[books] Add webdav koreader importer

vrobbler/apps/beers/models.py

@@ -1,5 +1,6 @@
 from uuid import uuid4
 
+from beers.untappd import get_beer_from_untappd_id, get_rating_from_soup
 from django.apps import apps
 from django.db import models
 from django.urls import reverse
@@ -8,10 +9,6 @@ from imagekit.models import ImageSpecField
 from imagekit.processors import ResizeToFit
 from scrobbles.dataclasses import BeerLogData
 from scrobbles.mixins import ScrobblableConstants, ScrobblableMixin
-from vrobbler.apps.beers.untappd import (
-    get_beer_from_untappd_id,
-    get_rating_from_soup,
-)
 
 BNULL = {"blank": True, "null": True}
 

vrobbler/apps/books/koreader.py

@@ -1,4 +1,3 @@
-from collections import OrderedDict
 import logging
 import re
 import sqlite3
@@ -7,12 +6,11 @@ from enum import Enum
 
 import pytz
 import requests
-from books.models import Author, Book
-from books.openlibrary import get_author_openlibrary_id
+from books.constants import BOOKS_TITLES_TO_IGNORE
 from django.apps import apps
 from django.contrib.auth import get_user_model
 from stream_sqlite import stream_sqlite
-from vrobbler.apps.books.constants import BOOKS_TITLES_TO_IGNORE
+from webdav.client import get_webdav_client
 
 logger = logging.getLogger(__name__)
 User = get_user_model()
@@ -63,6 +61,8 @@ def lookup_or_create_authors_from_author_str(ko_author_str: str) -> list:
     """Takes a string of authors from KoReader and returns a list
     of Authors from our database
     """
+    from books.models import Author
+
     author_str_list = ko_author_str.split(", ")
     author_list = []
     for author_str in author_str_list:
@@ -83,6 +83,8 @@ def lookup_or_create_authors_from_author_str(ko_author_str: str) -> list:
 
 
 def create_book_from_row(row: list):
+    from books.models import Book
+
     # No KoReader book yet, create it
     author_str = get_author_str_from_row(row).replace("\x00", "")
     total_pages = row[KoReaderBookColumn.PAGES.value]
@@ -131,6 +133,8 @@ def build_book_map(rows) -> dict:
     primary key IDs for page creation.
 
     """
+    from books.models import Book
+
     book_id_map = {}
 
     for book_row in rows:
@@ -148,7 +152,12 @@ def build_book_map(rows) -> dict:
         ).first()
 
         if not book:
-            title = book_row[KoReaderBookColumn.TITLE.value].split(" - ")[0].lower().replace("\x00", "")
+            title = (
+                book_row[KoReaderBookColumn.TITLE.value]
+                .split(" - ")[0]
+                .lower()
+                .replace("\x00", "")
+            )
             book = Book.objects.filter(title=title).first()
 
         if not book:
@@ -438,3 +447,19 @@ def process_koreader_sqlite_file(file_path, user_id) -> list:
             extra={"created_scrobbles": created},
         )
     return created
+
+
+def fetch_file_from_webdav(user_id: int) -> str:
+    file_path = f"/tmp/{user_id}-koreader-import.sqlite3"
+    client = get_webdav_client(user_id)
+
+    if not client:
+        logger.warning("could not get webdav client for user")
+        # TODO maybe we raise an exception here?
+        return ""
+
+    client.download_sync(
+        remote_path="var/koreader/statistics.sqlite3",
+        local_path=file_path,
+    )
+    return file_path

vrobbler/apps/scrobbles/management/commands/import_from_webdav.py

@@ -0,0 +1,18 @@
+from django.core.management.base import BaseCommand
+from vrobbler.apps.scrobbles.utils import import_from_webdav_for_all_users
+
+
+class Command(BaseCommand):
+    def add_arguments(self, parser):
+        parser.add_argument(
+            "--restart",
+            action="store_true",
+            help="Restart failed imports",
+        )
+
+    def handle(self, *args, **options):
+        restart = False
+        if options["restart"]:
+            restart = True
+        count = import_from_webdav_for_all_users(restart=restart)
+        print(f"Started {count} WeDAV imports")

vrobbler/apps/scrobbles/models.py

@@ -15,6 +15,7 @@ from books.models import Book
 from bricksets.models import BrickSet
 from django.conf import settings
 from django.contrib.auth import get_user_model
+from django.core.files import File
 from django.db import models
 from django.urls import reverse
 from django.utils import timezone
@@ -49,6 +50,7 @@ from videos.models import Series, Video
 from webpages.models import WebPage
 
 from vrobbler.apps.scrobbles.constants import MEDIA_END_PADDING_SECONDS
+from vrobbler.apps.scrobbles.utils import get_file_md5_hash
 
 logger = logging.getLogger(__name__)
 User = get_user_model()
@@ -186,6 +188,19 @@ class KoReaderImport(BaseFileImportMixin):
 
     sqlite_file = models.FileField(upload_to=get_path, **BNULL)
 
