def computeSimilarityScors(newReportSummary, newReportDescription , All_3partAPIinfo_list): #,Src_info_file_dir):
all_APIdescription = []
for ele in All_3partAPIinfo_list:
all_APIdescription.append(ele[5])
#版本二，数据已经经过文本处理，节省时间
newRportSD= computeSimilarity.tokenize_stopwords_stemmer( [newReportSummary,newReportDescription] )
scores = Half_computeSimilarity(newRportSD,all_APIdescription)
"""
#用于将数据归一化处理
Max_socre = scores[0]
for score in scores:
if Max_socre < score and score < 1 :#里面有一个是其本身。最大值为一
Max_socre = score
if Max_socre != 0:
for i in range(len(scores)):
scores[i] = scores[i] / Max_socre
"""
#print time(),44
Scores={}
for i in range(len(All_3partAPIinfo_list)):
API = All_3partAPIinfo_list[i][0]+ '.' + All_3partAPIinfo_list[i][1]
Scores[ API ] = scores[i]
"""
Scores = sorted(Scores.iteritems(), key = lambda asd:asd[1], reverse = True)
return Scores # [(key,value)]
"""
return Scores #直接用字典，这)# + ran样便于后续的关键字查找
sheet2.write(i,column+2,comments[column*30000 :].decode('utf-8'))
i = i + 1
f.save('/Users/apple/Documents/API推荐项目/hbase/SrcInfo.xls')#保存文件
"""
#版本2 文本预处理版
#project_dict= { "类的路径" : [【类名】，【方法名】，【( 变量类型, 变量名 ),...,( 变量类型, 变量名 )】，【注释】] , ... }
i=1
for key in project_dict:
sheet1.write(i,0,key.decode('utf-8'))
#sheet2.write(i,0,key.decode('utf-8'))
#类名
Class_name0=computeSimilarity.tokenize_stopwords_stemmer(project_dict[key][0])
Class_name=''
for word in Class_name0:
Class_name = Class_name+' ' + word
sheet1.write(i,1,Class_name.strip(' '))
#sheet2.write(i,1,Class_name.strip(' '))
#方法名
method_name=''
for name in project_dict[key][1]:
method_name = name + ' '+method_name
Method_name0=Tokenize_stopwords_stemmer([method_name])
Method_name=''
for word in Method_name0:
def computeSimilarityScors(newReportSummary, newReportDescription,
sheet1): #,Src_info_file_dir):
all_classdir = []
all_className = []
all_methodName = []
all_variableName = []
all_comments = []
"""
workbook = xlrd.open_workbook(Src_info_file_dir,'r')
sheet1 = workbook.sheet_by_name('sheet1')
"""
#sheet2 = workbook.sheet_by_name('comments')
#print 1
#print sheet1.cell(6,28).value.encode('utf-8')
for i in range(1, sheet1.nrows):
all_classdir.append(sheet1.cell(i, 0).value.encode('utf-8'))
className = []
className = sheet1.cell(i, 1).value.encode('utf-8').split(' ')
all_className.append(className)
methodName = []
methodName = sheet1.cell(i, 2).value.encode('utf-8').split(' ')
all_methodName.append(methodName)
variableName = []
variableName = sheet1.cell(i, 3).value.encode('utf-8').split(' ')
all_variableName.append(variableName)
comments_str = ''
j = 6
while 1:
try: #因为不确定 注释占了几格
comments_str = comments_str + ' ' + sheet1.cell(
i, j).value.encode('utf-8')
j = j + 1
except IndexError:
break
comments_str = comments_str.strip(' ')
comments = []
comments = comments_str.split(' ')
all_comments.append(comments)
#print i
#版本二，数据已经经过文本处理，节省时间
newRportS = computeSimilarity.tokenize_stopwords_stemmer(
[newReportSummary])
newRportD = computeSimilarity.tokenize_stopwords_stemmer(
[newReportDescription])
#print 11
all_classNameS_result = Half_computeSimilarity(newRportS, all_className)
all_methodNameS_result = Half_computeSimilarity(newRportS, all_methodName)
#print 22
all_variableNameS_result = Half_computeSimilarity(newRportS,
all_variableName)
all_commentsS_result = Half_computeSimilarity(newRportS, all_comments)
#print 33
all_classNameD_result = Half_computeSimilarity(newRportD, all_className)
all_methodNameD_result = Half_computeSimilarity(newRportD, all_methodName)
#print 44
all_variableNameD_result = Half_computeSimilarity(newRportD,
all_variableName)
all_commentsD_result = Half_computeSimilarity(newRportD, all_comments)
#print 55
"""
#版本一：数据无文本预处理
#print sheet1.cell(6,28).value.encode('utf-8')
for i in range(1,sheet1.nrows):
all_classdir.append(sheet1.cell(i,0).value.encode('utf-8'))
className=[]
className.append(sheet1.cell(i,1).value.encode('utf-8'))
