*/
package jalview.io;
+import jalview.analysis.Rna;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.AlignmentI;
+import jalview.datamodel.Annotation;
+import jalview.datamodel.DBRefEntry;
+import jalview.datamodel.DBRefSource;
+import jalview.datamodel.Mapping;
+import jalview.datamodel.Sequence;
+import jalview.datamodel.SequenceFeature;
+import jalview.datamodel.SequenceI;
+import jalview.schemes.ResidueProperties;
+import jalview.util.Comparison;
+import jalview.util.DBRefUtils;
+import jalview.util.Format;
+import jalview.util.MessageManager;
+import jalview.util.Platform;
+
import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import fr.orsay.lri.varna.exceptions.ExceptionUnmatchedClosingParentheses;
import fr.orsay.lri.varna.factories.RNAFactory;
import fr.orsay.lri.varna.models.rna.RNA;
-import jalview.analysis.Rna;
-import jalview.datamodel.AlignmentAnnotation;
-import jalview.datamodel.AlignmentI;
-import jalview.datamodel.Annotation;
-import jalview.datamodel.DBRefEntry;
-import jalview.datamodel.DBRefSource;
-import jalview.datamodel.Mapping;
-import jalview.datamodel.Sequence;
-import jalview.datamodel.SequenceFeature;
-import jalview.datamodel.SequenceI;
-import jalview.schemes.ResidueProperties;
-import jalview.util.Comparison;
-import jalview.util.DBRefUtils;
-import jalview.util.Format;
-import jalview.util.MessageManager;
// import org.apache.log4j.*;
private static final char UNDERSCORE = '_';
// WUSS extended symbols. Avoid ambiguity with protein SS annotations by using NOT_RNASS first.
+
public static final String RNASS_BRACKETS = "<>[](){}AaBbCcDdEeFfGgHhIiJjKkLlMmNnOoPpQqRrSsTtUuVvWwXxYyZz";
+ public static final int REGEX_STOCKHOLM = 0;
+
+ public static final int REGEX_BRACKETS = 1;
+
// use the following regex to decide an annotations (whole) line is NOT an RNA
// SS (it contains only E,H,e,h and other non-brace/non-alpha chars)
- private static final Regex NOT_RNASS = new Regex(
- "^[^<>[\\](){}A-DF-Za-df-z]*$");
+ public static final int REGEX_NOT_RNASS = 2;
+
+ private static final int REGEX_ANNOTATION = 3;
+
+ private static final int REGEX_PFAM = 4;
+
+ private static final int REGEX_RFAM = 5;
+
+ private static final int REGEX_ALIGN_END = 6;
+
+ private static final int REGEX_SPLIT_ID = 7;
+
+ private static final int REGEX_SUBTYPE = 8;
+
+ private static final int REGEX_ANNOTATION_LINE = 9;
+
+ private static final int REGEX_REMOVE_ID = 10;
+
+ private static final int REGEX_OPEN_PAREN = 11;
+
+ private static final int REGEX_CLOSE_PAREN = 12;
+
+ public static final int REGEX_MAX = 13;
+
+ private static Regex REGEX[] = new Regex[REGEX_MAX];
+
+ /**
+ * Centralize all actual Regex instantialization in Platform.
+ * // JBPNote: Why is this 'centralisation' better ?
+ * @param id
+ * @return
+ */
+ private static Regex getRegex(int id)
+ {
+ if (REGEX[id] == null)
+ {
+ String pat = null, pat2 = null;
+ switch (id)
+ {
+ case REGEX_STOCKHOLM:
+ pat = "# STOCKHOLM ([\\d\\.]+)";
+ break;
+ case REGEX_BRACKETS:
+ // for reference; not used
+ pat = "(<|>|\\[|\\]|\\(|\\)|\\{|\\})";
+ break;
+ case REGEX_NOT_RNASS:
+ pat = "^[^<>[\\](){}A-DF-Za-df-z]*$";
+ break;
+ case REGEX_ANNOTATION:
+ pat = "(\\w+)\\s*(.*)";
+ break;
+ case REGEX_PFAM:
+ pat = "PF[0-9]{5}(.*)";
+ break;
+ case REGEX_RFAM:
+ pat = "RF[0-9]{5}(.*)";
+ break;
+ case REGEX_ALIGN_END:
+ pat = "^\\s*\\/\\/";
+ break;
+ case REGEX_SPLIT_ID:
+ pat = "(\\S+)\\/(\\d+)\\-(\\d+)";
+ break;
+ case REGEX_SUBTYPE:
+ pat = "(\\S+)\\s+(\\S*)\\s+(.*)";
+ break;
+ case REGEX_ANNOTATION_LINE:
+ pat = "#=(G[FSRC]?)\\s+(.*)";
+ break;
+ case REGEX_REMOVE_ID:
+ pat = "(\\S+)\\s+(\\S+)";
+ break;
+ case REGEX_OPEN_PAREN:
+ pat = "(<|\\[)";
+ pat2 = "(";
+ break;
+ case REGEX_CLOSE_PAREN:
+ pat = "(>|\\])";
+ pat2 = ")";
+ break;
+ default:
+ return null;
+ }
+ REGEX[id] = Platform.newRegex(pat, pat2);
+ }
+ return REGEX[id];
+ }
StringBuffer out; // output buffer
// First, we have to check that this file has STOCKHOLM format, i.e. the
