = line.indexOf("\t"); |
| 86 | } | |
| 87 | } | |
| 88 | 41 | if (spces <= 0) |
| 89 | { | |
| 90 | // no sequence data to split on | |
| 91 | 1 | continue; |
| 92 | } | |
| 93 | 40 | String id = line.substring(0, spces); |
| 94 | 40 | StringBuffer tempseq; |
| 95 | ||
| 96 | 40 | if (seqhash.containsKey(id)) |
| 97 | { | |
| 98 | 0 | tempseq = seqhash.get(id); |
| 99 | } | |
| 100 | else | |
| 101 | { | |
| 102 | 40 | tempseq = new StringBuffer(); |
| 103 | 40 | seqhash.put(id, tempseq); |
| 104 | } | |
| 105 | ||
| 106 | 40 | if (!(headers.contains(id))) |
| 107 | { | |
| 108 | 40 | headers.add(id); |
| 109 | } | |
| 110 | 40 | if (spces + 1 < line.length()) |
| 111 | { | |
| 112 | 40 | tempseq.append(line.substring(spces + 1).trim()); |
| 113 | } | |
| 114 | } | |
| 115 | ||
| 116 | 9 | this.noSeqs = headers.size(); |
| 117 | ||
| 118 | 9 | if (noSeqs < 1) |
| 119 | { | |
| 120 | 0 | throw new IOException(MessageManager |
| 121 | .getString("exception.pfam_no_sequences_found")); | |
| 122 | } | |
| 123 | ||
| 124 | 49 | for (i = 0; i < headers.size(); i++) |
| 125 | { | |
| 126 | 40 | if (seqhash.get(headers.get(i)) != null) |
| 127 | { | |
| 128 | 40 | if (maxLength < seqhash.get(headers.get(i)).toString().length()) |
| 129 | { | |
| 130 | 10 | maxLength = seqhash.get(headers.get(i)).toString().length(); |
| 131 | } | |
| 132 | ||
| 133 | 40 | Sequence newSeq = parseId(headers.get(i).toString()); |
| 134 | 40 | newSeq.setSequence( |
| 135 | seqhash.get(headers.get(i).toString()).toString()); | |
| 136 | 40 | seqs.addElement(newSeq); |
| 137 | } | |
| 138 | else | |
| 139 | { | |
| 140 | 0 | System.err.println("PFAM File reader: Can't find sequence for " |
| 141 | + headers.get(i)); | |
| 142 | } | |
| 143 | } | |
| 144 | } | |
| 145 | ||
| 146 | 1 | @Override |
| 147 | public String print(SequenceI[] s, boolean jvsuffix) | |
| 148 | { | |
| 149 | 1 | StringBuffer out = new StringBuffer(""); |
| 150 | ||
| 151 | 1 | int max = 0; |
| 152 | 1 | int maxid = 0; |
| 153 | ||
| 154 | 1 | int i = 0; |
| 155 | ||
| 156 | 16 | while ((i < s.length) && (s[i] != null)) |
| 157 | { | |
| 158 | 15 | String tmp = printId(s[i], jvsuffix); |
| 159 | ||
| 160 | 15 | max = Math.max(max, s[i].getLength()); |
| 161 | ||
| 162 | 15 | if (tmp.length() > maxid) |
| 163 | { | |
| 164 | 3 | maxid = tmp.length(); |
| 165 | } | |
| 166 | ||
| 167 | 15 | i++; |
| 168 | } | |
| 169 | ||
| 170 | 1 | if (maxid < 15) |
| 171 | { | |
| 172 | 1 | maxid = 15; |
| 173 | } | |
| 174 | ||
| 175 | 1 | int j = 0; |
| 176 | ||
| 177 | 16 | while ((j < s.length) && (s[j] != null)) |
| 178 | { | |
| 179 | 15 | out.append(new Format("%-" + maxid + "s") |
| 180 | .form(printId(s[j], jvsuffix) + " ")); | |
| 181 | ||
| 182 | 15 | out.append(s[j].getSequenceAsString()); |
| 183 | 15 | out.append(newline); |
| 184 | 15 | j++; |
| 185 | } | |
| 186 | ||
| 187 | 1 | out.append(newline); |
| 188 | ||
| 189 | 1 | return out.toString(); |
| 190 | } | |
| 191 | } |