all_className.append(className)
methodName=[]
methodName.append(sheet1.cell(i,2).value.encode('utf-8'))
all_methodName.append(methodName)
variableName=[]
variableName.append(sheet1.cell(i,3).value.encode('utf-8'))
all_variableName.append(variableName)
comments_str=''
j=2
while 1 :
try:#因为不确定 注释占了几格
comments_str= comments_str + sheet2.cell(i,j).value.encode('utf-8')
j=j+1
except IndexError:
break
comments=[]
comments.append(comments_str)
all_comments.append(comments)
newRportS=[newReportSummary]
newRportD=[newReportDescription]
all_classNameS_result = computeSimilarity.all_computeSimilarity(newRportS,all_className)
all_methodNameS_result = computeSimilarity.all_computeSimilarity(newRportS,all_methodName)
all_variableNameS_result = computeSimilarity.all_computeSimilarity(newRportS,all_variableName)
all_commentsS_result = computeSimilarity.all_computeSimilarity(newRportS,all_comments)
all_classNameD_result = computeSimilarity.all_computeSimilarity(newRportD,all_className)
all_methodNameD_result = computeSimilarity.all_computeSimilarity(newRportD,all_methodName)
all_variableNameD_result = computeSimilarity.all_computeSimilarity(newRportD,all_variableName)
all_commentsD_result = computeSimilarity.all_computeSimilarity(newRportD,all_comments)
"""
"""
result = all_classNameS_result + all_classNameD_result
+ all_methodNameS_result + all_methodNameD_result
+ all_variableNameS_result + all_variableNameD_result
+ all_commentsS_result + all_commentsD_result
"""
result = []
result.append(all_classdir)
result.append(all_classNameS_result)
result.append(all_classNameD_result)
result.append(all_methodNameS_result)
result.append(all_methodNameD_result)
result.append(all_variableNameS_result)
result.append(all_variableNameD_result)
result.append(all_commentsS_result)
result.append(all_commentsD_result)
return result
def computeSimilarityScors(newReportSummary,newReportDescription ,sheet1): #,Src_info_file_dir):
all_classdir=[]
all_program_lanuage=[]
all_comments=[]
"""
workbook = xlrd.open_workbook(Src_info_file_dir,'r')
sheet1 = workbook.sheet_by_name('sheet1')
"""
#sheet2 = workbook.sheet_by_name('comments')
#print 1
#print sheet1.cell(6,28).value.encode('utf-8')
for i in range(1,sheet1.nrows):
all_classdir.append(sheet1.cell(i,0).value.encode('utf-8'))
className=[]
className = sheet1.cell(i,1).value.encode('utf-8').split(' ')
methodName=[]
methodName = sheet1.cell(i,2).value.encode('utf-8').split(' ')
variableName=[]
variableName = sheet1.cell(i,3).value.encode('utf-8').split(' ')
program_lanuage = className + methodName + variableName
all_program_lanuage.append(program_lanuage)
comments_str=''
j=6
while 1 :
try:#因为不确定 注释占了几格
comments_str= comments_str + ' ' + sheet1.cell(i,j).value.encode('utf-8')
j=j+1
except IndexError:
break
comments_str = comments_str.strip(' ')
comments=[]
comments = comments_str.split(' ')
all_comments.append(comments)
#print i
#版本二，数据已经经过文本处理，节省时间
newRportSD= computeSimilarity.tokenize_stopwords_stemmer( [newReportSummary + newReportDescription] )
#print 11
all_program_language_result = Half_computeSimilarity(newRportSD,all_program_lanuage)
all_commentsSD_result = Half_computeSimilarity(newRportSD,all_comments)
#归一化处理
Max_socre = all_program_language_result[0]
for score in all_program_language_result:
if Max_socre < score and score < 1 :#里面有一个是其本身。最大值为一
Max_socre = score
if Max_socre != 0:
for i in range(len(all_program_language_result)):
all_program_language_result[i] = all_program_language_result[i] / Max_socre
Max_socre = all_commentsSD_result[0]
for score in all_commentsSD_result:
if Max_socre < score and score < 1 :#里面有一个是其本身。最大值为一
Max_socre = score
if Max_socre != 0:
for i in range(len(all_commentsSD_result)):
all_commentsSD_result[i] = all_commentsSD_result[i] / Max_socre
result=[]
result.append(all_classdir)
result.append(all_program_language_result)
result.append(all_commentsSD_result)
return result