def downloadKnowled(self,
select_sql=SQL.SELECT_SUBJECT_RELATION,
inser_konw_sql=SQL.INSERT_21CNJY_KNOWLED,
know_url=URL.KNOW_URL,
know_child_url=URL.KNOW_CHILD_URL):
'''下载知识点'''
pg = PostgreSql()
try:
for row in pg.getAll(select_sql):
try:
subject_code, xd, subject_zname, course_21 = row
response = self.session.get(
know_url % (xd, course_21, Utils.getCurrMilliSecond()),
headers=self.headers)
rs = self.__recursiveKnowled(
response.json(), 1, subject_code,
know_child_url % ('%s', xd, course_21, '%s'))
if rs: pg.batchExecute(inser_konw_sql, rs)
pg.commit()
logger.info(
u'完成二一组卷网（学段:%s,学科:%s,线上学科名称:%s,线上学科代码:%s）知识点的导入，导入知识点数量:%d',
xd, course_21, subject_zname, subject_code, len(rs))
except Exception as e:
logger.exception(
u'二一组卷网（学段:%s,学科:%s,线上学科名称:%s,线上学科代码:%s）知识点的导入异常', xd,
course_21, subject_zname, subject_code)
pg.rollback()
finally:
pg.close()
def downloadSubject(self,
select_sql=SQL.SELECT_SUBJECT_RELATION,
update_sql=SQL.UPDATE_SUBJECT_RELATION,
subjects_url=URL.SUBJECTS_URL %
Utils.getCurrMilliSecond()):
'''下载21世纪学科学段，并更新与线网关系对应'''
response = self.session.get(subjects_url, headers=self.headers)
zj21cnjy_subject = response.json()
pg = PostgreSql()
try:
update_params = []
for row in pg.getAll(select_sql):
subject_code, xd, subject_zname, course_21 = row
if course_21: continue
course_21_name = None
for key, value in zj21cnjy_subject[str(xd)].iteritems():
value_temp = value if value != u'政治思品' else u'政治'
if subject_zname.find(value_temp) > -1:
course_21 = key
course_21_name = value
if course_21:
update_params.append(
(course_21, course_21_name, subject_code))
if update_params: pg.batchExecute(update_sql, update_params)
pg.commit()
logger.info(u'完成21cnjy与线网学科学段对应关系更新,更新数量：%d', len(update_params))
except Exception as e:
logger.exception(u'21cnjy与线网学科学段对应关系更新出现异常，异常信息:%s', e.message)
pg.rollback()
finally:
pg.close()
def execExtract(self,
select_main_sql=SQL.select_main_sql,
insert_image_sql=SQL.insert_image_sql,
update_main_sql=SQL.update_main_sql,
select_convert_sql=SQL.select_convert_sql):
'''执行提取图'''
postgreSql = PostgreSql()
count = 0
rs = True
try:
flag = True # 代表数据库里面还有需要处理的数据
while flag:
try:
flag = False
insert_image_params = []
update_main_params = []
for rows in postgreSql.getAll(select_main_sql):
flag = True
seq = rows[0]
qid = rows[1]
try:
urls = []
for col in rows[2:]:
for j_url in JyeooUtil.getJyeooImg(col):
if j_url not in urls:
urls.append(j_url)
# 生成临时的图片文件
self.__generateTmpImage(urls, postgreSql,
select_convert_sql)
# 插入数据到img表
insert_image_params.append(
(seq, qid, json.dumps(urls), 0 if urls else 2))
# 更新jyeoo主表的数据状态
update_main_params.append((1 if urls else 2, qid))
except Exception as ex:
rs = False
logger.exception(
u"提取图片-----处理qi=%s，创建题目的图片发生异常,异常信息：%s" %
(qid, ex.message))
return rs
if insert_image_params:
postgreSql.batchExecute(insert_image_sql,
insert_image_params)
if update_main_params:
postgreSql.batchExecute(update_main_sql,
update_main_params)
postgreSql.commit()
count += len(insert_image_params)
logger.info(u'提取图片-----已成功处理题目数量:%d' % count)
except Exception as e:
postgreSql.rollback()
rs = False
logger.exception(u"提取图片-----批量处理-异常信息:%s" % (e.message))
finally:
postgreSql.close()
return rs
def extractQuesImage(self,
rows=1000,
select_batch_ques=SQL.SELECT_BATCH_QUES,
insert_image_url=SQL.INSERT_IMAGE_URL,
update_status=SQL.UPDATE_STATUS):
'''分析提取题目图片'''
logger.info(u'开始分析提取题目图片')
seq = 0
try:
pg = PostgreSql()
