今天寫代碼時,編譯不過去,最後發現,配對符寫丟了一個:
import sys
import re
import codecs
class TestCls(object):
def __init__(self):
return None
def ReadSqlFile(self, filename, encoding):
with codecs.open(filename=filename, mode='r', encoding=encoding) as f:
content = f.read()
# sql文件,以"--"作爲註釋符(暫不支持其他註釋符),解析文件之前,先把註釋替換掉,否則影響解析
# 正則表達式,零寬斷言,零寬度正預測先行斷言
content = re.sub(r"--.*(?=[\r\n]+?)", "", content)
# 以"CREATE TABLE IF NOT EXISTS"作爲建表的開始標誌,以");"作爲結束標誌
patternTable = "(" + """CREATE TABLE IF NOT EXISTS[ \t]+""" + """(?P<tableName>[a-zA-Z0-9_]+)""" + """[ \t\r\n]*\(""" + """(?P<allFieldContent>[^;]+)""" + """\);""" + ")"
# 定義字段時必須要帶"NULL"或"NOT NULL",否則解析不出來,例:
# [colName] [colType] [NULL 或 NOT NULL] [後續字段(PRIMARY KEY等)]
patternField = """([ \t]*(?P<colName>[a-zA-Z0-9_]+)[ \t]+(?P<colType>[a-zA-Z0-9\(\)]+)[ \t]+((NULL)|(NOT NULL))(?P<otherStr>.*))"""
tableContentList = re.findall(patternTable, content)
for tableContent in list(zip(*tableContentList))[0]:
sreMatch = re.match(patternTable, tableContent)
tableName = sreMatch.groupdict()["tableName"]
allFieldContent = sreMatch.groupdict(["allFieldContent"]
#allFieldContent = sreMatch.groupdict()["allFieldContent"]
# 如果"()"等符號丟失了一個的話,會報錯出來很詭異的錯誤,此時應當意識到,可能丟失了配對符等.
# 此時,建議使用排除法,一點一點的註釋掉代碼,找到出錯的代碼塊,然後仔細查找原因.
fieldContentList = re.findall(patternField, allFieldContent)
for fieldContent in list(zip(*fieldContentList))[0]:
sreMatch = re.match(patternField, fieldContent)
colName = sreMatch.groupdict()["colName"]
colType = sreMatch.groupdict()["colType"]
otherStr = sreMatch.groupdict()["otherStr"]
print(colName)
print(colType)
print(otherStr)
return None
if __name__ == "__main__":
sqlStr = """
DROP TABLE IF EXISTS nnnnnn_etf_list;
CREATE TABLE IF NOT EXISTS nnnnnn_etf_list(
date INTEGER NOT NULL,--日期
code VARCHAR(32) NOT NULL,--code
field1 VARCHAR(64) NOT NULL,--字段1
field2 VARCHAR(64) NULL,--字段2
field3 VARCHAR(64) NULL,--字段3
field4 VARCHAR(64) NULL,--字段4
field5 VARCHAR(64) NULL,--字段5
field6 VARCHAR(64) NULL,--字段6
PRIMARY KEY(date,code,field1)
);"""
obj = TestCls()
fileName = r"D:\sql.sql"
obj.ReadSqlFile(fileName, 'utf8')
sys.exit(0)
完。