xx

README.md

@@ -0,0 +1,44 @@
+# xls2ddl
+
+a tool to create related database schema from xlsx file
+
+there is a xlsx template file called "model.xlst" in this repo
+
+## How to use it
+
+1 double click "model.xlst"
+2 in "sammary" datasheet, wirte the table name, table label, primary key
+3 in "fields" datasheet, create all the field in the table
+4 in "validation" datashhet identifies the index
+  first column is index name
+  second column choose the "idx"
+  thrid column for duplicatable index use "index:f1,f2 ..." or "unique:f1, f2, ..." for unique index
+
+5 save it with the tablenme as xlsx file's name
+6 repeat 1 to 5 for all the table.
+7 translates all xlsx file to ddl sql using
+in the folder hold all the xlsx file
+
+for mysql
+```
+python path/to/xls2ddl.py mysql .
+```
+for sqlite3
+```
+python path/to/xls2ddl.py sqlite3 .
+```
+for oracle
+```
+python path/to/xls2ddl.py oracle .
+```
+for postgresql
+```
+python path/to/xls2ddl.py postgresql .
+```
+for sqlserver
+```
+python path/to/xls2ddl.py sqlserver .
+```
+
+
+

requirements.txt

@@ -0,0 +1,4 @@
+xlrd
+openpyxl
+git+https://github.com/yumoqing/appPublic
+git+https://github.com/yumoqing/sqlor

xls2ddl

@@ -1 +0,0 @@
-Subproject commit a8be8655e9bde86fedc8a99a51f4e8c6de4697c4

xls2ddl.py

@@ -0,0 +1,67 @@
+# -*- coding:utf-8 -*-
+import io
+import sys
+from traceback import print_exc
+from xlsxData import CRUDData, xlsxFactory
+
+import codecs
+import json
+from sqlor.ddl_template_sqlserver import sqlserver_ddl_tmpl
+from sqlor.ddl_template_mysql import mysql_ddl_tmpl
+from sqlor.ddl_template_oracle import oracle_ddl_tmpl
+from sqlor.ddl_template_postgresql import postgresql_ddl_tmpl
+
+from appPublic.myTE import MyTemplateEngine
+from appPublic.folderUtils import listFile
+
+tmpls = {
+	"sqlserver":sqlserver_ddl_tmpl,
+	"mysql":mysql_ddl_tmpl,
+	"oracle":oracle_ddl_tmpl,
+	"postgresql":postgresql_ddl_tmpl
+}
+
+def xls2ddl(xlsfile,dbtype):
+	if xlsfile.endswith('json'):
+		with codecs.open(xlsfile,'r','utf-8') as f:
+			data = json.load(f)
+	else:
+		d = xlsxFactory(xlsfile)
+		if d is None:
+			return
+		data = d.read()
+	tmpl = tmpls.get(dbtype.lower())
+	if tmpl is None:
+		raise Exception('%s database not implemented' % dbtype)
+	e = MyTemplateEngine([])
+	s = e.renders(tmpl,data)
+	return s
+
+def model2ddl(folder,dbtype):
+	ddl_str = ''
+	for f in listFile(folder, suffixs=['xlsx','json']):
+		try:
+			ddl_str += f'-- {f}\n'
+			s = xls2ddl(f,dbtype)
+			ddl_str='%s%s' % (ddl_str, s)
+		except Exception as e:
+			print('Exception:',e,'f=',f)
+			print_exc()
+	return ddl_str
+
+if __name__ == '__main__':
+	import sys
+	##解决windows 终端中输出中文出现 
+	# UnicodeEncodeError: 'gbk' codec can't encode character '\xa0' in position 20249
+	# 错误
+	# BEGIN
+	sys.stdout = io.TextIOWrapper(sys.stdout.buffer,encoding='utf8')
+	#
+	# END
+	if len(sys.argv) < 3:
+		print('Usage:%s dbtype folder' % sys.argv[0])
+		sys.exit(1)
+
+	s = model2ddl(sys.argv[2], sys.argv[1])
+	print(s)
+

