Skip to main content

grs 台灣上市上櫃股票價格擷取

https://travis-ci.org/toomore/grs.svg?branch=master https://img.shields.io/pypi/dm/grs.svg?maxAge=2592000 https://img.shields.io/pypi/v/grs.svg?maxAge=2592000 https://img.shields.io/pypi/wheel/grs.svg?maxAge=2592000 https://img.shields.io/pypi/l/grs.svg?maxAge=2592000

主要開發擷取台灣股市(TWSE)股價資料

版本資訊

Authors:

Toomore Chiang

Version:

0.7.0 of 2016/09/11

Python Version:

Python 2.7

Docs:

http://grs-docs.toomore.net/

Requires

  • python-dateutil==1.5

  • ujson

  • urllib3

Report Issue or get involved

Web Demo

Quick Start

簡單計算

from grs import Stock

stock = Stock('2618')                         # 擷取長榮航股價
print stock.moving_average(5)                 # 計算五日均價與持續天數
print stock.moving_average_value(5)           # 計算五日均量與持續天數
print stock.moving_average_bias_ratio(5, 10)  # 計算五日、十日乖離值與持續天數

擷取 12 個月份資料

stock = Stock('2618', 12)

輸出 CSV 檔

stock.out_putfile('/dev/shm/2618.csv')

擷取上櫃股價資訊

from grs import Stock

stock = Stock('8446')                         # 擷取華研股價
print stock.moving_average(5)                 # 計算五日均價與持續天數
print stock.moving_average_value(5)           # 計算五日均量與持續天數
print stock.moving_average_bias_ratio(5, 10)  # 計算五日、十日乖離值與持續天數

如果已確定該代碼為上市或上櫃股票,可以直接指定參數跳過查表動作。

stock = Stock('2618', twse=True) # 擷取長榮航股價
stock = Stock('8446', otc=True)  # 擷取華研股價

其他功能

顯示台灣時間:TWTime

適用於其他時區查詢台灣當地時間。

from grs import TWTime

what_time = TWTime()
what_time.now()        # 顯示台灣此刻時間
what_time.localtime()  # 顯示當地此刻時間

判斷台灣股市是否開市:TWSEOpen

from grs import TWSEOpen
from datetime import datetime

open_or_not = TWSEOpen()

open_or_not.d_day(datetime.today())        # 判斷今天是否開市
                                           # 回傳 True or False
open_or_not.d_day(datetime(2012, 12, 22))  # 判斷 2012/12/22 是否開市

各股即時盤資訊:RealtimeTWSE / RealtimeOTC

上市即時資訊

from grs import RealtimeTWSE

realtime_stock = RealtimeTWSE('2618')   # 擷取長榮航即時股價
realtime_stock.raw                      # 原始資料
realtime_stock.data                     # 回傳 type: dict

上櫃即時資訊

from grs import RealtimeOTC

realtime_stock = RealtimeOTC('8446')    # 擷取華研即時股價
realtime_stock.raw                      # 原始資料
realtime_stock.data                     # 回傳 type: dict

大盤即時盤資訊:RealtimeWeight(加權指數、櫃檯指數、寶島指數)

from grs import RealtimeWeight

realtime_weight = RealtimeWeight()  # 擷取即時大盤資訊
realtime_weight.raw                 # 原始檔案
realtime_weight.data                # 回傳 type: dict

上市股票代碼列表:TWSENo

回傳上市股票代碼與搜尋

from grs import TWSENo


twse_no = TWSENo()
twse_no.all_stock       # 所有股票名稱、代碼 type: dict
twse_no.all_stock_no    # 所有股票代碼 type: list
twse_no.all_stock_name  # 所有股票名稱 type: list
twse_no.industry_code   # 回傳類別代碼 type: dict
twse_no.industry_comps  # 回傳類別所屬股票代碼 type: dict
twse_no.search(u'中')   # 搜尋股票名稱,回傳 type: dict
twse_no.searchbyno(23)  # 搜尋股票代碼,回傳 type: dict
twse_no.last_update     # 回傳列表最後更新時間(非同步)type: str

單日倒數時間:Countdown

適用於設定 cache 時間。

from grs import Countdown

countdown = Countdown(hour=14, minutes=30)  # 預設為 14:30
countdown.nextday    # 下一個 14:30 日期
countdown.countdown  # 到數秒數
countdown.exptime    # 下一個 14:30 日期時間(type: datetime)
countdown.lastmod    # 前一個 14:30 日期時間(type: datetime)

判斷乖離轉折點:Stock(no).check_moving_average_bias_ratio

判斷乖離轉折點

from grs import Stock

stock = Stock('2618')
data = stock.moving_average_bias_ratio(3, 6)[0]  # 取得 3-6 乖離值 type: list

# 計算五個區間負乖離轉折點
check_data = stock.check_moving_average_bias_ratio(data, sample=5,
                                                positive_or_negative= False)
print check_data  # (T/F, 第幾轉折日, 乖離轉折點值) type: tuple

四大買賣點判斷:BestFourPoint

判斷是否為技術分析的四大買賣點,條件成立,回傳條件結果,判斷結果僅供參考!

from grs import BestFourPoint
from grs import Stock

stock = Stock('2618')
result = BestFourPoint(stock)
result.best_four_point_to_buy()       # 判斷是否為四大買點
result.best_four_point_to_sell()      # 判斷是否為四大賣點
result.best_four_point()              # 綜合判斷

