@@ -106,8 +106,6 @@ public UniprotProxySequenceReader(String accession, CompoundSet<C> compoundSet)
106106
107107 /**
108108 * The xml is passed in as a DOM object so we know everything about the protein.
109- * Some uniprot records contain white space in the sequence. We must strip it out so setContents doesn't fail.
110- * TODO add simmilar logic to the other constructors
111109 * If an error occurs throw an exception. We could have a bad uniprot id
112110 * @param document
113111 * @param compoundSet
@@ -116,7 +114,7 @@ public UniprotProxySequenceReader(String accession, CompoundSet<C> compoundSet)
116114 public UniprotProxySequenceReader (Document document , CompoundSet <C > compoundSet ) throws CompoundNotFoundException {
117115 setCompoundSet (compoundSet );
118116 uniprotDoc = document ;
119- String seq = this .getSequence (uniprotDoc ). replaceAll ( " \\ s" , "" ) ;
117+ String seq = this .getSequence (uniprotDoc );
120118 setContents (seq );
121119 }
122120 /**
@@ -144,15 +142,16 @@ public void setCompoundSet(CompoundSet<C> compoundSet) {
144142
145143 /**
146144 * Once the sequence is retrieved set the contents and make sure everything this is valid
145+ * Some uniprot records contain white space in the sequence. We must strip it out so setContents doesn't fail.
147146 * @param sequence
148147 * @throws CompoundNotFoundException
149148 */
150149 @ Override
151150 public void setContents (String sequence ) throws CompoundNotFoundException {
152151 // Horrendously inefficient - pretty much the way the old BJ did things.
153152 // TODO Should be optimised.
154- // NOTE This chokes on whitespace in the sequence, so whitespace is stripped by the caller
155- this .sequence = sequence ;
153+ // NOTE This chokes on whitespace in the sequence, so whitespace is stripped
154+ this .sequence = sequence . replaceAll ( " \\ s" , "" ) ;
156155 this .parsedCompounds .clear ();
157156 for (int i = 0 ; i < sequence .length ();) {
158157 String compoundStr = null ;
0 commit comments