xlsxData.py

@@ -0,0 +1,287 @@
+import os
+import sys
+from openpyxl import load_workbook
+from appPublic.myjson import loadf,dumpf,dumps
+from appPublic.dictObject import DictObject
+	
+class TypeConvert:
+	def conv(self,typ,v):
+		if typ is None:
+			return v
+		f = getattr(self,'to_'+typ,None)
+		if f is None:
+			return v
+		return f(v)
+	
+	def to_int(self,v):
+		try:
+			return int(v)
+		except:
+			return 0
+	
+	def to_float(self,v):
+		try:
+			return float(v)
+		except:
+			return 0.0
+	
+	def to_str(self,v):
+		try:
+			return str(v)
+		except:
+			return ''
+	def to_json(self,v):
+		if v == '':
+			return v
+		try:
+			return loads(v)
+		except:
+			return v
+			
+	def to_date(self,v):
+		return v
+	
+	def to_time(self,v):
+		return v
+	
+	def to_timestamp(self,v):
+		return v
+
+	def to_cruddata(self,v):
+		vs = v.split('"',3)
+		if vs < 3:
+			return v
+		fn = vs[1]
+		d = CRUDData(fn)
+		try:
+			data = d.read()
+		except Exception as e:
+			return v
+		if vs[2] is None:
+			return data
+		cmd = "d%s" % vs[2]
+		ret=eval(cmd,{'d':data})
+		return ret
+	
+	def to_xlsxdata(self,v):
+		vs = v.split('"',3)
+		if vs < 3:
+			return v
+		fn = vs[1]
+		d = XLSXData(fn)
+		try:
+			data = d.read()
+		except Exception as e:
+			return v
+		if vs[2] is None:
+			return data
+		cmd = "d%s" % vs[2]
+		ret=eval(cmd,{'d':data})
+		return ret
+	
+class CRUDException(Exception):
+	def __init__(self,xlsfile,errmsg,*args,**argv):
+		Exception.__init__(self,*args,**argv)
+		self.xlsfile = xlsfile
+		self.errmsg = errmsg
+	
+	def __str__(self):
+		return 'filename:' + self.xlsfile+' error:' + self.errmsg
+
+class XLSXData(object):
+	def __init__(self,xlsxfile):
+		atype = type(xlsxfile) 
+		if atype == type('') or atype == type(u''):
+			self.xlsxfile = xlsxfile
+			self.book = load_workbook(filename=xlsxfile)
+		else:
+			self.book = xlsxfile # is from Factory
+		  
+	def readRecords(self,name,sheet):
+		i = 1
+		recs = []
+		fields = []
+		tc = TypeConvert()
+
+		for i,row in enumerate(sheet.values):
+			if i==0:
+				fields = self.getFieldNames(row)
+				continue
+			rec = {}
+			for j, a in enumerate(row):
+				if a is None:
+					continue
+				k = fields[j][0]
+				v = tc.conv(fields[j][1],a)
+				rec[k] = v
+			if rec == {}:
+				continue
+			o = DictObject(**rec)
+			recs.append(o)
+		return {name:recs}
+	 
+	def read(self):
+		ret = {}
+		for i,s in enumerate(self.book.worksheets):
+			ret.update(self.readRecords(self.book.sheetnames[i], s))
+		return DictObject(**ret)
+
+	def getFieldNames(self,row):
+		fs = []
+		for i,f in enumerate(row):
+			if f is None:
+				f = 'F_' + str(i)
+			else:
+				if type(f) != type(u""):
+					f = 'F_' + str(f)
+				"""
+				else:
+					f = f.encode('utf-8')
+				"""
+			b=f.split(':')
+			if len(b) < 2:
+				b.append(None)
+			fs.append(b)
+			i+= 1
+		return fs
+	  
+class CRUDData(XLSXData):
+	@classmethod
+	def isMe(self,book):
+		names = book.sheetnames
+		if 'summary' not in names:
+			return False
+		if 'fields' not in names:
+			return False
