Menu
▾
▴
[Obo-svn-commit] SF.net SVN: obo: [147] phenote/trunk
|
From: <mg...@us...> - 2006-10-03 23:11:41
|
Revision: 147
http://svn.sourceforge.net/obo/?rev=147&view=rev
Author: mgibson
Date: 2006-10-03 16:11:33 -0700 (Tue, 03 Oct 2006)
Log Message:
-----------
auto complete terms now put start with before contains (not startwith)
in the completion list
Modified Paths:
--------------
phenote/trunk/conf/initial-flybase.cfg
phenote/trunk/conf/initial-zfin.cfg
phenote/trunk/src/java/phenote/datamodel/Ontology.java
phenote/trunk/src/java/phenote/main/Phenote.java
phenote/trunk/src/java/test/phenote/gui/TestPhenote.java
Modified: phenote/trunk/conf/initial-flybase.cfg
===================================================================
--- phenote/trunk/conf/initial-flybase.cfg 2006-10-02 18:02:44 UTC (rev 146)
+++ phenote/trunk/conf/initial-flybase.cfg 2006-10-03 23:11:33 UTC (rev 147)
@@ -15,6 +15,7 @@
<postcomp relationship-ontology="relationship.obo"/>
<ontology name="Fly" file="fly_anatomy.obo" />
<ontology name="GO" file="gene_ontology.obo"/>
+ <ontology name="SP" file="spatial.obo"/>
</field>
<field name="Quality" file="quality.obo"/>
Modified: phenote/trunk/conf/initial-zfin.cfg
===================================================================
--- phenote/trunk/conf/initial-zfin.cfg 2006-10-02 18:02:44 UTC (rev 146)
+++ phenote/trunk/conf/initial-zfin.cfg 2006-10-03 23:11:33 UTC (rev 147)
@@ -16,6 +16,7 @@
<postcomp relationship-ontology="relationship.obo"/>
<ontology name="ZF" file="zebrafish_anatomy.obo" filter-out="ZFS:" />
<ontology name="GO" file="gene_ontology.obo"/>
+ <ontology name="SP" file="spatial.obo"/>
</field>
<field name="Quality" file="quality.obo"/>
Modified: phenote/trunk/src/java/phenote/datamodel/Ontology.java
===================================================================
--- phenote/trunk/src/java/phenote/datamodel/Ontology.java 2006-10-02 18:02:44 UTC (rev 146)
+++ phenote/trunk/src/java/phenote/datamodel/Ontology.java 2006-10-03 23:11:33 UTC (rev 147)
@@ -98,11 +98,13 @@
return searchTerms;
}
+ /** helper fn for getSearchTerms(String,SearhParamsI) */
private Vector<OBOClass> getSearchTerms(String input, List<OBOClass> ontologyTermList,
SearchParamsI searchParams) {
// need a unique list - UniqueTermList has quick check for uniqueness, checking
- // whole list is very very slow
- UniqueTermList uniqueTermList = new UniqueTermList();
+ // whole list is very very slow - how is it possible to get a dup term? i forget?
+ // the dup term was a BUG! in synonym - woops
+ SearchTermList uniqueTermList = new SearchTermList();
//Vector searchTerms = new Vector();
if (ontologyTermList == null)
return uniqueTermList.getVector();//searchTerms;
@@ -120,12 +122,6 @@
boolean termAdded = false;
- // SKIP PATO ATTRIBUTES - only want values - or do !contains "value"?
- // yes do contains value - as theres other crap to filter too
- // apparently curators want to see attribs as well according to Michael
- //if (filterAttributes() && isAttribute(originalTerm))
- //continue;
-
if (searchParams.searchTerms()) {
// adds originalTerm to searchTerms if match (1st if exact)
termAdded = compareAndAddTerm(input,originalTerm,oboClass,uniqueTermList);
@@ -136,14 +132,14 @@
if (searchParams.searchSynonyms()) {
Set synonyms = oboClass.getSynonyms();
- for (Iterator i = synonyms.iterator(); i.hasNext(); ) {
+ for (Iterator i = synonyms.iterator(); i.hasNext() &&!termAdded; ) {
String syn = i.next().toString();
- //System.out.println("syn "+syn+" for "+originalTerm);
+ //log().debug("syn "+syn+" for "+originalTerm);
termAdded = compareAndAddTerm(input,syn,oboClass,uniqueTermList);
- if (termAdded)
- continue;
+ //if (termAdded) continue; // woops continues this for not the outer!