flag = True
count = 0
while flag:
try:
flag = False
insert_params = []
update_params = []
for row in pg.getAll(select_batch_ques, (0, seq, rows)):
flag = True
qid = row[0]
old_id = row[1]
seq = row[2]
try:
urls = [row[3], row[4]]
for col in row[5:]:
if col is None: continue
urls.extend(self.__get21cnjyImg(col))
# 生成临时的图片文件
self.__generateTmpImage(urls)
# 插入数据到img表 存在图片状态为0，不存在图片状态为2
insert_params.append((seq, qid, Utils.toJson(urls),
0 if urls else 2))
# 更新21cnjy主表的数据状态 存在图片状态修改为1，不存在图片状态为2
update_params.append((1 if urls else 2, qid))
except Exception as ex:
logger.exception(
u"处理qi=%s，old_id=%s，创建题目的图片发生异常,异常信息：%s" %
(qid, old_id, ex.message))
if update_params:
pg.batchExecute(update_status, update_params)
if insert_params:
pg.batchExecute(insert_image_url, insert_params)
pg.commit()
count += len(update_params)
logger.info(u'已成功处理题目数量:%d' % count)
except Exception as e:
pg.rollback()
logger.exception("批量处理-异常信息:%s" % (e.message))
finally:
pg.close()
def findImags(self, subject):
seq_num = 0
rows = self.ROWS
try:
pg = PostgreSql()
flag = True
count = 0
while flag:
try:
flag = False
insert_params = []
update_params = []
for row in pg.getAll(self.SELECT_SQL,
(subject, seq_num, rows)):
flag = True
qid = row[0]
old_id = row[1]
seq_num = row[2]
try:
urls = []
for col in row[3:]:
if col is None: continue
urls.extend(self.getZjImg(col))
print(urls)
# 生成临时的图片文件
self.generateTmpImage(urls)
# 插入数据到img表 存在图片状态为0，不存在图片状态为2
insert_params.append(
(qid, json.dumps(urls), 0 if urls else 2))
# 更新jyeoo主表的数据状态 存在图片状态修改为1，不存在图片状态为2
update_params.append((1 if urls else 2, qid))
except Exception as ex:
logger.exception(
u"处理qi=%s，old_id=%s，创建题目的图片发生异常,异常信息：%s" %
(qid, old_id, ex.message))
if update_params:
pg.batchExecute(self.UPDATE_SQL, update_params)
if insert_params:
pg.batchExecute(self.INSERT_SQL, insert_params)
pg.commit()
count += len(update_params)
logger.info(u'已成功处理题目数量:%d' % count)
except Exception as e:
pg.rollback()
logger.exception("批量处理-异常信息:%s" % (e.message))
finally:
pg.close()
def downloadQuestions(
self,
ques_type='',
ques_pg_url=URL.QUES_PG_URL,
select_knowled_id=SQL.SELECT_KNOWLED_ID,
select_subject_sql=SQL.SELECT_SUBJECT_RELATION,
select_params_type=SQL.SELECT_PARAMS_TYPE,
update_knowled_downloded=SQL.UPDATE_KNOWLED_DOWNLODED):
'''按知识点下载题目'''
pg = PostgreSql()
try:
#获取题目类型信息
ques_type_dic = {}
for row_ques_types in pg.getAll(select_params_type,
('ques_type', 1)):
subject_code, code_21cnjy, name_21cnjy, code, name = row_ques_types
#按学科类型分组
if not ques_type_dic.has_key(subject_code):
ques_type_dic[subject_code] = {}
#具体学科的编码类型信息
ques_type_dic[subject_code][code_21cnjy] = {
'code_21cnjy': code_21cnjy,
'name_21cnjy': name_21cnjy,
'code': code,
'name': name
}
#获取学段信息
for row in pg.getAll(select_subject_sql):
try:
subject_code, xd, subject_zname, course_21 = row
self.__downloadQuestionsBySubject(
row, pg, ques_pg_url, ques_type_dic[subject_code],
ques_type, select_knowled_id, update_knowled_downloded)
except Exception as e:
logger.exception(
u'二一组卷网（学段:%s,学科:%s,线上学科名称:%s,线上学科代码:%s）题目导入异常', xd,
course_21, subject_zname, subject_code)
pg.rollback()
raise e
finally:
pg.close()
def findImags(subject):
rows = ROWS
try:
pg = PostgreSql()
flag = True
count = 0
qid = '0'
err_count = 0
while flag:
try:
flag = False
update_params = []
update_err_params = []
for row in pg.getAll(SELECT_SQL, (subject, qid, rows)):
flag = True
qid = row[0]
url = getZjImg(row[1])[0]
try:
