/*\r
* Jalview - A Sequence Alignment Editor and Viewer\r
-* Copyright (C) 2005 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle\r
+* Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle\r
*\r
* This program is free software; you can redistribute it and/or\r
* modify it under the terms of the GNU General Public License\r
*/\r
package jalview.io;\r
\r
-import jalview.analysis.*;\r
-\r
import jalview.datamodel.*;\r
\r
import java.io.*;\r
import java.util.*;\r
\r
\r
-public class FastaFile extends AlignFile {\r
- public FastaFile() {\r
- }\r
-\r
- public FastaFile(String inStr) {\r
- super(inStr);\r
+/**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @author $author$\r
+ * @version $Revision$\r
+ */\r
+public class FastaFile extends AlignFile\r
+{\r
+ /**\r
+ * Length of a sequence line\r
+ */\r
+ int len = 72;\r
+\r
+ StringBuffer out;\r
+\r
+ /**\r
+ * Creates a new FastaFile object.\r
+ */\r
+ public FastaFile()\r
+ {\r
}\r
\r
- public FastaFile(String inFile, String type) throws IOException {\r
+ /**\r
+ * Creates a new FastaFile object.\r
+ *\r
+ * @param inFile DOCUMENT ME!\r
+ * @param type DOCUMENT ME!\r
+ *\r
+ * @throws IOException DOCUMENT ME!\r
+ */\r
+ public FastaFile(String inFile, String type) throws IOException\r
+ {\r
super(inFile, type);\r
}\r
\r
- public void parse() throws IOException {\r
- String id = "";\r
- StringBuffer seq = new StringBuffer();\r
- int count = 0;\r
- boolean flag = false;\r
-\r
- int sstart = 0;\r
- int send = 0;\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @throws IOException DOCUMENT ME!\r
+ */\r
+ public void parse() throws IOException\r
+ {\r
+ StringBuffer sb = new StringBuffer();\r
+ boolean firstLine = true;\r
\r
String line;\r
-\r
- while ((line = nextLine()) != null) {\r
- if (line.length() > 0) {\r
- // Do we have an id line?\r
- // JBPNote - this code needs to be standardised to EBI/whatever for the\r
- // >dbref/dbref/dbref|refid1|refid2|refid3 'human-readable' style of naming (should it really exist)\r
-\r
- if (line.substring(0, 1).equals(">")) {\r
- if (count != 0) {\r
- if (sstart != 0) {\r
- seqs.addElement(new Sequence(id,\r
- seq.toString().toUpperCase(), sstart, send));\r
- } else {\r
- seqs.addElement(new Sequence(id,\r
- seq.toString().toUpperCase(), 1,\r
- seq.length()));\r
- }\r
- }\r
-\r
- count++;\r
-\r
- StringTokenizer str = new StringTokenizer(line, " ");\r
-\r
- id = str.nextToken();\r
- id = id.substring(1);\r
-\r
- com.stevesoft.pat.Regex dbId = new com.stevesoft.pat.Regex(\r
- "[A-Za-z-]+/?[A-Za-z-]+\\|(\\w+)\\|(.+)");\r
- // JBPNote At the moment - we don't get rid of the friendly names but this\r
- // behaviour is probably wrong in the long run.\r
- if (dbId.search(id)) {\r
- String dbid = dbId.stringMatched(1);\r
- String idname = dbId.stringMatched(2);\r
- if ( (idname.length() > 0) &&\r
- (idname.indexOf("_") > -1)) {\r
- id = idname; // use the friendly name - apparently no dbid\r
- } else\r
- if (dbid.length()>1) {\r
- id = dbid; // ignore the friendly name - we lose uniprot accession ID otherwise\r
- }\r
+ Sequence seq = null;\r
+\r
+ boolean annotation = false;\r
+\r
+ while ((line = nextLine()) != null)\r
+ {\r
+ line = line.trim();\r
+ if (line.length() > 0)\r
+ {\r
+ if (line.charAt(0)=='>')\r
+ {\r
+ if (line.startsWith(">#_"))\r
+ {\r
+ if (annotation)\r
+ {\r
+ Annotation[] anots = new Annotation[sb.length()];\r
+ String anotString = sb.toString();\r
+ for (int i = 0; i < sb.length(); i++)\r
+ {\r
+ anots[i] = new Annotation(anotString.substring(i, i+1),\r
+ null,\r
+ ' ', 0);\r
+ }\r
+ AlignmentAnnotation aa = new AlignmentAnnotation(\r
+ seq.getName().substring(2), seq.getDescription(),\r
+ anots);\r
+\r
+ annotations.addElement(aa);\r
}\r
+ }\r
+ else\r
+ annotation = false;\r
\r
- if (id.indexOf("/") > 0) {\r
