change

SkillExtractV1.py

@@ -43,7 +43,10 @@ class SkillExtractorDetailsV1:
         # list_of_tuples = list(zip(skills_list, skill_Typelst,skill_Scorelst))        
         # df = pd.DataFrame(list_of_tuples,
         #           columns=['Skill_Name', 'Skill_Type','Skill_Score'])
-        data = {'Data':['Required Skills', 'Soft Skills', 'Good to have Skills'], 'Values':[','.join(reuiredSkilllst), ','.join(softskillslst), ','.join(goodtohavelst)]}
+        newSkilllst = SkillExtractorDetailsV1.extractWords(inputData, db_params)
+        reqSkill = ','.join(reuiredSkilllst) + ','.join(newSkilllst)
+
+        data = {'Data':['Required Skills', 'Soft Skills', 'Good to have Skills'], 'Values':[reqSkill, ','.join(softskillslst), ','.join(goodtohavelst)]}
         df = pd.DataFrame(data)
         
         return df        
@@ -54,5 +57,51 @@ class SkillExtractorDetailsV1:
         query = "select upper(skilldetails) skill from skillmaster where weightage = 0"
         df = pd.read_sql_query(query, conn)
         return df
+    
+    def extractWords(job_description,db_params):
+        job_roles = []
+        job_description = job_description.replace(')',' ')
+        delimiters = ",", " ", " , ", ";","\n","/","\\"
+        regex_pattern = '|'.join(map(re.escape, delimiters))
+        df = SkillExtractorDetailsV1.getNewSkills(db_params)
+        data = re.split(regex_pattern, job_description)
+        #data = job_description.split(',')
+        for ds in data:
+            #print(ds)
+            try:
+                if(SkillExtractorDetailsV1.skill_Validate(df,ds.strip())):                
+                    job_roles.append(ds) 
+                    #SkillExtractorDetails.GetSkillId(ds.strip(),JdMasterid,db_params)
+                    print("Skills Identified* : " + ds)   
+            except Exception as error:
+                test = 1
+        return job_roles 
+
+    def getNewSkills(db_params):
+        query = "select skillid,skilldetails,skilltype,skill_score from skillmaster where weightage = -2"
+        conn = psycopg2.connect(**db_params)
+        cursor = conn.cursor()
+        df_skill_master = pd.read_sql_query(query, conn)
+        df_skill_master['skilldetails'] = df_skill_master['skilldetails'].str.upper()
+        cursor.close()
+        # Close the connection
+        conn.close()
 
-    
+        #print(df_skill_master)
+        return df_skill_master
+    def skill_Validate(df, skill):
+        skill = skill.upper()
+        if (len(skill.split()) < 2 and len(skill) < 3) or len(skill.split())==1:
+            df['skill_present'] = df['skilldetails'].apply(lambda x: re.match(rf'^{skill}$', x))
+            if any(df['skill_present']):
+                #print("Valid Skill")
+                return 1
+            else:
+                #print("Not a Skill")
+                return 0
+        elif df['skilldetails'].str.contains(skill.upper()).any():
+            #print("Valid Skill")
+            return 1
+        else:
+            # print("Not a Skill")
+            return 0