*/\r
package jalview.io;\r
\r
-import jalview.analysis.*;\r
-\r
import jalview.datamodel.*;\r
\r
import java.io.*;\r
\r
-import java.util.*;\r
-\r
\r
/**\r
* DOCUMENT ME!\r
*/\r
public void parse() throws IOException\r
{\r
- String id = "";\r
- StringBuffer seq = new StringBuffer();\r
+ StringBuffer sb = new StringBuffer();\r
int count = 0;\r
\r
- int sstart = 0;\r
- int send = 0;\r
-\r
String line;\r
+ Sequence seq = null;\r
\r
while ((line = nextLine()) != null)\r
{\r
if (line.length() > 0)\r
{\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
- if (line.substring(0, 1).equals(">"))\r
+ if (line.charAt(0)=='>')\r
{\r
if (count != 0)\r
{\r
- if (sstart != 0)\r
- {\r
- seqs.addElement(new Sequence(id, seq.toString(),\r
- sstart, send));\r
- }\r
- else\r
- {\r
- seqs.addElement(new Sequence(id, seq.toString(), 1,\r
- seq.length()));\r
- }\r
+ if (!isValidProteinSequence(sb.toString()))\r
+ {\r
+ throw new IOException(AppletFormatAdapter.INVALID_CHARACTERS\r
+ +" : "+seq.getName()\r
+ +" : "+invalidCharacter);\r
+ }\r
+\r
+ seq.setSequence(sb.toString());\r
+ seqs.addElement(seq);\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
-\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
- {\r
- String dbid = dbId.stringMatched(1);\r
- String idname = dbId.stringMatched(2);\r
-\r
- if ((idname.length() > 0) &&\r
- (idname.indexOf("_") > -1))\r
- {\r
- id = idname; // use the friendly name - apparently no dbid\r
- }\r
- else if (dbid.length() > 1)\r
- {\r
- id = dbid; // ignore the friendly name - we lose uniprot accession ID otherwise\r
- }\r
- }\r
-\r
- if (id.indexOf("/") > 0)\r
- {\r
- StringTokenizer st = new StringTokenizer(id, "/");\r
-\r
- if (st.countTokens() == 2)\r
- {\r
- id = st.nextToken();\r
-\r
- String tmp = st.nextToken();\r
+ seq = parseId(line.substring(1));\r
\r
- st = new StringTokenizer(tmp, "-");\r
-\r
- if (st.countTokens() == 2)\r
- {\r
- sstart = Integer.valueOf(st.nextToken())\r
- .intValue();\r
- send = Integer.valueOf(st.nextToken()).intValue();\r
- }\r
- }\r
- }\r
-\r
- seq = new StringBuffer();\r
+ count++;\r
+ sb = new StringBuffer();\r
}\r
else\r
{\r
- seq = seq.append(line);\r
+ sb.append(line);\r
}\r
}\r
}\r
\r
if (count > 0)\r
{\r
- if (!isValidProteinSequence(seq.toString().toUpperCase()))\r
+ if (!isValidProteinSequence(sb.toString()))\r
{\r
- throw new IOException("Invalid protein sequence");\r
+ throw new IOException(AppletFormatAdapter.INVALID_CHARACTERS\r
+ +" : "+seq.getName()\r
+ +" : "+invalidCharacter);\r
}\r
\r
- if (sstart != 0)\r
- {\r
- seqs.addElement(new Sequence(id, seq.toString().toUpperCase(),\r
- sstart, send));\r
- }\r
- else\r
- {\r
- seqs.addElement(new Sequence(id, seq.toString().toUpperCase(),\r
- 1, seq.length()));\r
- }\r
+ seq.setSequence(sb.toString());\r
+ seqs.addElement(seq);\r
}\r
}\r
\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param s DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public static String print(SequenceI[] s)\r
- {\r
- return print(s, 72);\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param s DOCUMENT ME!\r
- * @param len DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public static String print(SequenceI[] s, int len)\r
- {\r
- return print(s, len, true);\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param s DOCUMENT ME!\r
- * @param len DOCUMENT ME!\r
- * @param gaps DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public static String print(SequenceI[] s, int len, boolean gaps)\r
- {\r
- return print(s, len, gaps, true);\r
- }\r
\r
/**\r
* DOCUMENT ME!\r
*\r
* @return DOCUMENT ME!\r
*/\r
- public static String print(SequenceI[] s, int len, boolean gaps,\r
- boolean displayId)\r
+ public String print(SequenceI[] s)\r
{\r
+ int len = 72;\r
StringBuffer out = new StringBuffer();\r
int i = 0;\r
\r
while ((i < s.length) && (s[i] != null))\r
{\r
- String seq = "";\r
-\r
- if (gaps)\r
- {\r
- seq = s[i].getSequence();\r
- }\r
- else\r
- {\r
- seq = AlignSeq.extractGaps("-. ", s[i].getSequence());\r
- }\r
+ out.append(">" + printId(s[i]));\r
+ if(s[i].getDescription()!=null)\r
+ out.append(" "+s[i].getDescription());\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
+ out.append("\n");\r
\r
- int nochunks = (seq.length() / len) + 1;\r
+ int nochunks = (s[i].getLength() / len) + 1;\r
\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
+ if (end < s[i].getLength())\r
{\r
- out.append(seq.substring(start, end) + "\n");\r
+ out.append(s[i].getSequence(start, end) + "\n");\r
}\r
- else if (start < seq.length())\r
+ else if (start < s[i].getLength())\r
{\r
- out.append(seq.substring(start) + "\n");\r
+ out.append(s[i].getSequence(start, s[i].getLength()) + "\n");\r
}\r
}\r
\r