}
}
+ if (termAdded) continue;
if (searchParams.searchDefinitions()) {
@@ -151,7 +147,7 @@
if (definition != null & !definition.equals(""))
termAdded = compareAndAddTerm(input,definition,oboClass,uniqueTermList);
if (termAdded)
- continue;
+ continue; // not really necesary as its last
}
}
@@ -188,19 +184,11 @@
}
}
- // pato subclass?
-// private boolean isAttribute(String term) {
-// return contains(term.toLowerCase(),"attribute");
-// }
-
-
-
-
/** User input is already lower cased, this potentially adds oboClass to
* searchTerms if input & compareTerm match. Puts it first if exact.
* for term names comp = obo, for syns comp is the syn.
Returns true if term is a match & either gets added or already is added
- * Also checks if term is in list already
+ * Also checks if term is in list already - not needed - woops!
Theres a speed issue here - the vector needs to be unique. if check every time
with whole list very very slow. 2 alternatives to try.
1) have separate hash for checking uniqueness (downside 2 data structures).
@@ -209,9 +197,14 @@
nice - its 1 data structure BUT exact matches go 1st, no way in linked hash set
to insert 1st elements so would need to keep separate list of exact matches, which
i guess there can be more than one with synonyms (do syns count for exact matches?
- should they?) - downside of #2 is need Vector for combo box */
+ should they?) - downside of #2 is need Vector for combo box
+
+ this wont fly for having different display strings for syn & obs as compareTerm
+ can be syn obs term or def and its lost here
+ i think these methods need to be moved to gui.TermSearcher that utilizes Ontology
+ but produces CompListTerms */
private boolean compareAndAddTerm(String input, String compareTerm, OBOClass oboClass,
- UniqueTermList uniqueTermList) {
+ SearchTermList searchTermList) {
String oboTerm = oboClass.getName();
@@ -220,28 +213,22 @@
if (ignoreCase)
lowerComp = compareTerm.toLowerCase();
- boolean doContains = true; // discard? param for?
+ //boolean doContains = true; // discard? param for?
// exact match goes first in list
if (lowerComp.equals(input)) {
- //if (!searchTerms.contains(oboClass)) {// this takes a long time w long lists!
- //searchTerms.add(0,oboClass);
- uniqueTermList.addTermFirst(oboClass); // adds if not present
+ searchTermList.addTermFirst(oboClass); // adds if not present
return true;
- //}
}
+ // new paradigm - put starts with first
+ else if (lowerComp.startsWith(input)) {
+ searchTermList.addTerm(oboClass);
+ return true;
+ }
// Contains
- else if (doContains) {
- if (contains(lowerComp,input) && !termFilter(lowerComp)) {
- //if(!searchTerms.contains(oboClass))takes long time!searchTerms.add(oboClass);
- uniqueTermList.addTerm(oboClass); // does unique check
- return true;
- }
+ else if (contains(lowerComp,input) && !termFilter(lowerComp)) {
+ searchTermList.addContainsTerm(oboClass);
+ return true;
}
-// // Starts with - not doing this for now
-// else if (lowerComp.startsWith(input)) {
-// if (!searchTerms.contains(oboClass))
-// searchTerms.add(oboClass);
-// }
return false;
}
@@ -283,10 +270,14 @@
public String getSource() { return source; }
- /** does unique check w map */
- private class UniqueTermList {
+ /** does unique check w map - dont need unique check - it was a bug in synonyms
+ how silly! - but actually this data structure will be handy for putting starts with
+ before contains! UniqueTermList -> SearchTermList*/
+ private class SearchTermList {
private Vector<OBOClass> searchTerms = new Vector<OBOClass>();
- private Map<OBOClass,OBOClass> uniqueCheck = new HashMap<OBOClass,OBOClass>();
+ //private Map<OBOClass,OBOClass> uniqueCheck = new HashMap<OBOClass,OBOClass>();
+ // list of terms that are contained but NOT startsWith
+ private Vector<OBOClass> containTerms = new Vector<OBOClass>();
private void addTerm(OBOClass oboClass) {
addTerm(oboClass,false);
}
@@ -294,13 +285,21 @@
addTerm(oboClass,true);
}
private void addTerm(OBOClass oboClass,boolean first) {
- if (uniqueCheck.containsKey(oboClass))
- return;
+ //if (uniqueCheck.containsKey(oboClass)) {
+ //log().debug("dup term in search "+oboClass);
+ // new Throwable().printStackTrace(); return; }
if (first) searchTerms.add(0,oboClass);
else searchTerms.add(oboClass);
- uniqueCheck.put(oboClass,null); // dont need value
+ //uniqueCheck.put(oboClass,null); // dont need value
}
- private Vector<OBOClass> getVector() { return searchTerms; }
+ /** Add term thats not startsWith but contains */
+ private void addContainsTerm(OBOClass oboClass) {
+ containTerms.add(oboClass);
+ }
+ private Vector<OBOClass> getVector() {
+ searchTerms.addAll(containTerms);
+ return searchTerms;
+ }
}
@@ -333,6 +332,20 @@
// GARBAGE
+ //if (!searchTerms.contains(oboClass)) {// this takes a long time w long lists!