choice_answer = getTextByImageUrl2(url)
answer_arr = []
answer_arr.append(choice_answer)
update_params.append(
(json.dumps(answer_arr, ensure_ascii=False), qid))
count += 1
except Exception as ex:
err_count += 1
update_err_params.append((-1, qid))
logger.exception('异常的题目ID:%s,url:%s', qid, url)
if update_params: pg.batchExecute(UPDATE_SQL, update_params)
if update_err_params:
pg.batchExecute(UPDATE_STATUS_SQL, update_err_params)
pg.commit()
logger.info(u'学科编码：%d，已成功处理题目数量:%d，错误数量：%d' %
(subject, count, err_count))
except Exception as e:
pg.rollback()
logger.exception("学科编码：%d，批量处理-异常信息:%s" % (subject, e.message))
finally:
pg.close()
def downloadQueryParams(self,
select_sql=SQL.SELECT_SUBJECT_RELATION,
insert_sql=SQL.INSERT_21CNJY_TYPE,
query_param_url=URL.QUERY_PARAM_URL,
ques_query_type=QUES_QUERY_TYPE):
'''下载题目查询参数----如题目类型、难度等等'''
pg = PostgreSql()
count = 0
try:
for row in pg.getAll(select_sql):
try:
subject_code, xd, subject_zname, course_21 = row
insert_params = []
response = self.session.get(
query_param_url %
(xd, course_21, Utils.getCurrMilliSecond()),
headers=self.headers)
for param_type, values in response.json().iteritems():
if not ques_query_type.has_key(param_type):
continue
for code, name in values.iteritems():
count += 1
id = count
insert_params.append(
(id, ques_query_type[param_type], code, name,
subject_code))
if insert_params:
pg.batchExecute(insert_sql, insert_params)
pg.commit()
logger.info(
u'完成二一组卷网（学段:%s,学科:%s,线上学科名称:%s,线上学科代码:%s）查询参数的导入，导入参数的数量:%d，所有学科处理总数%d',
xd, course_21, subject_zname, subject_code,
len(insert_params), count)
except Exception as e:
logger.exception(
u'二一组卷网（学段:%s,学科:%s,线上学科名称:%s,线上学科代码:%s）查询参数的导入异常', xd,
course_21, subject_zname, subject_code)
pg.rollback()
finally:
pg.close()
def main(self,startTime=start_time, root_path=PATH.rootImagPath,
pic_new_path=PATH.pic_new_path):
select_sql = self.SELECT_SQL_IMG
update_sql = self.UPDATE_SQL
update_sql_img = self.UPDATE_SQL_IMG
insert_sql = self.INSERT_SQL_CONVERT
curr_time = time.time()
curr_time_strft = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(curr_time))
# 记录当前分析时间
logger.info(u'本次分析时间：%s,秒：%.2f' % (curr_time_strft, curr_time))
postgreSql = PostgreSql()
count = 0
total = 0
try:
flag = True # 代表数据库里面还有需要处理的数据
id = 0
while flag:
try:
flag = False
update_params = []
update_image_params = []
insert_params = []
for rows in postgreSql.getAll(select_sql % id):
flag = True
total += 1
id = rows[0] if rows[0] > id else id
qid = rows[1]
urls = rows[2]
try:
isDownloadFinish = True
urlMap = {}
for url in json.loads(urls):
url_path = urlparse.urlsplit(url)
fileName = os.path.join(root_path, url_path.path[1:])
if os.path.exists(fileName):
mtime = os.path.getmtime(fileName)
if mtime >= curr_time:
isDownloadFinish = False
elif start_time <= mtime:
(temp, extension) = os.path.splitext(fileName)
# 新文件名称
file_new_name = "%s%s" % (Utils.getStrMD5(url + "-mqm"), extension)
# 新文件名称 - 全名
file_new_name_all = os.path.join(pic_new_path, file_new_name)
# 新的url
url_new = image_url + file_new_name
urlMap[url] = url_new
if not os.path.exists(file_new_name_all):
shutil.copy2(fileName, file_new_name_all)
Utils.modifyMD5(file_new_name_all)
insert_params.append((url, url_new))
else:
# 表示为之前处理过的图片
sql_url = self.SQL_URL
urlMap[url] = postgreSql.getOne(sql_url, (url,))[0]
# if urlMap[url]:
# logger.error(u'oldurl:%s,数据不存在'% url)
else:
isDownloadFinish = False
# 下载完成就更新t_jyeoo_img_url
if isDownloadFinish:
