Improved cookiejar handling for now

obsidian_import/__init__.py

@@ -261,17 +261,22 @@ PATH_WATCHED = Path('output/show_episodes_watched.csv')
 PATH_PLAYED = Path('output/games_played.csv')
 PATH_WORKOUT = Path('/home/jmaa/Notes/workout.csv')
 PATH_STEP_COUNTS = Path(
-    '/home/jmaa/personal-archive/misc-data/step_counts_2023-07-26_to_2024-09-21.csv',
+    '/home/jmaa/Notes/Rawbackupdata/Steps/exportStepCount_2025-03-15_22-58-20',
 )
 PATH_STEPMANIA = Path('output/stepmania.csv')
 
 
 IMPORTERS = [
-    {'path': PATH_WORKOUT, 'import_rows': import_workout_csv},
+    {'path': PATH_WORKOUT, 'standard_variant': True, 'import_rows': import_workout_csv},
     {'path': PATH_STEP_COUNTS, 'import_rows': import_step_counts_csv},
-    {'path': PATH_STEPMANIA, 'import_rows': import_stepmania_steps_csv},
+    {
+        'path': PATH_STEPMANIA,
+        'standard_variant': True,
+        'import_rows': import_stepmania_steps_csv,
+    },
     {
         'path': PATH_PLAYED,
+        'standard_variant': True,
         'import_rows': lambda vault, rows: import_activity_sample_csv(
             vault,
             rows,
@@ -281,6 +286,7 @@ IMPORTERS = [
     },
     {
         'path': PATH_WATCHED,
+        'standard_variant': True,
         'import_rows': lambda vault, rows: import_activity_sample_csv(
             vault,
             rows,
@@ -301,7 +307,9 @@ def import_data(obsidian_path: Path, dry_run=True):
                 import_def['path'],
             )
             continue
-        rows = load_csv_file(import_def['path'])
+        rows = load_csv_file(
+            import_def['path'], sniff=not import_def.get('standard_variant'),
+        )
         logger.info('Loaded CSV with %d lines', len(rows))
         num_files_updated = import_def['import_rows'](vault, rows)
         logger.info('Updated %d files', num_files_updated)

personal_data/csv_import.py

@@ -2,6 +2,7 @@ import csv
 import dataclasses
 import datetime
 import decimal
+import logging
 import typing
 import urllib.parse
 from collections.abc import Callable
@@ -11,6 +12,8 @@ from typing import Any
 
 from frozendict import frozendict
 
+logger = logging.getLogger(__name__)
+
 CSV_DIALECT = 'one_true_dialect'
 csv.register_dialect(CSV_DIALECT, lineterminator='\n', skipinitialspace=True)
 
@@ -86,10 +89,12 @@ def load_csv_file(csv_file: Path, sniff=False) -> list[frozendict[str, typing.An
     dicts: list[frozendict] = []
     with open(csv_file) as csvfile:
         if sniff:
-            dialect = csv.Sniffer().sniff(csvfile.read(1024))
+            logger.warning('Sniffing CSV variant: %s', csv_file)
+            dialect = csv.Sniffer().sniff(csvfile.read(1024), delimiters=',;')
             csvfile.seek(0)
         else:
             dialect = CSV_DIALECT
+        logger.warning('Loading CSV file: %s', csv_file)
         reader = csv.DictReader(csvfile, dialect=dialect)
         for row in reader:
             for k in list(row.keys()):

personal_data/fetchers/gitea.py

@@ -4,9 +4,9 @@ from collections.abc import Iterator, Mapping
 from typing import Any
 
 from personal_data.data import DeduplicateMode, Scraper
-from ..util import safe_del
 
 from .. import secrets
+from ..util import safe_del
 
 logger = logging.getLogger(__name__)
 

personal_data/fetchers/youtube.py

@@ -1,41 +1,87 @@
-import csv
+import datetime
 import json
 import logging
 import subprocess
 from dataclasses import dataclass
+from typing import ClassVar
 
 from personal_data.data import DeduplicateMode, Scraper
+
 from ..util import safe_del
 
 logger = logging.getLogger(__name__)
 
 PLAYLIST_ID = 'PLAfDVJvDKCvOMvfoTL7eW8GkWNJwd90eV'
-#PLAYLIST_ID='LL'
 
-@dataclass(frozen=True)
-class YoutubeFavoritesScraper(Scraper):
-    dataset_name: str = 'youtube_favorites'
-    deduplicate_mode: DeduplicateMode = DeduplicateMode.BY_ALL_COLUMNS
-    deduplicate_ignore_columns = []
 
