From 263b0b284364cb89b6dfa3ed30382e739747904b Mon Sep 17 00:00:00 2001
From: Jon Michael Aanes <jonjmaa@gmail.com>
Date: Sun, 1 Sep 2024 16:55:45 +0200
Subject: [PATCH] Implemented partial Steam support

---
 personal_data/fetchers/steam_community.py | 70 +++++++++++++++++++++++
 personal_data/parse_util.py               | 18 +++++-
 personal_data/secrets.py                  |  3 +
 test/test_deduplicate.py                  |  4 +-
 test/test_parsing.py                      |  2 +-
 5 files changed, 91 insertions(+), 6 deletions(-)
 create mode 100644 personal_data/fetchers/steam_community.py

diff --git a/personal_data/fetchers/steam_community.py b/personal_data/fetchers/steam_community.py
new file mode 100644
index 0000000..a975e33
--- /dev/null
+++ b/personal_data/fetchers/steam_community.py
@@ -0,0 +1,70 @@
+import dataclasses
+import datetime
+import logging
+import re
+import bs4
+from typing import Any
+from collections.abc import Iterator
+
+from ..data import DeduplicateMode, Scraper
+from .. import secrets, parse_util, html_util
+
+logger = logging.getLogger(__name__)
+
+URL_SITE_ROOT = 'https://steamcommunity.com/'
+
+URL_GAME_ACHIVEMENTS = URL_SITE_ROOT+'id/{username}/stats/appid/{appid}'
+
+FORMAT_DATE_HEADER = '%d/%m/%YYYY'
+
+
+@dataclasses.dataclass(frozen=True)
+class SteamAchievementScraper(Scraper):
+    dataset_name = 'games_played_TODO'
+    deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
+
+    def scrape(self) -> Iterator[dict[str, Any]]:
+        yield from self.scrape_app(105600)
+
+    def scrape_app(self, appid: int) -> Iterator[dict[str, Any]]:
+        url = URL_GAME_ACHIVEMENTS.format(
+                username=secrets.STEAM_USERNAME,
+                appid=appid,
+        )
+        response = self.session.get(url)
+        response.raise_for_status()
+
+        NOW = parse_util.parse_response_datetime(response)
+
+        # Parse data
+        soup = bs4.BeautifulSoup(response.content, 'lxml')
+
+        game_name: str = re.match(r'Steam Community :: (.+) :: Jmaa', soup.head.title.get_text()).group(1)
+
+        soup = html_util.normalize_soup_slightly(
+            soup,
+            classes=False,
+        )
+
+
+        for entry in soup.select('.achieveRow'):
+            trophy_name: str = entry.select_one('h3').get_text()
+            trophy_desc: str = entry.select_one('h5').get_text()
+            trophy_icon: str = entry.select_one('img').src
+
+            time_acquired_html: str = entry.select_one('.achieveUnlockTime')
+            if time_acquired_html is None:
+                continue
+            time_acquired_text: str = time_acquired_html.get_text().strip().removeprefix('Unlocked ')
+            time_acquired: datetime.datetime = parse_util.parse_time(time_acquired_text)
+
+            yield {
+                'game.name': game_name,
+                'me.last_played_time': time_acquired,
+                # Trophy Data
+                'trophy.name': trophy_name,
+                'trophy.desc': trophy_desc,
+                'trophy.icon': trophy_icon,
+            }
+
+            del entry, time_acquired
diff --git a/personal_data/parse_util.py b/personal_data/parse_util.py
index d23c4a0..4f42bc2 100644
--- a/personal_data/parse_util.py
+++ b/personal_data/parse_util.py
@@ -36,14 +36,26 @@ def parse_response_datetime(response) -> datetime.datetime:
     ).replace(tzinfo=datetime.UTC)
 
 
-LOCAL_TIMEZONE = datetime.datetime.now(datetime.timezone.utc).astimezone().tzinfo
+LOCAL_TIMEZONE = datetime.datetime.now(datetime.UTC).astimezone().tzinfo
+
+
+def try_parse(text:str, fmt:str) -> datetime.datetime | None:
+    try:
+        time = datetime.datetime.strptime(text, fmt)
+        if time.tzinfo is None:
+            time = time.replace(tzinfo=LOCAL_TIMEZONE)
+    except:
+        time = None
+    return time
 
 
 def parse_time(text: str) -> datetime.datetime:
     text = text.replace('\n', ' ')
     text = text.strip()
-    time = datetime.datetime.strptime(text, '%d %b %Y %I:%M:%S %p')
-    time = time.replace(tzinfo=LOCAL_TIMEZONE)
+
+    time = try_parse(text, '%d %b %Y %I:%M:%S %p')
+    time = time or try_parse(text, '%d %b, %Y @ %I:%M%p')
+
     assert time.tzinfo is not None, time
     return time
 
diff --git a/personal_data/secrets.py b/personal_data/secrets.py
index a56c1d5..83af2d5 100644
--- a/personal_data/secrets.py
+++ b/personal_data/secrets.py
@@ -15,6 +15,9 @@ PLAYSTATION_PSN_ID = load_secret('PLAYSTATION_PSN_ID')
 # Partisia Blockchain
 PBC_ACCOUNT_ADDRESS = load_secret('PBC_ACCOUNT_ADDRESS')
 
+# Steam
+STEAM_USERNAME = load_secret('STEAM_USERNAME')
+
 # Kucoin
 KUCOIN_KEY = load_secret('KUCOIN_KEY')
 KUCOIN_SECRET = load_secret('KUCOIN_SECRET')
diff --git a/test/test_deduplicate.py b/test/test_deduplicate.py
index 2dabd59..10250f9 100644
--- a/test/test_deduplicate.py
+++ b/test/test_deduplicate.py
@@ -60,7 +60,7 @@ LIST_2 = [
                 27,
                 31,
                 134506,
-                tzinfo=datetime.timezone.utc,
+                tzinfo=datetime.UTC,
             ),
             'weight.kg': Decimal('73.6'),
         },
@@ -75,7 +75,7 @@ LIST_2 = [
                 36,
                 9,
                 590355,
-                tzinfo=datetime.timezone.utc,
+                tzinfo=datetime.UTC,
             ),
             'weight.kg': Decimal('74.7'),
         },
diff --git a/test/test_parsing.py b/test/test_parsing.py
index b082da9..ce6d15c 100644
--- a/test/test_parsing.py
+++ b/test/test_parsing.py
@@ -8,7 +8,7 @@ from personal_data.main import to_value
 PARSE_MAPPINGS = [
     (
         '2024-04-28 21:35:40+00:00',
-        datetime.datetime(2024, 4, 28, 21, 35, 40, tzinfo=datetime.timezone.utc),
+        datetime.datetime(2024, 4, 28, 21, 35, 40, tzinfo=datetime.UTC),
     ),
     (
         '0003791e9f5f3691b8bbbe0d12a7ae9c3f2e89db38',