Merge pull request #1745 from ranaroussi/main

sync main -> dev

CHANGELOG.rst

@@ -1,6 +1,15 @@
 Change Log
 ===========
 
+0.2.32
+------
+Add cookie & crumb to requests #1657
+
+0.2.31
+------
+- Fix TZ cache exception blocking import #1705 #1709
+- Fix merging pre-market events with intraday prices #1703
+
 0.2.30
 ------
 - Fix OperationalError #1698

meta.yaml

@@ -1,5 +1,5 @@
 {% set name = "yfinance" %}
-{% set version = "0.2.31b1" %}
+{% set version = "0.2.32" %}
 
 package:
   name: "{{ name|lower }}"

tests/ticker.py

@@ -18,6 +18,8 @@
 import unittest
 import requests_cache
 from typing import Union, Any
+import re
+from urllib.parse import urlparse, parse_qs, urlencode, urlunparse
 
 ticker_attributes = (
     ("major_holders", pd.DataFrame),
@@ -76,7 +78,7 @@ def test_getTz(self):
         tkrs = ["IMP.JO", "BHG.JO", "SSW.JO", "BP.L", "INTC"]
         for tkr in tkrs:
             # First step: remove ticker from tz-cache
-            yf.utils.get_tz_cache().store(tkr, None)
+            yf.cache.get_tz_cache().store(tkr, None)
 
             # Test:
             dat = yf.Ticker(tkr, session=self.session)
@@ -181,14 +183,25 @@ def test_no_expensive_calls_introduced(self):
         will quickly trigger spam-block when doing bulk download of history data.
         """
         symbol = "GOOGL"
-        range = "1y"
+        period = "1y"
         with requests_cache.CachedSession(backend="memory") as session:
             ticker = yf.Ticker(symbol, session=session)
-            ticker.history(range)
-            actual_urls_called = tuple([r.url for r in session.cache.filter()])
+            ticker.history(period=period)
+            actual_urls_called = [r.url for r in session.cache.filter()]
+
+        # Remove 'crumb' argument
+        for i in range(len(actual_urls_called)):
+            u = actual_urls_called[i]
+            parsed_url = urlparse(u)
+            query_params = parse_qs(parsed_url.query)
+            query_params.pop('crumb', None)
+            query_params.pop('cookie', None)
+            u = urlunparse(parsed_url._replace(query=urlencode(query_params, doseq=True)))
+            actual_urls_called[i] = u
+        actual_urls_called = tuple(actual_urls_called)
 
         expected_urls = (
-            f"https://query2.finance.yahoo.com/v8/finance/chart/{symbol}?events=div%2Csplits%2CcapitalGains&includePrePost=False&interval=1d&range={range}",
+            f"https://query2.finance.yahoo.com/v8/finance/chart/{symbol}?events=div%2Csplits%2CcapitalGains&includePrePost=False&interval=1d&range={period}",
         )
         self.assertEqual(
             expected_urls,
@@ -211,76 +224,77 @@ def test_actions(self):
         self.assertFalse(data.empty, "data is empty")
 
 
-# Below will fail because not ported to Yahoo API
-# class TestTickerEarnings(unittest.TestCase):
-#     session = None
+class TestTickerEarnings(unittest.TestCase):
+    session = None
+
+    @classmethod
+    def setUpClass(cls):
+        cls.session = session_gbl
 
-#     @classmethod
-#     def setUpClass(cls):
-#         cls.session = session_gbl
+    @classmethod
+    def tearDownClass(cls):
+        if cls.session is not None:
+            cls.session.close()
 
-#     @classmethod
-#     def tearDownClass(cls):
-#         if cls.session is not None:
-#             cls.session.close()
+    def setUp(self):
+        self.ticker = yf.Ticker("GOOGL", session=self.session)
 
-#     def setUp(self):
-#         self.ticker = yf.Ticker("GOOGL", session=self.session)
+    def tearDown(self):
+        self.ticker = None
 
-#     def tearDown(self):
-#         self.ticker = None
+    def test_earnings_dates(self):
+        data = self.ticker.earnings_dates
+        self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+        self.assertFalse(data.empty, "data is empty")
 
-#     def test_earnings(self):
-#         data = self.ticker.earnings
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
+    def test_earnings_dates_with_limit(self):
+        # use ticker with lots of historic earnings
+        ticker = yf.Ticker("IBM")
+        limit = 110
+        data = ticker.get_earnings_dates(limit=limit)
+        self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+        self.assertFalse(data.empty, "data is empty")
+        self.assertEqual(len(data), limit, "Wrong number or rows")
 
-#         data_cached = self.ticker.earnings
-#         self.assertIs(data, data_cached, "data not cached")
+        data_cached = ticker.get_earnings_dates(limit=limit)
+        self.assertIs(data, data_cached, "data not cached")
 
-#     def test_quarterly_earnings(self):
-#         data = self.ticker.quarterly_earnings
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
+    # Below will fail because not ported to Yahoo API
 
-#         data_cached = self.ticker.quarterly_earnings
-#         self.assertIs(data, data_cached, "data not cached")
+    # def test_earnings(self):
+    #     data = self.ticker.earnings
+    #     self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+    #     self.assertFalse(data.empty, "data is empty")
 
