def highlight_file(path):
file_content = read_file(path)
return highlight(
file_content, JavaLexer(),
HtmlFormatter(linenos=True, anchorlinenos=True, lineanchors="foo"))
def highlight_code(self, search_count, rank, targetpath):
html_snippets = []
matched_line_numbers = self.matched_line_number()
if matched_line_numbers:
snippet_cluster_lns = self.compute_lines_to_highlight(
self.adjacent_line_numbers())
snippets = []
##################### Data provision ######################
pure_snippets_for_data_requirement = [] ##
#snippet_cluster_lns ##lines_to_be_highlighted
#self.file_content_lines ##entire code
for snippet_cluster_ln in snippet_cluster_lns:
snippet = []
for n in snippet_cluster_ln:
snippet.append(self.file_content_lines[n])
pure_snippets_for_data_requirement.append(
self.file_content_lines[n])
start_line = min(snippet_cluster_ln)
# end_line = max(snippet_cluster_ln)
highlight_lines = map(lambda x: x - start_line + 1,
self.matching_line_numbers)
snippets.append(
("\n".join(snippet), start_line, highlight_lines))
# #Data provision_ Defect4J (모든) 각 쿼리별 랭킹순으로 하이라이트 라인번호 + 전체코드 파일
final_str = "\n".join(self.file_content_lines)
original_project_list = [
'knutwalker_google-closure-compiler',
'google_closure-compiler', 'weitzj_closure-compiler',
'jfree_jfreechart-fse', 'jfree_jfreechart',
'apache_commons-math', 'apache_commons-lang',
'mockito_mockito', 'bryceguo_mockito'
]
if not self.file_path.split(
'/'
)[6] in original_project_list: #Check the duplicate projects.
purepath = targetpath[:-4]
if not os.path.exists(purepath):
os.makedirs(purepath)
testcode_path = purepath + '_result_testcode'
if not os.path.exists(testcode_path):
os.makedirs(testcode_path)
if self.file_path.split(
'/')[-1] == 'Test.java': # 결과 파일이 Test.java 인 것들 걸러내기
pass
else:
final_path = purepath + "/" + str(rank) + '_' + str(
"||".join(self.file_path.split('/')[6:]))
write_file(final_path, str(final_str))
write_file(final_path + "_", str(snippet_cluster_lns))
print "*****************************", final_path, "is Done.."
# 여기서 test code 또한 찾아서 셋트로 돌려줘보자.
# >> 각각 결과코드에 해당하는 프로젝트 경로를 새로 다 뒤져서 현재 결과파일 앞뒤로 test 키워드 들어있는 파일들을 찾아본다.
# >> 테스트 파일 찾았으면 복붙
result_file_name = self.file_path.split('/')[-1]
result_pure_file_name = (
(self.file_path.split('/')[-1]).split('.')
)[0] #pure name of the java file (e.g., ABC.java -> ABC)
stopwords = ['A', 'a', 'test']
javafiles = java_files_from_dir('/'.join(
self.file_path.split('/')[:7]))
for javafile in javafiles:
if result_pure_file_name in stopwords:
continue
name_of_javafile = javafile.split('/')[-1].split(
'.')[0]
if 'test' in name_of_javafile and result_pure_file_name in name_of_javafile: #해당 결과파일의 이름과 'test' 라는 키워드가 들어가있는 파일이면 (e.g., xxxtest.java or testxxx.java)
content = read_file(javafile)
testcode_path = testcode_path + '/' + str(
rank) + '_' + str('||'.join(
javafile.split('/')[6:]))
write_file(testcode_path, content)
write_file('/Users/Falcon/Desktop/count.txt',
testcode_path)
write_file('/Users/Falcon/Desktop/count.txt',
javafile)
write_file('/Users/Falcon/Desktop/count.txt',
'**********************')
html_snippets = [
highlight(snippet[0], JavaLexer(),
LatexFormatter(linenos=True, linenostart=snippet[1]))
for snippet in snippets
]
self.code_snippets = [
GitSearchItemSnippet(self.hl_snippet(snippet[0], snippet[1]),
snippet[1]) for snippet in snippets
]
if not html_snippets:
html_snippets.append(
highlight(self.file_content, JavaLexer(),
HtmlFormatter(linenos=True, anchorlinenos=True)))
self.code_snippets.append(
GitSearchItemSnippet(self.hl_snippet(self.file_content, 0), 0))
return "".join(html_snippets)