def deleteblank(readcsvname, writecsvname):
print 'deleteblank'
csvdata = web_data_process.read_csv(readcsvname)
finalList = []
for item in csvdata:
if len(item) != 1:
finalList.append(item)
print 'lenth', len(finalList)
web_data_process.write_in_csv(writecsvname, finalList)
def webSix(readcsvname, writecsvname):
print 'webFive'
csvdata = web_data_process.read_csv(readcsvname)
data = []
num = 0
for content in csvdata:
j = 0
for item in content:
item = item.decode('utf-8')
str = match1(item)
content[j] = str
# print 'item - x',num,j,item,x
j += 1
data.append(content)
num += 1
web_data_process.write_in_csv(writecsvname, data)
def deletebianhao(readcsvname1, readcsvname2):
print 'deletebianhao'
csvdata1 = web_data_process.read_csv(readcsvname1)
csvdata2 = web_data_process.read_csv(readcsvname2)
formulaList = []
functionList = []
for item in csvdata1:
item.pop(0)
formulaList.append(item)
for item in csvdata2:
item.pop(0)
functionList.append(item)
writecsvname1 = 'webFormula_final_2.csv'
writecsvname2 = 'webFunction_3.csv'
web_data_process.write_in_csv(writecsvname1, formulaList)
web_data_process.write_in_csv(writecsvname2, functionList)
def webFive(readcsvname, writecsvname):
print 'webFive'
csvdata = web_data_process.read_csv(readcsvname)
data = []
num = 0
for content in csvdata:
j = 0
for item in content:
item = item.decode('utf-8')
pos = item.find('去')
if pos > -1:
item = item[0:pos]
x = wordmatch(item)
# x = item.replace('炙', '')
# x = x.replace('不', '')
# x = x.replace('蒸', '')
# x = x.replace('炒', '')
# x = x.replace('熬', '')
# x = x.replace('锉', '')
# x = x.replace('炒香', '')
# x = x.replace('炮', '')
# x = x.replace('切', '')
# x = x.replace('轧细', '')
# x = x.replace('捣碎', '')
# x = x.replace('裹煨', '')
# x = x.replace('研粉', '')
# x = x.replace('调下', '')
# x = x.replace('另研', '')
# x = x.replace('碎绵裹', '')
# 通过正则表达去除多余的单位，只保留数值+g的单位。
str = match(x)
content[j] = str
# print 'item - x',num,j,item,x
j += 1
data.append(content)
num += 1
web_data_process.write_in_csv(writecsvname, data)
def UnifiedDose(readcsvname, writecsvname):
print 'UnifiedDose'
csvdata = web_data_process.read_csv(readcsvname)
normalList = []
for item in csvdata:
# print 'item',item
midList = []
for itemdata in item:
# print 'itemdata', itemdata
itemdata = itemdata.decode('utf8')
if (itemdata.find('两') > 0):
try:
zz = itemdata.split('两')
# print 'split itemdata', itemdata
unit = float(zz[0]) * 50
# print 'unit', unit
changeunit = str(unit) + 'g'
# print 'changeunit', changeunit
midList.append(changeunit)
except:
midList.append(itemdata)
elif (itemdata.find('钱') > 0):
try:
zz = itemdata.split('钱')
unit = float(zz[0]) * 3.125
# print 'unit',unit
changeunit = str(unit) + 'g'
midList.append(changeunit)
except:
midList.append(itemdata)
elif (itemdata.find('kg') > 0):
try:
zz = itemdata.split('kg')
unit = float(zz[0]) * 1000
changeunit = str(unit) + 'g'
midList.append(changeunit)
except:
midList.append(itemdata)
elif (itemdata.find('Kg') > 0):
try:
zz = itemdata.split('Kg')
unit = float(zz[0]) * 1000
changeunit = str(unit) + 'g'
midList.append(changeunit)
except:
midList.append(itemdata)
elif (itemdata.find('斤') > 0):
