def _get_report_data(year, quarter, pageNo, dataArr, retry_count=3, pause=0.001): ct._write_console() for _ in range(retry_count): time.sleep(pause) try: request = Request(ct.REPORT_URL%(ct.P_TYPE['http'], ct.DOMAINS['vsf'], ct.PAGES['fd'], year, quarter, pageNo, ct.PAGE_NUM[1])) text = urlopen(request, timeout=10).read() text = text.decode('GBK') text = text.replace('--', '') html = lxml.html.parse(StringIO(text)) res = html.xpath("//table[@class=\"list_table\"]/tr") if ct.PY3: sarr = [etree.tostring(node).decode('utf-8') for node in res] else: sarr = [etree.tostring(node) for node in res] sarr = ''.join(sarr) sarr = '<table>%s</table>'%sarr df = pd.read_html(sarr)[0] df = df.drop(11, axis=1) df.columns = ct.REPORT_COLS dataArr = dataArr.append(df, ignore_index=True) nextPage = html.xpath('//div[@class=\"pages\"]/a[last()]/@onclick') if len(nextPage)>0: pageNo = re.findall(r'\d+', nextPage[0])[0] return _get_report_data(year, quarter, pageNo, dataArr) else: return dataArr except Exception as e: pass raise IOError(ct.NETWORK_URL_ERROR_MSG)
def _today_ticks(symbol, tdate, pageNo, retry_count, pause): ct._write_console() for _ in range(retry_count): time.sleep(pause) try: html = lxml.html.parse( ct.TODAY_TICKS_URL % (ct.P_TYPE['http'], ct.DOMAINS['vsf'], ct.PAGES['t_ticks'], symbol, tdate, pageNo)) res = html.xpath('//table[@id=\"datatbl\"]/tbody/tr') if ct.PY3: sarr = [etree.tostring(node).decode('utf-8') for node in res] else: sarr = [etree.tostring(node) for node in res] sarr = ''.join(sarr) sarr = '<table>%s</table>' % sarr sarr = sarr.replace('--', '0') df = pd.read_html(StringIO(sarr), parse_dates=False)[0] df.columns = ct.TODAY_TICK_COLUMNS df['pchange'] = df['pchange'].map(lambda x: x.replace('%', '')) except Exception as e: print(e) else: return df raise IOError(ct.NETWORK_URL_ERROR_MSG)
def _parsing_dayprice_json(types=None, page=1): """ 处理当日行情分页数据,格式为json Parameters ------ pageNum:页码 return ------- DataFrame 当日所有股票交易数据(DataFrame) """ ct._write_console() request = Request( ct.SINA_DAY_PRICE_URL % (ct.P_TYPE['http'], ct.DOMAINS['vsf'], ct.PAGES['jv'], types, page)) text = urlopen(request, timeout=10).read() if text == 'null': return None # reg = re.compile(r'\,(.*?)\:') # text = reg.sub(r',"\1":', text.decode('gbk') if ct.PY3 else text) # text = text.replace('"{symbol', '{"symbol') # text = text.replace('{symbol', '{"symbol"') if ct.PY3: text = text.decode('UTF8') jstr = json.dumps(text) else: jstr = json.dumps(text, encoding='GBK') js = json.loads(jstr) df = pd.DataFrame(pd.read_json(js, dtype={'code': object}), columns=ct.DAY_TRADING_COLUMNS) df = df.drop('symbol', axis=1) # df = df.ix[df.volume > 0] return df
def _get_stock_hq_list(pageNo, dataArr): ct._write_console() try: #param:["hq","hs_a","{sort}",{asc},{page},{num}] hq_list_param = '["hq","hs_a","",0,%d,%d]'%(pageNo,ct.OPEN_API_PAGE_NUM) request = Request(ct.SINA_OPEN_API_URL%(quote(hq_list_param,',[]'))) text = urlopen(request, timeout=10).read() text = text.decode('gbk') if ct.PY3 else text js = json.loads(text.strip()) if js is None: return dataArr df = pd.DataFrame(js[0]['items'], columns=js[0]['fields']) dataArr = dataArr.append(df, ignore_index=True) if int(js[0]['count']) > pageNo * ct.OPEN_API_PAGE_NUM : pageNo = pageNo+1 return _get_stock_hq_list(pageNo, dataArr) else: return dataArr except Exception as e: print(e)
