Skip to content

Commit a17a617

Browse files
committed
There can be more than one space as well
1 parent be56559 commit a17a617

File tree

1 file changed

+4
-1
lines changed

1 file changed

+4
-1
lines changed

src/edu/pitt/cs/nih/backend/feedback/TextFileFeedbackManager_LibSVM_WordTree.java

Lines changed: 4 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -834,6 +834,7 @@ protected String getStartEndPosition(String docID, Map<String,String> spanMap) t
834834
String fn_pathology = Storage_Controller.getPathologyReportFn();
835835
String docText;
836836
Pattern pattern = getSearchPatternFromSpanMap(spanMap);
837+
System.out.println(pattern);
837838
Matcher m;
838839
// search text in colonoscopy text, remove header footer
839840
// docText = Preprocess.separateReportHeaderFooter(
@@ -939,6 +940,8 @@ public String wordTreeSkippedNGramPatternString(Map<String, String> spanMap)
939940
String[] matchedTokenList = TextUtil.escapeRegex(spanMap.get("matched")).split(" ");
940941

941942

943+
System.out.println("Received span map: " + spanMap);
944+
942945
StringBuilder sb = new StringBuilder();
943946
// matchedTokenList.length >= selectedTokenList.length
944947
int skippedN = 0;
@@ -978,7 +981,7 @@ public String wordTreeSkippedNGramPatternString(Map<String, String> spanMap)
978981
patternStr = sb.toString().trim().replaceAll(whiteSpaceBeforePunc, "\\\\s{0,1}");
979982
// in case the first skipped n-gram is a punctuation
980983
// there would be no white space before the n-gram
981-
patternStr = patternStr.replaceAll(" (?=(\\(\\\\S\\+))", "\\\\s{0,1}");
984+
patternStr = patternStr.replaceAll(" (?=(\\(\\\\S\\+))", "\\\\s*");
982985
// // quote the string
983986
// patternStr = TextUtil.escapeRegex(patternStr);
984987
// reverse 's

0 commit comments

Comments
 (0)