-#     def test_earnings_forecasts(self):
-#         data = self.ticker.earnings_forecasts
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
+    #     data_cached = self.ticker.earnings
+    #     self.assertIs(data, data_cached, "data not cached")
 
-#         data_cached = self.ticker.earnings_forecasts
-#         self.assertIs(data, data_cached, "data not cached")
+    # def test_quarterly_earnings(self):
+    #     data = self.ticker.quarterly_earnings
+    #     self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+    #     self.assertFalse(data.empty, "data is empty")
 
-#     def test_earnings_dates(self):
-#         data = self.ticker.earnings_dates
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
+    #     data_cached = self.ticker.quarterly_earnings
+    #     self.assertIs(data, data_cached, "data not cached")
 
-#         data_cached = self.ticker.earnings_dates
-#         self.assertIs(data, data_cached, "data not cached")
+    # def test_earnings_forecasts(self):
+    #     data = self.ticker.earnings_forecasts
+    #     self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+    #     self.assertFalse(data.empty, "data is empty")
 
-#     def test_earnings_trend(self):
-#         data = self.ticker.earnings_trend
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
+    #     data_cached = self.ticker.earnings_forecasts
+    #     self.assertIs(data, data_cached, "data not cached")
 
-#         data_cached = self.ticker.earnings_trend
-#         self.assertIs(data, data_cached, "data not cached")
+    #     data_cached = self.ticker.earnings_dates
+    #     self.assertIs(data, data_cached, "data not cached")
 
-#     def test_earnings_dates_with_limit(self):
-#         # use ticker with lots of historic earnings
-#         ticker = yf.Ticker("IBM")
-#         limit = 110
-#         data = ticker.get_earnings_dates(limit=limit)
-#         self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
-#         self.assertFalse(data.empty, "data is empty")
-#         self.assertEqual(len(data), limit, "Wrong number or rows")
+    # def test_earnings_trend(self):
+    #     data = self.ticker.earnings_trend
+    #     self.assertIsInstance(data, pd.DataFrame, "data has wrong type")
+    #     self.assertFalse(data.empty, "data is empty")
 
-#         data_cached = ticker.get_earnings_dates(limit=limit)
-#         self.assertIs(data, data_cached, "data not cached")
+    #     data_cached = self.ticker.earnings_trend
+    #     self.assertIs(data, data_cached, "data not cached")
 
 
 class TestTickerHolders(unittest.TestCase):

yfinance/__init__.py

@@ -23,7 +23,8 @@
 from .ticker import Ticker
 from .tickers import Tickers
 from .multi import download
-from .utils import set_tz_cache_location, enable_debug_mode
+from .utils import enable_debug_mode
+from .cache import set_tz_cache_location
 
 __version__ = version.version
 __author__ = "Ran Aroussi"

yfinance/base.py

@@ -22,6 +22,7 @@
 from __future__ import print_function
 
 import datetime as _datetime
+from io import StringIO
 import json as _json
 import logging
 import time as _time
@@ -34,9 +35,8 @@
 import pandas as pd
 import requests
 
-from . import shared
-from . import utils
-from .data import TickerData
+from . import shared, utils, cache
+from .data import YfData
 from .scrapers.analysis import Analysis
 from .scrapers.fundamentals import Fundamentals
 from .scrapers.holders import Holders
@@ -69,12 +69,12 @@ def __init__(self, ticker, session=None, proxy=None):
         if utils.is_isin(self.ticker):
             self.ticker = utils.get_ticker_by_isin(self.ticker, None, session)
 
-        self._data: TickerData = TickerData(self.ticker, session=session)
+        self._data: YfData = YfData(session=session)
 
-        self._analysis = Analysis(self._data)
-        self._holders = Holders(self._data)
-        self._quote = Quote(self._data)
-        self._fundamentals = Fundamentals(self._data)
+        self._analysis = Analysis(self._data, ticker)
+        self._holders = Holders(self._data, ticker)
+        self._quote = Quote(self._data, ticker)
+        self._fundamentals = Fundamentals(self._data, ticker)
 
         self._fast_info = None
 
@@ -1644,20 +1644,20 @@ def _get_ticker_tz(self,timeout, proxy=None):
         proxy = proxy or self.proxy
         if self._tz is not None:
             return self._tz
-        cache = utils.get_tz_cache()
-        tz = cache.lookup(self.ticker)
+        c = cache.get_tz_cache()
+        tz = c.lookup(self.ticker)
 
         if tz and not utils.is_valid_timezone(tz):
             # Clear from cache and force re-fetch
-            cache.store(self.ticker, None)
+            c.store(self.ticker, None)
             tz = None
 
         if tz is None:
             tz = self._fetch_ticker_tz(proxy, timeout)
 
             if utils.is_valid_timezone(tz):
                 # info fetch is relatively slow so cache timezone
-                cache.store(self.ticker, tz)
+                c.store(self.ticker, tz)
             else:
                 tz = None
 
@@ -2093,7 +2093,7 @@ def get_earnings_dates(self, limit=12, proxy=None) -> Optional[pd.DataFrame]:
                                    "the issue. Thank you for your patience.")
 
             try:
-                data = pd.read_html(data)[0]
+                data = pd.read_html(StringIO(data))[0]
             except ValueError:
                 if page_offset == 0:
                     # Should not fail on first page