// first line must match
- r = new Regex("# STOCKHOLM ([\\d\\.]+)");
+ r = getRegex(REGEX_STOCKHOLM);
if (!r.search(nextLine()))
{
throw new IOException(MessageManager
// logger.debug("Stockholm version: " + version);
}
- // We define some Regexes here that will be used regularly later
- rend = new Regex("^\\s*\\/\\/"); // Find the end of an alignment
- p = new Regex("(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
+ // We define some Regexes here that will be used regularily later
+ rend = getRegex(REGEX_ALIGN_END);//"^\\s*\\/\\/"); // Find the end of an alignment
+ p = getRegex(REGEX_SPLIT_ID);//"(\\S+)\\/(\\d+)\\-(\\d+)"); // split sequence id in
// id/from/to
- s = new Regex("(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses annotation subtype
- r = new Regex("#=(G[FSRC]?)\\s+(.*)"); // Finds any annotation line
- x = new Regex("(\\S+)\\s+(\\S+)"); // split id from sequence
+ s = getRegex(REGEX_SUBTYPE);// "(\\S+)\\s+(\\S*)\\s+(.*)"); // Parses
+ // annotation subtype
+ r = getRegex(REGEX_ANNOTATION_LINE);// "#=(G[FSRC]?)\\s+(.*)"); // Finds any
+ // annotation line
+ x = getRegex(REGEX_REMOVE_ID);// "(\\S+)\\s+(\\S+)"); // split id from
+ // sequence
// Convert all bracket types to parentheses (necessary for passing to VARNA)
- Regex openparen = new Regex("(<|\\[)", "(");
- Regex closeparen = new Regex("(>|\\])", ")");
+ Regex openparen = getRegex(REGEX_OPEN_PAREN);//"(<|\\[)", "(");
+ Regex closeparen = getRegex(REGEX_CLOSE_PAREN);//"(>|\\])", ")");
// // Detect if file is RNA by looking for bracket types
-// Regex detectbrackets = new Regex("(<|>|\\[|\\]|\\(|\\))");
+ // Regex detectbrackets = getRegex("(<|>|\\[|\\]|\\(|\\))");
rend.optimize();
p.optimize();
this.noSeqs = seqs.size();
String dbsource = null;
- Regex pf = new Regex("PF[0-9]{5}(.*)"); // Finds AC for Pfam
- Regex rf = new Regex("RF[0-9]{5}(.*)"); // Finds AC for Rfam
+ Regex pf = getRegex(REGEX_PFAM); // Finds AC for Pfam
+ Regex rf = getRegex(REGEX_RFAM); // Finds AC for Rfam
if (getAlignmentProperty("AC") != null)
{
String dbType = getAlignmentProperty("AC").toString();
if (accAnnotations != null && accAnnotations.containsKey("AC"))
{
- String dbr = (String) accAnnotations.get("AC");
- if (dbr != null)
- {
- // we could get very clever here - but for now - just try to
+ String dbr = (String) accAnnotations.get("AC");
+ if (dbr != null)
+ {
+ // we could get very clever here - but for now - just try to
// guess accession type from type of sequence, source of alignment plus
// structure
- // of accession
- guessDatabaseFor(seqO, dbr, dbsource);
+ // of accession
+ guessDatabaseFor(seqO, dbr, dbsource);
}
// else - do what ? add the data anyway and prompt the user to
// specify what references these are ?
*/
// Let's save the annotations, maybe we'll be able to do something
// with them later...
- Regex an = new Regex("(\\w+)\\s*(.*)");
+ Regex an = getRegex(REGEX_ANNOTATION);
if (an.search(annContent))
{
if (an.stringMatched(1).equals("NH"))
if (type.equalsIgnoreCase("secondary structure"))
{
ss = true;
- isrnass = !NOT_RNASS.search(annots); // sorry about the double negative
+ isrnass = !getRegex(REGEX_NOT_RNASS).search(annots); // sorry about the double
+ // negative
// here (it's easier for dealing with
// other non-alpha-non-brace chars)
}
return ref.getSource().toString() + " ; "
+ ref.getAccessionId().toString();
}
+
@Override
public String print(SequenceI[] s, boolean jvSuffix)
{
}
else
{
- for (int idb = 0; idb < seq.getDBRefs().size(); idb++)
+ for (int idb = 0; idb < ndb; idb++)
{
- DBRefEntry dbref = seq.getDBRefs().get(idb);
+ DBRefEntry dbref = seqrefs.get(idb);
dataRef.put(tmp, dbref_to_ac_record(dbref));
// if we put in a uniprot or EMBL record then we're done:
- if (isAA && DBRefSource.UNIPROT
- .equals(DBRefUtils.getCanonicalName(dbref.getSource())))
- {
- break;
- }
- if (!isAA && DBRefSource.EMBL
+ if ((isAA ? DBRefSource.UNIPROT : DBRefSource.EMBL)
.equals(DBRefUtils.getCanonicalName(dbref.getSource())))
{
break;