def parse_schema_from_rows(self, rows):
shema = []
column_names = []
for row in rows:
item = {'name': row[0], 'type': row[1], 'comment': row[-1]}
if item['name'] in column_names:
raise ParseSchemaError('%s column has the same name' %
item['name'])
column_names.append(item['name'])
keyword = reg_keyword(item['comment'])
ctype = reg_cmd(item['type'])
if not keyword:
keyword = item['type']
cmd = reg_cmd(keyword) if keyword else ctype
rets = reg_args(keyword)
if cmd == 'enum' or cmd == 'order_enum':
if len(rets) == 0:
raise EnumMustNotEmptyError
# 如果enum类型只有一个值，则产生固定值
# 如果enum类型只有一个值，且以file://开头，则读取文件
if len(rets) == 1 and rets[0].startswith(ENUM_FILE):
rets = read_file_lines(rets[0][len(ENUM_FILE):])
if ctype in INT_TYPES:
args = [int(ret) for ret in rets]
elif ctype in FLOAT_TYPES:
args = [float(ret) for ret in rets]
else:
args = rets
elif cmd in INT_TYPES:
args = [int(ret) for ret in rets]
args.append(True) if 'unsigned' in keyword else args.append(
False)
elif cmd == 'op':
args = [
process_op_args(rets[0], 'columns'),
]
else:
try:
args = [int(ret) for ret in rets]
except:
args = rets
item['cmd'] = cmd
item['ctype'] = ctype
item['args'] = args
shema.append(item)
return shema
def construct_meta_rows(self):
"""
元数据文件中每行中每个字段以||分割，一共有三列：
第一列表示：字段名
第二列表示：字段类型
第三列表示：带标记的字段注释
:return:
"""
filepath = self.args.meta
lines = read_file_lines(filepath)
rows = []
for line in lines:
words = line.split("||")
if len(words) != 3:
raise ParseSchemaError('parse schema error, %s' % line)
rows.append([word.strip() for word in words])
return rows
def parse_schema_from_text(self, text):
keywords = reg_all_keywords(text)
schema = []
for keyword in keywords:
cmd = reg_cmd(keyword)
rets = reg_args(keyword)
if cmd == 'enum' or cmd == 'order_enum':
if len(rets) == 0:
raise EnumMustNotEmptyError
# 如果enum类型只有一个值，则产生固定值
# 如果enum类型只有一个值，且以file://开头，则读取文件
if len(rets) == 1 and rets[0].startswith(ENUM_FILE):
args = read_file_lines(rets[0][len(ENUM_FILE):])
else:
# 枚举全部当做字符类型
args = rets
elif cmd in INT_TYPES:
args = [int(ret) for ret in rets]
args.append(True) if 'unsigned' in keyword else args.append(
False)
elif cmd == 'op':
args = [
process_op_args(rets[0], 'columns'),
]
else:
try:
args = [int(ret) for ret in rets]
except:
args = rets
item = {'cmd': cmd, 'args': args}
schema.append(item)
return schema
def test_read_file_lines(self):
filepath = "__init__.py"
read_file_lines(filepath)