- StringTokenizer st = new StringTokenizer(id, "/");\r
-\r
- if (st.countTokens() == 2) {\r
- id = st.nextToken();\r
+ if (!firstLine)\r
+ {\r
+ seq.setSequence(sb.toString());\r
\r
- String tmp = st.nextToken();\r
+ if (!annotation)\r
+ seqs.addElement(seq);\r
+ }\r
\r
- st = new StringTokenizer(tmp, "-");\r
+ seq = parseId(line.substring(1));\r
+ firstLine = false;\r
\r
- if (st.countTokens() == 2) {\r
- sstart = Integer.valueOf(st.nextToken())\r
- .intValue();\r
- send = Integer.valueOf(st.nextToken()).intValue();\r
- }\r
- }\r
- }\r
+ sb = new StringBuffer();\r
\r
- seq = new StringBuffer();\r
- } else {\r
- seq = seq.append(line);\r
+ if (line.startsWith(">#_"))\r
+ annotation = true;\r
+ }\r
+ else\r
+ {\r
+ sb.append(line);\r
}\r
}\r
}\r
\r
- if (count > 0) {\r
- if (!isValidProteinSequence(seq.toString().toUpperCase())) {\r
- throw new IOException("Invalid protein sequence");\r
- }\r
-\r
- if (sstart != 0) {\r
- seqs.addElement(new Sequence(id, seq.toString().toUpperCase(),\r
- sstart, send));\r
- } else {\r
- seqs.addElement(new Sequence(id, seq.toString().toUpperCase(),\r
- 1, seq.length()));\r
- }\r
+ if (annotation)\r
+ {\r
+ Annotation[] anots = new Annotation[sb.length()];\r
+ String anotString = sb.toString();\r
+ for (int i = 0; i < sb.length(); i++)\r
+ {\r
+ anots[i] = new Annotation(anotString.substring(i, i + 1),\r
+ null,\r
+ ' ', 0);\r
+ }\r
+ AlignmentAnnotation aa = new AlignmentAnnotation(\r
+ seq.getName().substring(2), seq.getDescription(),\r
+ anots);\r
+\r
+ annotations.addElement(aa);\r
}\r
- }\r
\r
- public static String print(SequenceI[] s) {\r
- return print(s, 72);\r
- }\r
-\r
- public static String print(SequenceI[] s, int len) {\r
- return print(s, len, true);\r
- }\r
-\r
- public static String print(SequenceI[] s, int len, boolean gaps) {\r
- return print(s, len, gaps, true);\r
+ else if (!firstLine)\r
+ {\r
+ seq.setSequence(sb.toString());\r
+ seqs.addElement(seq);\r
+ }\r
}\r
\r
- public static String print(SequenceI[] s, int len, boolean gaps,\r
- boolean displayId) {\r
- StringBuffer out = new StringBuffer();\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @param s DOCUMENT ME!\r
+ * @param len DOCUMENT ME!\r
+ * @param gaps DOCUMENT ME!\r
+ * @param displayId DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ public String print(SequenceI[] s)\r
+ {\r
+ out = new StringBuffer();\r
int i = 0;\r
\r
- while ((i < s.length) && (s[i] != null)) {\r
- String seq = "";\r
+ while ((i < s.length) && (s[i] != null))\r
+ {\r
+ out.append(">" + printId(s[i]));\r
+ if(s[i].getDescription()!=null)\r
+ out.append(" "+s[i].getDescription());\r
\r
- if (gaps) {\r
- seq = s[i].getSequence();\r
- } else {\r
- seq = AlignSeq.extractGaps("-. ", s[i].getSequence());\r
- }\r
+ out.append("\n");\r
\r
- // used to always put this here: + "/" + s[i].getStart() + "-" + s[i].getEnd() +\r
- out.append(">" +\r
- ((displayId) ? s[i].getDisplayId() : s[i].getName()) + "\n");\r
+ int nochunks = (s[i].getLength() / len) + 1;\r
\r
- int nochunks = (seq.length() / len) + 1;\r
-\r
- for (int j = 0; j < nochunks; j++) {\r
+ for (int j = 0; j < nochunks; j++)\r
+ {\r
int start = j * len;\r
int end = start + len;\r
\r
- if (end < seq.length()) {\r
- out.append(seq.substring(start, end) + "\n");\r
- } else if (start < seq.length()) {\r
- out.append(seq.substring(start) + "\n");\r
+ if (end < s[i].getLength())\r
+ {\r
+ out.append(s[i].getSequenceAsString(start, end) + "\n");\r
+ }\r
+ else if (start < s[i].getLength())\r
+ {\r
+ out.append(s[i].getSequenceAsString(start, s[i].getLength()) + "\n");\r
}\r
}\r
\r
return out.toString();\r
}\r
\r
- public String print() {\r
+ /**\r
+ * DOCUMENT ME!\r
+ *\r
+ * @return DOCUMENT ME!\r
+ */\r
+ public String print()\r
+ {\r
return print(getSeqsAsArray());\r
}\r
}\r
git://source.jalview.org / jalview.git / blobdiff