+		if 'validation' not in names:
+			return False
+		return True
+	
+	def read(self):
+		d = XLSXData.read(self)
+		if not 'summary' in d.keys():
+			raise CRUDException(self.xlsxfile,'summary sheet missing')
+		if not 'fields' in d.keys():
+			raise CRUDException(self.xlsxfile,'fields sheet missing')
+		if not 'validation' in d.keys():
+			raise CRUDException(self.xlsxfile,'validation sheet missing')
+			
+		v = d['summary'][0]['primary']
+		v = v.split(',')
+		d['summary'][0]['primary'] = v
+		d = self.convForeignkey(d)
+		d = self.convIndex(d)
+		return d
+  
+	def convForeignkey(self,data):
+		vs = data['validation']
+		nvs = []
+		for v in vs:
+			if v['oper'] == 'fk':
+				m = v['value']
+				des= m.split(':')
+				if len(des) != 3:
+					raise CRUDException(self.xlsxfile,'fk value error:%s' % m)
+				v['value'] = {'table':des[0],'value':des[1],'title':des[2]}
+			nvs.append(v)
+		data['validation'] = nvs
+		return data
+ 
+	def getFieldByNmae(self,fields,name):
+		for f in fields:
+			if f['name'] == name:
+				return f
+
+	def getFKs(self,validation):
+		fks = []
+		for v in validation:
+			if v['oepr'] == 'fk':
+				fks.append(v)
+		return fks
+
+	def getIDXs(self,validation):
+		idxs = []
+		for v in validation:
+			if v['oper'] == 'idx':
+				idxs.append(v)
+		return idxs
+
+	def isFieldExist(self,fields,name):
+		for f in fields:
+			if f['name'] == name:
+				return True
+		return False
+
+	def convIndex(self,data):
+		vs = data['validation']
+		nvs = []
+		for v in vs:
+			if v['oper'] == 'idx':
+				idx = {}
+				idx['name'] = v['name']
+				m = v['value']
+				des= m.split(':')
+				if len(des) != 2:
+					raise CRUDException(self.xlsxfile,'idx value format:idx_type:keylist:%s' % m)
+				idx['idxtype'] = des[0]
+				idx['idxfields'] = des[1].split(',')
+				nvs.append(idx)
+		data['indexes'] = nvs
+		return data
+
+def xlsxFactory(xlsxfilename):
+	def findSubclass(name,klass):
+		for k in klass.__subclasses__():
+			if k.isMe(name):
+				return k
+			k1 = findSubclass(name,k)
+			if k1 is not None:
+				return k1
+			return None
+	try:
+		book = load_workbook(filename=xlsxfilename)
+		k = findSubclass(book,XLSXData)
+		if k is not None:
+			xlsx = k(book)
+			xlsx.xlsxfile = xlsxfilename
+			return xlsx
+		return XLSXData(book)
+
+	except Exception as e:
+		# print(xlsxfilename, 'load failed\n%s' % str(e))
+		return None
+
+def ValueConvert(s):
+	if s[:9] == 'xlsfile::':
+		d = xlsxFactory(s[9:])
+		return d.read()
+	if s[:10] == 'jsonfile::':
+		return loadf(s[10:])
+	return s
+	
+def paramentHandle(ns):
+	for k,v in ns.items():
+		ns[k] = ValueConvert(v)
+	return ns
+
+if __name__ == '__main__':
+	retData = {}
+	ns = {}
+	datafiles = []
+	for a in sys.argv[1:]:
+		m = a.split('=',1)
+		if len(m)>1:
+			ns[m[0]] = m[1]
+		else:
+			datafiles.append(a)
+	
+	ns = paramentHandle(ns)
+	for f in datafiles:
+		ext = os.path.splitext(f)[-1]
+		if ext in ['.xlsx','.xls' ]:
+			d = xlsxFactory(f)
+			data = d.read()
+			retData.update(data)
+	retData.update(ns)
+	print( dumps(retData))
+