update_image_params.append((1, qid))
# 设置替换的图片url、更新原始数据表的状态为3（有图片、图片下载完成）
update_params.append((json.dumps(urlMap), 3, qid))
except Exception as ex:
logger.exception(u"处理qi=%s，校验题目的所有图片下载是否完成发生异常,异常信息：%s" % (qid, ex.message))
if update_params: postgreSql.batchExecute(update_sql, update_params)
if update_image_params: postgreSql.batchExecute(update_sql_img, update_image_params)
if insert_params: postgreSql.batchExecute(insert_sql, insert_params)
postgreSql.commit()
count += len(update_image_params)
logger.info(u'已成功处理题目数量:%d，校验题目数量总数:%d' % (count, total))
except Exception as e:
postgreSql.rollback()
logger.exception("批量处理-异常信息:%s" % (e.message))
finally:
postgreSql.close()
def execParseImage(self,
select_image_sql=SQL.select_image_sql,
select_convert_sql=SQL.select_convert_sql,
update_main_url_sql=SQL.update_main_url_sql,
update_image_sql=SQL.update_image_sql,
insert_convert_sql=SQL.insert_convert_sql,
picture_path=PATH.picture_path,
pic_new_path=PATH.pic_new_path,
pic_relative_path=PATH.pic_relative_path,
image_url=image_url):
pic_new_real_path = os.path.join(pic_new_path, pic_relative_path)
image_real_url = urlparse.urljoin(image_url, pic_relative_path)
logger.info(u'进入处理图片流程，原始图片路径：%s，处理后图片存放路径：%s，图片url前缀地址：%s',
picture_path, pic_new_real_path, image_real_url)
if not os.path.exists(pic_new_real_path):
os.makedirs(pic_new_real_path)
postgreSql = PostgreSql()
count = 0
total = 0
rs = True
try:
flag = True # 代表数据库里面还有需要处理的数据
id = 0
while flag:
try:
flag = False
update_main_params = []
update_image_params = []
insert_convert_params = []
for rows in postgreSql.getAll(select_image_sql % id):
flag = True
total += 1
id = rows[0] if rows[0] > id else id
qid = rows[1]
urls = rows[2]
try:
isDownloadFinish = True
urlMap = {}
for url in json.loads(urls):
url_path = urlparse.urlsplit(url)
fileName = os.path.join(
picture_path, url_path.path[1:])
if os.path.exists(fileName):
(temp,
extension) = os.path.splitext(fileName)
# 新文件名称
file_new_name = "%s%s" % (Utils.getStrMD5(
url + "-mqm"), extension)
# 新文件名称 - 全名
file_new_name_all = os.path.join(
pic_new_path, file_new_name)
# 新的url
url_new = image_real_url + file_new_name
urlMap[url] = url_new
if os.path.exists(file_new_name_all):
if not postgreSql.getOne(
select_convert_sql, (url, )):
insert_convert_params.append(
url, url_new)
else:
shutil.copy2(fileName,
file_new_name_all)
Utils.modifyMD5(file_new_name_all)
insert_convert_params.append(
(url, url_new))
else:
#查询
rs = postgreSql.getOne(
select_convert_sql, (url, ))
if rs:
urlMap[url] = rs[0]
else:
isDownloadFinish = False
# 下载完成就更新t_jyeoo_img_url
if isDownloadFinish:
update_image_params.append((1, qid))
# 设置替换的图片url、更新原始数据表的状态为3（有图片、图片下载完成）
update_main_params.append(
(json.dumps(urlMap), 3, qid))
except Exception as ex:
rs = False
logger.exception(
u"处理图片流程，qi=%s，校验题目的所有图片下载是否完成发生异常,异常信息：%s" %
(qid, ex.message))
return rs
if update_main_params:
postgreSql.batchExecute(update_main_url_sql,
update_main_params)
if update_image_params:
postgreSql.batchExecute(update_image_sql,
update_image_params)
if insert_convert_params:
postgreSql.batchExecute(insert_convert_sql,
insert_convert_params)
postgreSql.commit()
count += len(update_image_params)
logger.info(u'处理图片流程，已成功处理题目数量:%d，校验题目数量总数:%d' %
(count, total))
except Exception as e:
rs = False
postgreSql.rollback()
logger.exception(u"处理图片流程，批量处理-异常信息:%s" % (e.message))
finally:
postgreSql.close()
return rs