try:
zz = itemdata.split('斤')
unit = float(zz[0]) * 500
changeunit = str(unit) + 'g'
midList.append(changeunit)
except:
midList.append(itemdata)
elif (itemdata.find('分') > 0):
try:
zz = itemdata.split('斤')
unit = float(zz[0]) * 0.3
changeunit = str(unit) + 'g'
midList.append(changeunit)
except:
midList.append(itemdata)
else:
midList.append(itemdata)
normalList.append(midList)
web_data_process.write_in_csv(writecsvname, normalList)
def webProcessNum(readcsvname, writecsvname):
print 'webProcessNum'
csvdata = web_data_process.read_csv(readcsvname)
finaldata = []
num = 0
pattern3 = re.compile(
ur'\d+(?:g|kg|ml|l|个|Kg|钱|片|根|条|份|张|枚|具|朵|只|粒|茎|两|斤|挺|对|头|L|ML|分|节|cm|握|株)'
)
pattern4 = re.compile(
ur'\d+.\d+(?:g|kg|ml|Kg|l|个|钱|片|根|条|份|张|枚|具|朵|只|粒|茎|两|斤|挺|对|头|L|ML|分|节|cm|握|株)'
)
pattern1 = re.compile(ur'[\u4e00-\u9fa5]+(\(|（)')
pattern2 = re.compile(ur'(?:\(|（|）|\))')
pattern5 = re.compile(
ur'(?<![\u4e00-\u9fa5])(?:\(|（)\d+(?:g|kg|ml|l|个|Kg|钱|片|根|条|份|张|枚|具|朵|只|粒|茎|两|斤|挺|对|头|L|ML|分|节|cm|握|株)(?:\)|）)(?![\u4e00-\u9fa5])'
)
pattern6 = re.compile(ur'(?:\)|）)(?=[\u4e00-\u9fa5]+)')
pattern7 = re.compile(ur'(?:\(|（)')
for content in csvdata:
j = 0
medicallist = []
yaowulist = []
try:
content.remove('')
except:
pass
for item in content:
item = item.decode('utf-8')
if j == 0:
medicallist.append(content[0])
else:
#情况一：升麻 =》只有字
if not pattern3.search(item):
# print '1', item
word = item
wordnumber = 'None'
medicallist.append(word)
medicallist.append(wordnumber)
#情况三：（6g）=>只有数量
if pattern5.search(item):
# print '3',item
wordnumber = re.sub(pattern2, '', item)
k = 0
for item in medicallist:
if item == 'None':
medicallist[k] = wordnumber
k += 1
#情况五：处理‘各’：各(30g) or 焦栀各(各9g)
if item.find('各') > -1:
# print '各',item
item = item.replace('各', '')
item = re.sub(pattern2, '', item)
#找到整数 或者 小数
weight1 = pattern4.findall(item) #小数
weight2 = pattern3.findall(item) #整数
# 把正确的值放在变量weight中
if (weight1):
wordnumber = weight1[0]
yaowulist = pattern4.split(item)
elif (weight2):
wordnumber = weight2[0]
yaowulist = pattern3.split(item)
# 把处方的每味药提出来重新放在medicallist列表元素[0]里，同时已经去除了药的数量单位
if (yaowulist):
try:
for zz in yaowulist:
if zz != u'':
word = zz
# print 'word',word
medicallist.append(word)
medicallist.append(wordnumber)
except:
pass
k = 0
for item in medicallist:
if item == 'None':
medicallist[k] = wordnumber
k += 1
#情况二：麻黄（6g）
if pattern1.search(item):
# print '麻黄（6g）',item
word = item[0:pattern7.search(item).start()]
# print 'bb',word
wordnumber = item[pattern7.search(item).end():]
wordnumber = re.sub(pattern2, '', wordnumber)
# print 'nn',wordnumber
medicallist.append(word)
medicallist.append(wordnumber)
#情况四：（6g）麻黄
if pattern6.search(item):
wordnumber = item[:pattern6.search(item).start()]
wordnumber = re.sub(pattern2, '', wordnumber)
word = item[pattern6.search(item).end():]
medicallist.append(wordnumber)
medicallist.append(word)
j += 1
finaldata.append(medicallist)
num += 1
web_data_process.write_in_csv(writecsvname, finaldata)
def webFour(readcsvname, writecsvname):
print 'webFour'
data = web_dataDetailProcess.composition_process(readcsvname)
web_data_process.write_in_csv(writecsvname, data)