Minor fix

intersystems-ib · Aug 23, 2024 · aa2179f · aa2179f
1 parent f017bcb
commit aa2179f
Show file tree

Hide file tree

Showing 3 changed files with 14 additions and 12 deletions.
diff --git a/encoder-ui/src/app/analysis/analysis.component.ts b/encoder-ui/src/app/analysis/analysis.component.ts
@@ -85,32 +85,34 @@ export class AnalysisComponent implements OnInit{
     var textHTML = textSelected;
     this.textToMark = textSelected;
     this.loading = true;
+    var regex = /[.,;:¿?!¡]/g;
+
     this.irisService.getAnalysisDetails(idAnalysis).subscribe({next: res => {
     if (res.length > 0){
       this.diagnostics = this.diagnostics.concat(res);
     }
     this.diagnostics.forEach((diagnostic, indexDiag) => {              
         let phrase: string = "";
         if (diagnostic.RawText.split(" ").length == 3){
-          const matchValue = textSelected.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[2],"ig"));
+          const matchValue = textSelected.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[2],"ig"));
           if (matchValue) {
             phrase = matchValue[0];
           }
         }
         else if (diagnostic.RawText.split(" ").length == 2){
-          const matchValue = textSelected.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[1],"ig"));
+          const matchValue = textSelected.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[1],"ig"));
           if (matchValue) {
             phrase = matchValue[0];
           }
         }
         else if (diagnostic.RawText.split(" ").length == 4){
-          const matchValue = textSelected.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[3],"ig"));
+          const matchValue = textSelected.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[3],"ig"));
           if (matchValue) {
             phrase = matchValue[0];
           }
         }
         else if (diagnostic.RawText.split(" ").length == 1){
-          const matchValue = textSelected.match(diagnostic.RawText);
+          const matchValue = textSelected.toLowerCase().replace(regex,"").match(diagnostic.RawText);
           if (matchValue) {
             phrase = matchValue[0];
           }

diff --git a/encoder-ui/src/app/analyzer/analyzer.component.ts b/encoder-ui/src/app/analyzer/analyzer.component.ts
@@ -53,6 +53,7 @@ export class AnalyzerComponent {
     this.totalReceived = 0;
     this.error = false;
     this.loading = true;
+    var regex = /[.,;:¿?!¡]/g;
     this.textAndDiagnosticList = [];
     const rawText = {
       "Text": textOriginal
@@ -77,25 +78,25 @@ export class AnalyzerComponent {
                 this.diagnostics.forEach((diagnostic, indexDiag) => {              
                     let phrase: string = "";
                     if (diagnostic.RawText.split(" ").length == 3){
-                      const matchValue = textOriginal.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[2],"ig"));
+                      const matchValue = textOriginal.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[2],"ig"));
                       if (matchValue) {
                         phrase = matchValue[0];
                       }
                     }
                     else if (diagnostic.RawText.split(" ").length == 2){
-                      const matchValue = textOriginal.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[1],"ig"));
+                      const matchValue = textOriginal.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[1],"ig"));
                       if (matchValue) {
                         phrase = matchValue[0];
                       }
                     }
                     else if (diagnostic.RawText.split(" ").length == 4){
-                      const matchValue = textOriginal.match(new RegExp(diagnostic.RawText.split(" ")[0] + " (.{0,100}) " +diagnostic.RawText.split(" ")[3],"ig"));
+                      const matchValue = textOriginal.toLowerCase().replace(regex,"").match(new RegExp(diagnostic.RawText.split(" ")[0] + "(.{0,100})" +diagnostic.RawText.split(" ")[3],"ig"));
                       if (matchValue) {
                         phrase = matchValue[0];
                       }
                     }
                     else if (diagnostic.RawText.split(" ").length == 1){
-                      const matchValue = textOriginal.match(diagnostic.RawText);
+                      const matchValue = textOriginal.toLowerCase().replace(regex,"").match(diagnostic.RawText);
                       if (matchValue) {
                         phrase = matchValue[0];
                       }

diff --git a/src/ENCODER/BP/AnalyzeTextProcess.cls b/src/ENCODER/BP/AnalyzeTextProcess.cls
@@ -35,14 +35,13 @@ Method AnalyzeText(text As %String, analysisId As %String, language As %String)
                 lexical_tokens = [t.lower() for t in words if t.isalpha()]
                 phrase = " ".join(lexical_tokens)
                 result = ""
-                words = phrase.split(" ")
-                totalWords = len(words)
+                totalWords = len(lexical_tokens)
                 i = 0
                 texts = []
                 while i + 3 <= totalWords :
-                    texts.append(words[i]+" "+words[i+1]+" "+words[i+2])
+                    texts.append(lexical_tokens[i]+" "+lexical_tokens[i+1]+" "+lexical_tokens[i+2])
                     i = i + 1
-                if len(texts) == 0 :
+                if totalWords < 3 :
                     texts.append(phrase)
 
                 embeddings = model.encode(texts, normalize_embeddings=True)