From f35bd7373186cb3262858b97480b962a64b65ae1 Mon Sep 17 00:00:00 2001 From: gmungoc Date: Thu, 21 Jan 2016 14:38:21 +0000 Subject: [PATCH 1/1] JAL-1705 canonicalise ENSEMBL name; code tidy --- src/jalview/util/DBRefUtils.java | 101 +++++++++++++++++---------------- test/jalview/util/DBRefUtilsTest.java | 1 + 2 files changed, 54 insertions(+), 48 deletions(-) diff --git a/src/jalview/util/DBRefUtils.java b/src/jalview/util/DBRefUtils.java index cdf2325..c85a489 100755 --- a/src/jalview/util/DBRefUtils.java +++ b/src/jalview/util/DBRefUtils.java @@ -24,18 +24,24 @@ import jalview.datamodel.DBRefEntry; import jalview.datamodel.DBRefSource; import jalview.datamodel.PDBEntry; import jalview.datamodel.SequenceI; -import jalview.ws.seqfetcher.DbSourceProxy; import java.util.ArrayList; import java.util.HashMap; +import java.util.HashSet; import java.util.Hashtable; import java.util.List; import java.util.Map; import com.stevesoft.pat.Regex; +/** + * Utilities for handling DBRef objects and their collections. + */ public class DBRefUtils { + /* + * lookup from lower-case form of a name to its canonical (standardised) form + */ private static Map canonicalSourceNameLookup = new HashMap(); private static Map dasCoordinateSystemsLookup = new HashMap(); @@ -47,6 +53,7 @@ public class DBRefUtils DBRefSource.UNIPROT); canonicalSourceNameLookup.put("uniprotkb/trembl", DBRefSource.UNIPROT); canonicalSourceNameLookup.put("pdb", DBRefSource.PDB); + canonicalSourceNameLookup.put("ensembl", DBRefSource.ENSEMBL); dasCoordinateSystemsLookup.put("pdbresnum", DBRefSource.PDB); dasCoordinateSystemsLookup.put("uniprot", DBRefSource.UNIPROT); @@ -55,39 +62,33 @@ public class DBRefUtils } /** - * Utilities for handling DBRef objects and their collections. - */ - /** * * @param dbrefs - * Vector of DBRef objects to search + * array of DBRef objects to search * @param sources * String[] array of source DBRef IDs to retrieve - * @return Vector + * @return */ public static DBRefEntry[] selectRefs(DBRefEntry[] dbrefs, String[] sources) { - if (dbrefs == null) - { - return null; - } - if (sources == null) + if (dbrefs == null || sources == null) { return dbrefs; } - Map srcs = new HashMap(); - ArrayList res = new ArrayList(); - - for (int i = 0; i < sources.length; i++) + HashSet srcs = new HashSet(); + for (String src : sources) { - srcs.put(new String(sources[i]), new Integer(i)); + srcs.add(src); } - for (int i = 0, j = dbrefs.length; i < j; i++) + + List res = new ArrayList(); + for (DBRefEntry dbr : dbrefs) { - if (srcs.containsKey(dbrefs[i].getSource())) + String source = getCanonicalName(dbr.getSource()); + if (srcs.contains(source)) { - res.add(dbrefs[i]); + res.add(dbr); } } @@ -96,8 +97,6 @@ public class DBRefUtils DBRefEntry[] reply = new DBRefEntry[res.size()]; return res.toArray(reply); } - res = null; - // there are probable memory leaks in the hashtable! return null; } @@ -201,8 +200,10 @@ public class DBRefUtils /** * match on all non-null fields in refa */ + // TODO unused - remove? public static DbRefComp matchNonNullonA = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { if (refa.getSource() == null @@ -231,27 +232,18 @@ public class DBRefUtils * either field is null or field matches for all of source, version, accession * id and map. */ + // TODO unused - remove? public static DbRefComp matchEitherNonNull = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { - if ((refa.getSource() == null || refb.getSource() == null) - || refb.getSource().equals(refa.getSource())) + if (nullOrEqual(refa.getSource(), refb.getSource()) + && nullOrEqual(refa.getVersion(), refb.getVersion()) + && nullOrEqual(refa.getAccessionId(), refb.getAccessionId()) + && nullOrEqual(refa.getMap(), refb.getMap())) { - if ((refa.getVersion() == null || refb.getVersion() == null) - || refb.getVersion().equals(refa.getVersion())) - { - if ((refa.getAccessionId() == null || refb.getAccessionId() == null) - || refb.getAccessionId().equals(refa.getAccessionId())) - { - if ((refa.getMap() == null || refb.getMap() == null) - || (refb.getMap() != null && refb.getMap().equals( - refa.getMap()))) - { - return true; - } - } - } + return true; } return false; } @@ -261,18 +253,18 @@ public class DBRefUtils * accession ID and DB must be identical. Version is ignored. Map is either * not defined or is a match (or is compatible?) */ + // TODO unused - remove? public static DbRefComp matchDbAndIdAndEitherMap = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { if (refa.getSource() != null && refb.getSource() != null && refb.getSource().equals(refa.getSource())) { // We dont care about version - // if ((refa.getVersion()==null || refb.getVersion()==null) - // || refb.getVersion().equals(refa.getVersion())) - // { if (refa.getAccessionId() != null && refb.getAccessionId() != null + // FIXME should be && not || here? || refb.getAccessionId().equals(refa.getAccessionId())) { if ((refa.getMap() == null || refb.getMap() == null) @@ -292,17 +284,16 @@ public class DBRefUtils * or map but no maplist on either or maplist of map on a is the complement of * maplist of map on b. */ + // TODO unused - remove? public static DbRefComp matchDbAndIdAndComplementaryMapList = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { if (refa.getSource() != null && refb.getSource() != null && refb.getSource().equals(refa.getSource())) { // We dont care about version - // if ((refa.getVersion()==null || refb.getVersion()==null) - // || refb.getVersion().equals(refa.getVersion())) - // { if (refa.getAccessionId() != null && refb.getAccessionId() != null || refb.getAccessionId().equals(refa.getAccessionId())) { @@ -329,8 +320,10 @@ public class DBRefUtils * or or map but no maplist on either or maplist of map on a is equivalent to * the maplist of map on b. */ + // TODO unused - remove? public static DbRefComp matchDbAndIdAndEquivalentMapList = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { if (refa.getSource() != null && refb.getSource() != null @@ -369,17 +362,13 @@ public class DBRefUtils */ public static DbRefComp matchDbAndIdAndEitherMapOrEquivalentMapList = new DbRefComp() { + @Override public boolean matches(DBRefEntry refa, DBRefEntry refb) { - // System.err.println("Comparing A: "+refa.getSrcAccString()+(refa.hasMap()?" has map.":".")); - // System.err.println("Comparing B: "+refb.getSrcAccString()+(refb.hasMap()?" has map.":".")); if (refa.getSource() != null && refb.getSource() != null && refb.getSource().equals(refa.getSource())) { // We dont care about version - // if ((refa.getVersion()==null || refb.getVersion()==null) - // || refb.getVersion().equals(refa.getVersion())) - // { if (refa.getAccessionId() != null && refb.getAccessionId() != null && refb.getAccessionId().equals(refa.getAccessionId())) { @@ -473,4 +462,20 @@ public class DBRefUtils return ref; } + /** + * Returns true if either object is null, or they are equal + * + * @param o1 + * @param o2 + * @return + */ + public static boolean nullOrEqual(Object o1, Object o2) + { + if (o1 == null || o2 == null) + { + return true; + } + return (o1 == null ? o2.equals(o1) : o1.equals(o2)); + } + } diff --git a/test/jalview/util/DBRefUtilsTest.java b/test/jalview/util/DBRefUtilsTest.java index 4a55a0d..b560eb8 100644 --- a/test/jalview/util/DBRefUtilsTest.java +++ b/test/jalview/util/DBRefUtilsTest.java @@ -96,6 +96,7 @@ public class DBRefUtilsTest assertEquals("UNIPROT", DBRefUtils.getCanonicalName("UNIPROTKB/TREMBL")); assertEquals("UNIPROTKB/SWISS-CHEESE", DBRefUtils.getCanonicalName("UNIPROTKB/SWISS-CHEESE")); + assertEquals("ENSEMBL", DBRefUtils.getCanonicalName("Ensembl")); } @Test(groups = { "Functional" }) -- 1.7.10.2