+ //searchTerms.add(0,oboClass);
+ //if (!searchTerms.contains(oboClass)) searchTerms.add(oboClass);
+ //if(!searchTerms.contains(oboClass))takes long time!searchTerms.add(oboClass);
+ // SKIP PATO ATTRIBUTES - only want values - or do !contains "value"?
+ // yes do contains value - as theres other crap to filter too
+ // apparently curators want to see attribs as well according to Michael
+ //if (filterAttributes() && isAttribute(originalTerm))
+ //continue;
+
+ // pato subclass?
+// private boolean isAttribute(String term) {
+// return contains(term.toLowerCase(),"attribute");
+// }
// part of search params?
//private boolean filterAttributes() { return isPato(); }
Modified: phenote/trunk/src/java/phenote/main/Phenote.java
===================================================================
--- phenote/trunk/src/java/phenote/main/Phenote.java 2006-10-02 18:02:44 UTC (rev 146)
+++ phenote/trunk/src/java/phenote/main/Phenote.java 2006-10-03 23:11:33 UTC (rev 147)
@@ -6,6 +6,8 @@
import java.awt.Frame;
import java.awt.GridBagConstraints;
import java.awt.GridBagLayout;
+import java.awt.event.WindowAdapter;
+import java.awt.event.WindowEvent;
import javax.swing.BoxLayout;
import javax.swing.JFrame;
import javax.swing.JPanel;
@@ -30,9 +32,10 @@
public class Phenote {
- private static final String VERSION = "0.8 dev";
+ private static final String VERSION = "0.8.1 dev";
//private static final String DEFAULT_CONFIG_FILE = Config.DEFAULT_CONFIG_FILE;
private static final Logger LOG = Logger.getLogger(Phenote.class);
+ private static boolean standalone = false; // default for servlet
private CharacterTablePanel characterTablePanel;
private TermPanel termPanel;
@@ -43,6 +46,7 @@
public static void main(String[] args) {
+ standalone = true; // i think this is ok
System.out.println("This is Phenote version "+VERSION);
// default mac lok & feel is "Mac OS X", but the JComboBox is buggy
try {
@@ -161,10 +165,21 @@
frame.getContentPane().add(makeMainPanel());
MenuManager.createMenuManager(frame);
frame.setPreferredSize(new Dimension(1000,550));
+ if (standalone) // if stand alone exit java on window close
+ frame.addWindowListener(new WindowExit());
frame.pack();
frame.setVisible(true);
}
+ /** for standalone & webstart - not for servlet as i think it will bring the
+ whole servlet down */
+ private class WindowExit extends WindowAdapter {
+ public void windowClosing(WindowEvent e) {
+ if (standalone)
+ System.exit(0);
+ }
+ }
+
/** main panel contains TermPanel CharTablePanel & TermInfo */
private JPanel makeMainPanel() {
JPanel mainPanel = new JPanel(new GridBagLayout()); // ??
Modified: phenote/trunk/src/java/test/phenote/gui/TestPhenote.java
===================================================================
--- phenote/trunk/src/java/test/phenote/gui/TestPhenote.java 2006-10-02 18:02:44 UTC (rev 146)
+++ phenote/trunk/src/java/test/phenote/gui/TestPhenote.java 2006-10-03 23:11:33 UTC (rev 147)
@@ -162,7 +162,7 @@
qualityComboBox.simulateLKeyStroke();
qualityComboBox.setSelectedIndex(2);
// this is admittedly presumptious of quality
- assertEquals("acute angle",qualityComboBox.getText());
+ assertEquals("large volume",qualityComboBox.getText());
System.out.println("comp list sel ok "+qualityComboBox.getText());
}
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|