ValueError: unsupported format character in mysql scrapy pipline

Question

I am working on a scrapy crawler, and this issue really bothers me since I already been trapped by this for days.

This placeholders function works fine when I using "?" instead of "%s" for SQLite db. But while using "?" as the database switching to MySQL, it shows:

"TypeError: not all arguments converted during string formatting "

even I pay lots of effort modifying codes and changing the placeholder(supposedly?) it still shows :

" query = query % self._escape_args(args, conn) ValueError: unsupported format character ',' "

more specifically :

Traceback (most recent call last):

  File "/usr/lib64/python3.4/dist-packages/twisted/internet/defer.py", line 653, in _runCallbacks
    current.result = callback(current.result, *args, **kw)

File "/home/ec2-user/lulu_testing/get_download_file/hello_scrapy/hello/hello/pipelines.py", line 42, in process_item
    self.cur.execute(insert_query, insert_values)

File "/usr/lib/python3.4/dist-packages/pymysql/cursors.py", line 163, in execute
    query = self.mogrify(query, args)

File "/usr/lib/python3.4/dist-packages/pymysql/cursors.py", line 142, in mogrify
    query = query % self._escape_args(args, conn)

ValueError: unsupported format character ',' (0x2c) at index 94

the pipline for mysql version

import pymysql
import scrapy
from hello.items import HelloItem

class HelloPipeline(object):

def __init__(self):#

    self.conn = pymysql.connect(host="localhost", port=3306, user="root", passwd="lulu", db="test", charset="utf8", use_unicode=True)
    self.cur = self.conn.cursor()


    self.cur.execute("drop table IF EXISTS test;")
    self.conn.commit()

    self.cur.execute("create table if not EXISTS table_test_4(test0 text, test1 text, test2 text, test3 text,test4 text, test5 text, test6 text, test7 text, test8 text, test9 text);")
    self.conn.commit()

    #pass


def process_item(self, item, spider):#

    col = ",".join(item.keys())
    placeholders = ",".join(len(item) * "%s")

    insert_query = "INSERT INTO test_table_4({0}) VALUES({1});".format(col,placeholders)

    insert_values = tuple(item.values())


    self.cur.execute(insert_query, insert_values)

    return item

def close_spider(self, spider):#
    self.cur.close()
    self.conn.close()
    #pass

the SQLite version(what i was using b4)

import sqlite3
import scrapy
from hello.items import HelloItem

class HelloPipeline(object):

def open_spider(self, spider):#
    self.conn = sqlite3.connect("test_database_ver_2018_03_31.sqlite")
    self.cur = self.conn.cursor()
    self.cur.execute("create table if not exists test_table(test0 text, test1 text, test2 text, test3 text,test4 text, test5 text, test6 text, test7 text, test8 text, test9 text);")        
    #pass

def close_spider(self, spider):#
    self.conn.commit()
    self.conn.close()
    #pass

def process_item(self, item, spider):#

    col = ",".join(item.keys())       
    placeholders = ",".join(len(item) * "?")
    sql = "insert into test_table({}) values({})"


    self.cur.execute(sql.format(col, placeholders), tuple(item.values()))


    return item

data settings of main scrapy crawler program

    testitem = HelloItem()

    testitem["test0"] = house_detail.select(".houseInfoTitle")[0].text        
    testitem["test1"] = house_detail.select(".pageView")[0].text 
    testitem["test2"] = house_detail.select(".detailInfo")[0].text
    testitem["test3"] = house_detail.select(".houseIntro")[0].text
    testitem["test4"] = house_detail.select(".lifeBox")[0].text
    testitem["test5"] = house_detail.select(".labelList")[0].text
    testitem["test6"] = house_detail.select(".facility")[0].text
    testitem["test7"] = str(house_detail.select(".userInfo"))
    testitem["test8"] = str(house_detail.select(".banner"))
    testitem["test9"] = str(house_detail.select("#show"))

    return testitem

item setting

import scrapy


class HelloItem(scrapy.Item):

test0 = scrapy.Field()
test1 = scrapy.Field()
test2 = scrapy.Field()
test3 = scrapy.Field()
test4 = scrapy.Field()
test5 = scrapy.Field()
test6 = scrapy.Field()
test7 = scrapy.Field()
test8 = scrapy.Field()
test9 = scrapy.Field()

Answer 1

The problem is that this line:

placeholders = ",".join(len(item) * "%s")

isn't doing what you expect.

>>> item = {'a': 1, 'b': 2, 'c': 3}
>>> placeholders = ",".join(len(item) * "%s")
>>> print(placeholders)
%,s,%,s,%,s     

",".join(len(item) * "%s") does two things - computes len(item) * "%s" , then joins the result with ',' .

The result of len(item) * '%s' is the string (or iterable ) '%s%s%s' . str.join(iterable) returns a string consisting of all the elements of iterable separated by the string that provides the method . So the result of calling

','.join('%s%s%s') is

'%,s,%,s,%,s' , not '%s,%s,%s'

You want to do

>>> ",".join(len(item) * ["%s"])
'%s,%s,%s'

or

>>> ",".join('%s' for _ in item)
'%s,%s,%s'

so that str.join is operating on an iterable of '%s' strings rather than a single string like '%s%s%s' . `