Fix reading timezone from earning dates

CHANGELOG.rst

@@ -1,6 +1,15 @@
 Change Log
 ===========
 
+0.2.50
+------
+Fixes:
+- price repair #2111 #2139
+- download() appearance 2109
+- isin() error #2099
+- growth_estimates #2127
+Also new docs #2132
+
 0.2.49
 ------
 Fix prices-clean rarely discarding good data #2122

meta.yaml

@@ -1,5 +1,5 @@
 {% set name = "yfinance" %}
-{% set version = "0.2.49" %}
+{% set version = "0.2.50" %}
 
 package:
   name: "{{ name|lower }}"

requirements.txt

@@ -8,4 +8,4 @@ pytz>=2022.5
 frozendict>=2.3.4
 beautifulsoup4>=4.11.1
 html5lib>=1.1
-peewee>=3.16.2
+peewee>=3.16.2

tests/test_ticker.py

@@ -8,18 +8,14 @@
    python -m unittest tests.ticker.TestTicker
 
 """
-import pandas as pd
-
-from tests.context import yfinance as yf
-from tests.context import session_gbl
-from yfinance.exceptions import YFPricesMissingError, YFInvalidPeriodError, YFNotImplementedError, YFTickerMissingError, YFTzMissingError, YFDataException
-
 
 import unittest
-import requests_cache
 from typing import Union, Any, get_args, _GenericAlias
 from urllib.parse import urlparse, parse_qs, urlencode, urlunparse
 
+import pandas as pd
+import requests_cache
+
 ticker_attributes = (
     ("major_holders", pd.DataFrame),
     ("institutional_holders", pd.DataFrame),
@@ -305,6 +301,7 @@ 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")
+        self.assertEqual(data.index.tz.zone, "America/New_York")
 
     def test_earnings_dates_with_limit(self):
         # use ticker with lots of historic earnings
@@ -314,6 +311,7 @@ def test_earnings_dates_with_limit(self):
         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")
+        self.assertEqual(data.index[0].tz.zone, "America/New_York")
 
         data_cached = ticker.get_earnings_dates(limit=limit)
         self.assertIs(data, data_cached, "data not cached")
@@ -339,6 +337,15 @@ def test_earnings_dates_with_limit(self):
     #     data_cached = self.ticker.earnings_trend
     #     self.assertIs(data, data_cached, "data not cached")
 
+    def test_ticker_has_tz(self):
+        test_data = {"AMZN": "America/New_York", "LHA.DE": "Europe/Berlin", "6758.T": "Asia/Tokyo"}
+        for symbol, tz in test_data.items():
+            with self.subTest(f"{symbol}-{tz}"):
+                ticker = yf.Ticker(symbol)
+                data = ticker.get_earnings_dates(limit=1)
+                self.assertIsNotNone(data.index.tz)
+                self.assertEqual(data.index.tz.zone, tz)
+
 
 class TestTickerHolders(unittest.TestCase):
     session = None

yfinance/base.py

@@ -21,26 +21,25 @@
 
 from __future__ import print_function
 
-from io import StringIO
 import json as _json
 import warnings
+from io import StringIO
 from typing import Optional, Union
 from urllib.parse import quote as urlencode
 
 import pandas as pd
 import requests
 
 from . import utils, cache
+from .const import _BASE_URL_, _ROOT_URL_
 from .data import YfData
 from .exceptions import YFEarningsDateMissing
 from .scrapers.analysis import Analysis
 from .scrapers.fundamentals import Fundamentals
+from .scrapers.funds import FundsData
+from .scrapers.history import PriceHistory
 from .scrapers.holders import Holders
 from .scrapers.quote import Quote, FastInfo
-from .scrapers.history import PriceHistory
-from .scrapers.funds import FundsData
-
-from .const import _BASE_URL_, _ROOT_URL_
 
 
 class TickerBase:
@@ -575,6 +574,15 @@ def get_earnings_dates(self, limit=12, proxy=None) -> Optional[pd.DataFrame]:
 
         logger = utils.get_yf_logger()
 
+        ticker_tz = ""
+
+        def get_ticker_tz():
+            nonlocal ticker_tz
+            if ticker_tz == "":
+                self._quote.proxy = proxy or self.proxy
+                ticker_tz = self._get_ticker_tz(proxy=proxy, timeout=30)
+            return ticker_tz
+
         page_size = min(limit, 100)  # YF caps at 100, don't go higher
         page_offset = 0
         dates = None
@@ -639,10 +647,48 @@ def get_earnings_dates(self, limit=12, proxy=None) -> Optional[pd.DataFrame]:
         # - combine and parse
         dates[cn] = dates[cn] + ' ' + tzinfo["AM/PM"]
         dates[cn] = pd.to_datetime(dates[cn], format="%b %d, %Y, %I %p")
-        # - instead of attempting decoding of ambiguous timezone abbreviation, just use 'info':
+
+        # Try to remap all ambiguous timezone values:
+        tzinfo['TZ'] = tzinfo['TZ'].str.replace('BST', 'Europe/London')
+        tzinfo['TZ'] = tzinfo['TZ'].str.replace('GMT', 'Europe/London')
+        if '.' not in self.ticker:
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('EST', 'America/New_York')
+        elif self.ticker.endswith(".AX"):
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('EST', 'Australia/Sydney')
+        tzinfo['TZ'] = tzinfo['TZ'].str.replace('MST', 'America/Denver')
+        tzinfo['TZ'] = tzinfo['TZ'].str.replace('PST', 'America/Los_Angeles')
+        if'.' not in self.ticker:
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('CST', 'America/Chicago')
+        else:
+            # Revisit if Cuba get a stock exchange
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('CST', 'Asia/Shanghai')
+        if self.ticker.endswith('.TA'):
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('IST', 'Asia/Jerusalem')
+        elif self.ticker.endswith('.IR'):
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('IST', 'Europe/Dublin')
+        elif self.ticker.endswith('.NS'):
+            tzinfo['TZ'] = tzinfo['TZ'].str.replace('IST', 'Asia/Kolkata')
+
+        # But in case still ambiguity that pytz cannot parse, have a backup:
         self._quote.proxy = proxy or self.proxy
-        tz = self._get_ticker_tz(proxy=proxy, timeout=30)
-        dates[cn] = dates[cn].dt.tz_localize(tz)
+        tz_backup = self._get_ticker_tz(proxy=proxy, timeout=30)
+
+        if len(tzinfo['TZ'].unique())==1:
+            try:
+                dates[cn] = dates[cn].dt.tz_localize(tzinfo['TZ'].iloc[0])
+            except Exception:
+                dates[cn] = dates[cn].dt.tz_localize(tz_backup)
+        else:
+            dates2 = []
+            for i in range(len(dates)):
+                dt = dates[cn].iloc[i]
+                tz = tzinfo['TZ'].iloc[i]
+                try:
+                    dt = dt.tz_localize(tz)
+                except Exception:
+                    dt = dt.tz_localize(tz_backup)
+                dates2.append(dt)
+            dates[cn] = pd.to_datetime(dates2, utc=True).tz_convert(dates2[0].tzinfo)
 
         dates = dates.set_index("Earnings Date")
 

yfinance/version.py

@@ -1 +1 @@
-version = "0.2.49"
+version = "0.2.50"