-    def scrape(self) -> list[dict]:
+def scrape(watch_history: bool) -> list[dict[str, str]]:
     """Use yt-dlp to fetch the list of favorited videos. This is a placeholder for invoking yt-dlp and parsing its output."""
-        result = subprocess.run(
+    if watch_history:
-            [
+        url = 'https://www.youtube.com/feed/history'
+        ytdlp_args = [
+            'yt-dlp',
+            url,
+            '--dump-json',
+            '--cookies-from-browser',
+            'firefox:/home/jmaa/.cachy/mbui5xg7.default-release',
+        ]
+    else:
+        url = f'https://www.youtube.com/playlist?list={PLAYLIST_ID}'
+        ytdlp_args = [
             'yt-dlp',
             '--flat-playlist',
             '--dump-json',
-                f'https://www.youtube.com/playlist?list={PLAYLIST_ID}',
+            url,
-            ],
+        ]
+
+    result = subprocess.run(
+        ytdlp_args,
         capture_output=True,
         text=True,
     )
 
     if result.returncode != 0:
-            raise RuntimeError(f'Non-zero returncode in command: {result.returncode}\n\n{result.stderr}')
+        message = (
+            'Non-zero returncode in command: '
+            + str(result.returncode)
+            + '\n\n'
+            + result.stderr
+        )
+        raise RuntimeError(message)
 
+    output = []
     for line in result.stdout.splitlines():
         data = json.loads(line)
+        if watch_history:
+            if data.get('thumbnails'):
+                data['thumbnail'] = data['thumbnails'][-1]['url']
+            if data.get('timestamp'):
+                data['watch_datetime'] = datetime.datetime.fromtimestamp(
+                    int(data['timestamp']),
+                    tz=datetime.timezone.utc,
+                ).isoformat()
+        else:
             data['thumbnail'] = data['thumbnails'][-1]['url']
         safe_del(data, '_type', '_version', 'thumbnails')
-            yield data
+        output.append(data)
+    return output
+
+
+@dataclass(frozen=True)
+class YoutubeFavoritesScraper(Scraper):
+    dataset_name: str = 'youtube_favorites'
+    deduplicate_mode: DeduplicateMode = DeduplicateMode.BY_ALL_COLUMNS
+    deduplicate_ignore_columns: ClassVar[list[str]] = []
+
+    def scrape(self) -> list[dict]:
+        yield from scrape(watch_history=False)
+
+
+@dataclass(frozen=True)
+class YoutubeWatchHistoryScraper(Scraper):
+    dataset_name: str = 'youtube_watch_history'
+    deduplicate_mode: DeduplicateMode = DeduplicateMode.BY_ALL_COLUMNS
+    deduplicate_ignore_columns: ClassVar[list[str]] = []
+
+    def scrape(self) -> list[dict]:
+        yield from scrape(watch_history=True)

personal_data/main.py

@@ -91,6 +91,19 @@ def available_scraper_names() -> list[str]:
     return [scraper_cls.__name__ for scraper_cls in available_scrapers()]
 
 
+def get_cookiejar(use_cookiejar: bool):
+    if use_cookiejar:
+        logger.warning('Got cookiejar from firefox')
+        cookiejar = browsercookie.firefox()
+        if len(cookiejar) > 10:
+            return cookiejar
+        browsercookie.firefox(['/home/jmaa/.cachy/mbui5xg7.default-release/cookies.sqlite'])
+        if len(cookiejar) > 10:
+            return cookiejar
+    logger.warning('No cookiejar is used')
+    return []
+
+
 def main(
     scraper_filter: frozenset[str],
     *,
@@ -98,12 +111,8 @@ def main(
     ignore_cache: bool,
     notification_types: frozenset[notification.NotificationType],
 ) -> None:
-    if use_cookiejar:
+    cookiejar = get_cookiejar(use_cookiejar)
-        cookiejar = browsercookie.firefox()
+    logger.warning('Cookiejar has %s cookies', len(cookiejar))
-        logger.info('Got cookiejar from firefox: %s cookies', len(cookiejar))
-    else:
-        cookiejar = []
-        logger.warning('No cookiejar is used')
 
     if len(notification_types) == 0:
         logger.info('No notifications enabled: Notifications will not be sent!')

personal_data/util.py

@@ -20,7 +20,6 @@ def safe_del(d: dict, *keys: str):
             del d[key]
 
 
-
 def equals_without_fields(
     a: Mapping[str, Any],
     b: Mapping[str, Any],