全部上市股票檢視

from grs import BestFourPoint
from grs import Stock
from grs import TWSENo

stock_no_list = TWSENo().all_stock_no

for i in stock_no_list:
    try:
        best_point, info = BestFourPoint(Stock(i)).best_four_point()
        if best_point:  # 買點
            print 'Buy: {0} {1}'.format(i, info)
        else:   # 賣點
            print 'Sell: {0} {1}'.format(i, info)
    except:     # 不作為或資料不足
        print 'X: {0}'.format(i)

擴充月份資料:Stock(no).plus_mons(month)

當原有的月份資料不夠時,不需要從頭抓取,只需要給予增額月份值即可。

from grs import Stock

stock = Stock('2618')                # 預設為抓取3個月份資料
stock.moving_average(60)
IndexError: list index out of range  # 資料不足
len(stock.raw)                       # 回傳 51 個值
stock.plus_mons(1)                   # 在抓取一個月資料
len(stock.raw)                       # 回傳 66 個值
stock.moving_average(60)             # 計算成功

Change Logs

0.7.0 2016/09/11

  • 修正:上市(TWSE)擷取連結

0.6.2 2015/03/20

  • 修正:上櫃(OTC)擷取連結

0.6.1 2014/06/11

  • 修正:安裝時錯誤的套件載入

0.6.0 2014/06/10

  • 修正:使用 urllib3 取代 urllib2

  • 新增:新格式的即時盤擷取資訊,包含加權指數、櫃檯指數、寶島指數

0.5.6 2014/06/01

  • 修正:tools 儲存路徑

  • 新增:日常交易的代碼與名稱(grs.twseno.ImportCSV.get_stock_list)

  • 新增:日常交易的類別代碼與名稱(grs.twseno.ImportCSV.get_stock_comps_list)

  • 已知問題:盤中即時資訊擷取無法使用 grs.RealtimeStock/RealtimeWeight

0.5.5 2014/05/18

  • 修正: grs.fetch_data.SimpleAnalytics.CKMAO to be classmethod.

0.5.4 2014/05/12

  • 新增:MA, MAO, MAV, CKMAO into grs.fetch_data.SimpleAnalytics.

0.5.3 2014/04/17

  • 修正:離線時的錯誤訊息

  • 修正:realtime str format.

0.5.2 2014/04/12

  • 修正:字串判斷使用 basestring.

0.5.1 2014/04/08

  • 修正:套件遺漏 csv 檔案

0.5.0 2014/03/04

  • 新增:上櫃資訊( 櫃台買賣中心 )

  • 修正:股票代碼列表回傳(TWSENo)代碼值改為 string.

0.4.3 2014/01/22

0.4.2 2014/01/11

  • 修正:Stock stock_no, RealtimeStock no 必須為 string. Issues #9

0.4.1 2014/01/02

  • 修正:Countdown().countdown 秒數問題

  • 新增:twse_no, twse_open, twse_realtime, countdown into unittest

  • 移除:Support Python 2.6

0.4.0 2013/12/30

  • 修正:Naming Convention

  • 修正:Coding style to fit PEP8

  • 新增:For PyPy

0.3.0 2013/12/18

  • 更新:股票代碼列表

  • 更新:2014 年集中交易市場開(休)市日期表

0.2.1 2013/12/16

  • 修正:部分資料改用 tuple

0.2.0 2012/04/13

  • 修正:輸出中文統一使用 Unicode

  • 修正:需要套件 python-dateutil 調整為 1.5

  • 修正:Web Demo 網站網址

  • 新增:Stock.plusMons() 擴充月份資料

0.1.4 2012/04/01

  • 修正:每月首日無資料抓取問題

0.1.3 2012/03/31

  • 修正:Countdown 倒數時間計算錯誤(dateutil.relativedelta)

0.1.2 2012/03/31

  • 修正:grs 倒數時間計算錯誤(dateutil.relativedelta)

Metadata

Release files for grs 0.7.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for grs 0.7.0
File Size Uploaded Provenance
grs-0.7.0.tar.gz 96.1 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for grs 0.7.0
File Interpreter ABI Platform
grs-0.7.0-py2-none-any.whl Python 2 none any Details

Total release size: 206.1 kB

Release files / grs-0.7.0.tar.gz

Download URL grs-0.7.0.tar.gz
Size 96.1 kB
Tags Source
SHA-256 checksum
How to use checksums
be95feff86d624468bc5ab5573bd7737b398d7500d324cf2314412ecbd58505a
BLAKE2b-256 checksum
How to use checksums
46f81deaa8cc07dcd0d946a9b0767f7920fe0b9706a0dbd41ec82f73b280715d
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No

Release files / grs-0.7.0-py2-none-any.whl

Download URL grs-0.7.0-py2-none-any.whl
Size 110.1 kB
Tags Python 2
SHA-256 checksum
How to use checksums
f07e5d1a1f72254c420b9cf203b09a9410e69fdb45104ee3808bf9a7e1d065ec
BLAKE2b-256 checksum
How to use checksums
c0e455d5ae7d11d94c3f656a57e67a2eb2b8293e9d0ab33faa3e6d8b3ca4de3c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No

Security for grs 0.7.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page