作者funningboy (funningboy)
看板Python
标题Re: [分享] Python练习-取得股票资讯
时间Tue Apr 2 14:22:49 2013
※ 引述《mantix (= =)》之铭言：
: ※ 引述《luke93 (不废江河万古流)》之铭言：
: : 标题: [分享] Python练习-取得股票资讯
: : 时间: Fri Apr 24 22:53:54 2009
: : """
小弟再来一发好了,底下用既有的套件,来开发自己的程式交易系统,
真的只要30分钟就好了!!!. 无须知道底层API怎麽运作,只要知道你自己的交易策略就好.
http://funningboy.blogspot.tw/2013/04/ta-lib-bigdata-stock-analysis-tool.html
: : 最近的练习，发现python真的满好用的
: : 改变stock_ids的内容就可以增减想看的股票
: : 一部分的code也是参考某位乡民的，感谢
: : 如果有人愿意提供什麽想法的话，非常感谢
: 最近做研究需要股票的资料，想说到底要怎麽取得
: 刚好看到原po 提供的code，弄了好久
: 终於改成可以抓google finance history 的版本
: 这里分享给大家
: -------------fetch_gf.py-OB-----------------
: #!/usr/bin/python
: ### fetch google finance historical data ##
: import httplib
: import urllib2
: import re
: def debug_print( s, msg = None):
: print "DEBUG", msg, s
: pass
: # get url
: # https://www.google.com/finance/historical?q=TPE:2330&startdate=1/31/1986&start=30&num=30
: def get_site( stkID = 2330, month = 1, day = 1, year = 1985, start = 0, num =
: 30):
: return 'https://www.google.com/finance/historical?q=TPE:' + \
: str( stkID ) + '&startdate=' + \
: str(day) + '/' + str(month) + '/' + str(year) + \
: '&start=' + str(start) + '&num=' + str(num)
: # testing
: #debug_print(get_site(2485, 1, 31, 1986, 0, 200))
: #debug_print(get_site())
: # target source:
: # date, open, high, low, close, volume
: #<td class="lm">Jun 8, 2012
: #<td class="rgt">77.00
: #<td class="rgt">78.70
: #<td class="rgt">77.00
: #<td class="rgt">77.90
: #<td class="rgt rm">34,628,000
: pattern = r'<td class="lm">(.+)' + \
: '\s<td class="rgt">([\d\.]+)' + \
: '\s<td class="rgt">([\d\.]+)' + \
: '\s<td class="rgt">([\d\.]+)' + \
: '\s<td class="rgt">([\d\.]+)' + \
: '\s<td class="rgt rm">([\d\,]+)'
: reg_price = re.compile( pattern )
: pattern_total_size =
: r'google\.finance\.applyPagination\(\s+\d+,\s+\d+,\s+(\d+),\s+'
: reg_row_size = re.compile( pattern_total_size )
: def write_fetch_data(fileName, stock_id = 0):
: f = open( fileName, 'a')
: if stock_id < 1000 :
: return None
: # Get web page content
: urlsite = get_site(stock_id, 1, 31, 1986, 0, 200)
: content = opener.open( urlsite ).read()
: match_r_sz = reg_row_size.search( content )
: if (match_r_sz is None):
: print "stock_id: "+ str(stock_id) + " not found! "
: return None
: ## write ID SZ line
: row_size = int (match_r_sz.groups()[0])
: id_line = "*ID:" + str(stock_id) + " SZ: " + str(row_size)
: print id_line
: f.write(id_line)
: f.write('\n')
: ## page number , 200rows/page
: page_num = row_size/200 + 1;
: print "pages: " + str(page_num)
: page_range = range(page_num)
: for i in page_range:
: start_pos = i*200
: site = get_site(stock_id, 1, 31, 1986, start_pos, start_pos+200)
: cnt = opener.open( site ).read()
: stock_data = reg_price.findall( cnt )
: length = len(stock_data)
: if (length is 0 ):
: print "stock_id: "+ str(stock_id) + " not found! "
: return None
: else:
: for i in range(length):
: f.write(str(stock_data[i]))
: f.write("\n")
: #### Main ####
: fileName = 'stock.data'
: f = open(fileName, 'w')
: f.close() ## rewrite data
: httplib.HTTPConnection.debuglevel = 1
: opener = urllib2.build_opener()
: opener.addheaders = [('User-agent', 'Mozilla/5.0')]
: #stock_ids = range(9999)
: stock_ids = (1111, 2330, 2485)
: print 'Start Fetch!'
: for stock_id in stock_ids: # fetch all
: write_fetch_data(fileName, stock_id)
--
※ 发信站: 批踢踢实业坊(ptt.cc)
◆ From: 36.224.38.125
1F：推 gozule:TA-lib真是神物~~ 04/03 01:58
2F：推 u9211008:这个模组看起来好像不错用,不过有没有办法在XP下安装呢?? 06/03 22:08