def _get_growth_data(year, quarter,pageNo,dataArr): ct._write_console() try: #param:["cnl","行业","地域","概念","年","季度","{sort}",{asc},{page},{num}] list_param = '["cnl","","","","%d","%d","",0,%d,%d]'%(year, quarter,pageNo,ct.OPEN_API_PAGE_NUM) request = Request(ct.SINA_OPEN_API_URL%(quote(list_param,',[]'))) text = urlopen(request, timeout=10).read() text = text.decode('gbk') if ct.PY3 else text js = json.loads(text.strip()) if js is None: return dataArr df = pd.DataFrame(js[0]['items'],columns=ct.GROWTH_COLS) dataArr = dataArr.append(df, ignore_index=True) if int(js[0]['count']) > pageNo * ct.OPEN_API_PAGE_NUM : pageNo = pageNo+1 return _get_growth_data(year, quarter,pageNo, dataArr) else: return dataArr except Exception as e: print(e)
def get_h_data(code, start=None, end=None, autype='qfq', index=False, retry_count=3, pause=0.001, drop_factor=True): ''' 获取历史复权数据 Parameters ------ code:string 股票代码 e.g. 600848 start:string 开始日期 format:YYYY-MM-DD 为空时取当前日期 end:string 结束日期 format:YYYY-MM-DD 为空时取去年今日 autype:string 复权类型,qfq-前复权 hfq-后复权 None-不复权,默认为qfq retry_count : int, 默认 3 如遇网络等问题重复执行的次数 pause : int, 默认 0 重复请求数据过程中暂停的秒数,防止请求间隔时间太短出现的问题 drop_factor : bool, 默认 True 是否移除复权因子,在分析过程中可能复权因子意义不大,但是如需要先储存到数据库之后再分析的话,有该项目会更加灵活 return ------- DataFrame date 交易日期 (index) open 开盘价 high 最高价 close 收盘价 low 最低价 volume 成交量 amount 成交金额 ''' start = du.today_last_year() if start is None else start end = du.today() if end is None else end qs = du.get_quarts(start, end) qt = qs[0] ct._write_head() data = _parse_fq_data(_get_index_url(index, code, qt), index, retry_count, pause) if data is None: data = pd.DataFrame() if len(qs) > 1: for d in range(1, len(qs)): qt = qs[d] ct._write_console() df = _parse_fq_data(_get_index_url(index, code, qt), index, retry_count, pause) if df is None: # 可能df为空,退出循环 break else: data = data.append(df, ignore_index=True) if len(data) == 0 or len( data[(data.date >= start) & (data.date <= end)]) == 0: return pd.DataFrame() data = data.drop_duplicates('date') if index: data = data[(data.date >= start) & (data.date <= end)] data = data.set_index('date') data = data.sort_index(ascending=False) return data if autype == 'hfq': if drop_factor: data = data.drop('factor', axis=1) data = data[(data.date >= start) & (data.date <= end)] for label in ['open', 'high', 'close', 'low']: data[label] = data[label].map(ct.FORMAT) data[label] = data[label].astype(float) data = data.set_index('date') data = data.sort_index(ascending=False) return data else: if autype == 'qfq': if drop_factor: data = data.drop('factor', axis=1) df = _parase_fq_factor(code, start, end) df = df.drop_duplicates('date') df = df.sort_values('date', ascending=False) firstDate = data.head(1)['date'] frow = df[df.date == firstDate[0]] rt = get_realtime_quotes(code) if rt is None: return pd.DataFrame() if ((float(rt['high']) == 0) & (float(rt['low']) == 0)): preClose = float(rt['pre_close']) else: if du.is_holiday(du.today()): preClose = float(rt['price']) else: if (du.get_hour() > 9) & (du.get_hour() < 18): preClose = float(rt['pre_close']) else: preClose = float(rt['price']) rate = float(frow['factor']) / preClose data = data[(data.date >= start) & (data.date <= end)] for label in ['open', 'high', 'low', 'close']: data[label] = data[label] / rate data[label] = data[label].map(ct.FORMAT) data[label] = data[label].astype(float) data = data.set_index('date') data = data.sort_index(ascending=False) return data else: for label in ['open', 'high', 'close', 'low']: data[label] = data[label] / data['factor'] if drop_factor: data = data.drop('factor', axis=1) data = data[(data.date >= start) & (data.date <= end)] for label in ['open', 'high', 'close', 'low']: data[label] = data[label].map(ct.FORMAT) data = data.set_index('date') data = data.sort_index(ascending=False) data = data.astype(float) return data