From 842bb5d609af44d333d867536b894fa2da2aa5f2 Mon Sep 17 00:00:00 2001
From: Jon Michael Aanes <jonjmaa@gmail.com>
Date: Fri, 25 Oct 2024 21:29:44 +0200
Subject: [PATCH] Code quality and release year parsing

---
 personal_data/fetchers/psnprofiles.py     | 21 +++++++++++++++++++++
 personal_data/fetchers/steam_community.py | 14 ++++++++------
 personal_data/parse_util.py               | 14 ++++++++------
 3 files changed, 37 insertions(+), 12 deletions(-)

diff --git a/personal_data/fetchers/psnprofiles.py b/personal_data/fetchers/psnprofiles.py
index 4cd3b12..4cd66b0 100644
--- a/personal_data/fetchers/psnprofiles.py
+++ b/personal_data/fetchers/psnprofiles.py
@@ -2,6 +2,7 @@ import dataclasses
 import logging
 import re
 from collections.abc import Iterator
+import datetime
 
 import bs4
 import requests_util
@@ -31,6 +32,8 @@ MAX_NUMBER_GAMES_TO_PARSE = 10000
 
 @dataclasses.dataclass(frozen=True)
 class PsnProfilesScraper(Scraper):
+    """Downloads all trophies for the given user."""
+
     dataset_name = 'games_played_playstation'
     deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
 
@@ -124,6 +127,16 @@ class PsnProfilesScraper(Scraper):
                 d['me.last_played_time'] = time_played
             yield d
 
+    def _parse_game_release_date(self, soup: bs4.BeautifulSoup) -> datetime.date:
+        table_rows = soup.select('table.gameInfo tr')
+        for row in table_rows :
+            cells = row.select('td')
+            if cells[0].get_text() in {'Release', 'Releases'}:
+                text = cells[1].get_text()
+                dates = re.findall(r'\w+\s+\d+,\s+\d{4}', text)
+                return min(parse_util.parse_date(date) for date in dates)
+        assert False, 'Could not find release date'
+
     def _scrape_game_trophies(
         self,
         psnprofiles_id: int,
@@ -143,8 +156,14 @@ class PsnProfilesScraper(Scraper):
 
         # Parse data
         soup = bs4.BeautifulSoup(response.content, 'lxml')
+
+        # Normalize before parsing trophies
         soup = personal_data.html_util.normalize_soup_slightly(soup, classes=False)
 
+        # Parse release year
+        game_release_date = self._parse_game_release_date(soup)
+        assert game_release_date
+
         # Remove redundant
         for redundant in soup.select('.wide-ad'):
             redundant.extract()
@@ -174,11 +193,13 @@ class PsnProfilesScraper(Scraper):
 
             yield {
                 'game.name': game_name,
+                'game.release_date': game_release_date,
                 'me.last_played_time': gotten_at,
                 # Trophy Data
                 'trophy.name': trophy_name,
                 'trophy.desc': trophy_desc,
                 'trophy.icon': trophy_icon,
+                # Ids
                 'psnprofiles.game_id': psnprofiles_id,
             }
 
diff --git a/personal_data/fetchers/steam_community.py b/personal_data/fetchers/steam_community.py
index 5290987..a7ea8c3 100644
--- a/personal_data/fetchers/steam_community.py
+++ b/personal_data/fetchers/steam_community.py
@@ -22,15 +22,17 @@ FORMAT_DATE_HEADER = '%d/%m/%YYYY'
 
 @dataclasses.dataclass(frozen=True)
 class SteamAchievementScraper(Scraper):
+    """Downloads all trophies for the given user."""
+
     dataset_name = 'games_played_TODO'
     deduplicate_mode = DeduplicateMode.BY_ALL_COLUMNS
 
     def scrape(self) -> Iterator[dict[str, Any]]:
         username = secrets.STEAM_USERNAME
-        for appid in self.determine_appids_from_recent_activity(username):
-            yield from self.scrape_app(username, appid)
+        for appid in self._determine_appids_from_recent_activity(username):
+            yield from self._scrape_app_achievements(username, appid)
 
-    def determine_appids_from_recent_activity(self, username: str) -> Iterator[int]:
+    def _determine_appids_from_recent_activity(self, username: str) -> Iterator[int]:
         url = URL_USER_RECENT_ACTIVITY.format(
             username=username,
         )
@@ -47,7 +49,7 @@ class SteamAchievementScraper(Scraper):
             appid = int(href.split('/')[-1])
             yield appid
 
-    def scrape_app(self, username: str, appid: int) -> Iterator[dict[str, Any]]:
+    def _scrape_app_achievements(self, username: str, appid: int) -> Iterator[dict[str, Any]]:
         url = URL_GAME_ACHIVEMENTS.format(
             username=username,
             appid=appid,
@@ -55,8 +57,6 @@ class SteamAchievementScraper(Scraper):
         response = self.session.get(url)
         response.raise_for_status()
 
-        NOW = parse_util.parse_response_datetime(response)
-
         # Parse data
         soup = bs4.BeautifulSoup(response.content, 'lxml')
 
@@ -90,6 +90,8 @@ class SteamAchievementScraper(Scraper):
                 'trophy.name': trophy_name,
                 'trophy.desc': trophy_desc,
                 'trophy.icon': trophy_icon,
+                # Ids
+                'steam.appid': appid,
             }
 
             del entry, time_acquired
diff --git a/personal_data/parse_util.py b/personal_data/parse_util.py
index 475da67..0e0bf76 100644
--- a/personal_data/parse_util.py
+++ b/personal_data/parse_util.py
@@ -19,8 +19,6 @@ DATETIME_UNITS = {
 
 FORMAT_DATE_HEADER = '%a, %d %b %Y %H:%M:%S GMT'
 
-FORMAT_DAY_MONTH_YEAR = '%d %B %Y'
-
 
 def parse_duration(text: str) -> datetime.timedelta:
     (num, unit) = text.split(' ')
@@ -69,7 +67,11 @@ def parse_time(text: str) -> datetime.datetime:
 
 
 def parse_date(text: str) -> datetime.date:
-    return datetime.datetime.strptime(
-        text.strip(),
-        FORMAT_DAY_MONTH_YEAR,
-    ).date()
+    text = text.strip()
+    if dt := try_parse(text, '%d %B %Y'):
+        return dt.date()
+    if dt := try_parse(text, '%b %d, %Y'):
+        return dt.date()
+    if dt := try_parse(text, '%B %d, %Y'):
+        return dt.date()
+    assert False, text