+    def save_sqlite_file_to_self(self, file_path):
+        with open(file_path, "rb") as f:
+            self.sqlite_file.save(
+                f"{self.user_id}-koreader-statistics.sqlite",
+                File(f),
+                save=True,
+            )
+
+    def file_md5_hash(self) -> str:
+        if self.sqlite_file:
+            return get_file_md5_hash(self.sqlite_file.path)
+        return ""
+
     def process(self, force=False):
 
         if self.processed_finished and not force:

vrobbler/apps/scrobbles/utils.py

@@ -1,8 +1,10 @@
+import hashlib
 import logging
 import re
 from datetime import datetime, timedelta, tzinfo
 
 import pytz
+from books.koreader import fetch_file_from_webdav
 from django.apps import apps
 from django.contrib.auth import get_user_model
 from django.db import models
@@ -11,6 +13,8 @@ from profiles.models import UserProfile
 from profiles.utils import now_user_timezone
 from scrobbles.constants import LONG_PLAY_MEDIA
 from scrobbles.tasks import process_lastfm_import, process_retroarch_import
+from vrobbler.apps.scrobbles.tasks import process_koreader_import
+from webdav.client import get_webdav_client
 
 logger = logging.getLogger(__name__)
 User = get_user_model()
@@ -178,5 +182,87 @@ def delete_zombie_scrobbles(dry_run=True):
     return zombies_found
 
 
+def import_from_webdav_for_all_users(restart=False):
+    """Grab a list of all users with WebDAV enabled and kickoff imports for them"""
+    from scrobbles.models import KoReaderImport
+
+    # LastFmImport = apps.get_model("scrobbles", "LastFMImport")
+    webdav_enabled_user_ids = UserProfile.objects.filter(
+        webdav_url__isnull=False,
+        webdav_user__isnull=False,
+        webdav_pass__isnull=False,
+        webdav_auto_import=True,
+    ).values_list("user_id", flat=True)
+    logger.info(
+        f"start import of {webdav_enabled_user_ids.count()} webdav accounts"
+    )
+
+    koreader_import_count = 0
+
+    for user_id in webdav_enabled_user_ids:
+        webdav_client = get_webdav_client(user_id)
+
+        try:
+            webdav_client.info("var/koreader/statistics.sqlite3")
+            koreader_found = True
+        except:
+            koreader_found = False
+            logger.info(
+                "no koreader stats file found on webdav",
+                extra={"user_id": user_id},
+            )
+
+        if koreader_found:
+            last_import = (
+                KoReaderImport.objects.filter(
+                    user_id=user_id, processed_finished__isnull=False
+                )
+                .order_by("processed_finished")
+                .last()
+            )
+
+            koreader_file_path = fetch_file_from_webdav(1)
+            new_hash = get_file_md5_hash(koreader_file_path)
+            old_hash = None
+            if last_import:
+                old_hash = last_import.file_md5_hash()
+
+            if old_hash and new_hash == old_hash:
+                logger.info(
+                    "koreader stats file has not changed",
+                    extra={
+                        "user_id": user_id,
+                        "new_hash": new_hash,
+                        "old_hash": old_hash,
+                        "last_import_id": last_import.id,
+                    },
+                )
+                continue
+
+            koreader_import, created = KoReaderImport.objects.get_or_create(
+                user_id=user_id, processed_finished__isnull=True
+            )
+
+            if not created and not restart:
+                logger.info(
+                    f"Not resuming failed KoReader import {koreader_import.id} for user {user_id}, use restart=True to restart"
+                )
+                continue
+
+            koreader_import.save_sqlite_file_to_self(koreader_file_path)
+
+            process_koreader_import.delay(koreader_import.id)
+            koreader_import_count += 1
+    return koreader_import_count
+
+
 def media_class_to_foreign_key(media_class: str) -> str:
     return re.sub(r"(?<!^)(?=[A-Z])", "_", media_class).lower()
+
+
+def get_file_md5_hash(file_path: str) -> str:
+    with open(file_path, "rb") as f:
+        file_hash = hashlib.md5()
+        while chunk := f.read(8192):
+            file_hash.update(chunk)
+    return file_hash.hexdigest()

vrobbler/apps/scrobbles/views.py

@@ -25,7 +25,7 @@ from rest_framework.decorators import (
     permission_classes,
 )
 from rest_framework.parsers import MultiPartParser
-from rest_framework.permissions import IsAuthenticated
+from rest_framework.permissions import IsAuthenticated, AllowAny
 from rest_framework.response import Response
 from scrobbles.api import serializers
 from scrobbles.constants import (

vrobbler/apps/webdav/client.py

@@ -1,23 +1,33 @@
-from webdav3.client import Client
+import logging
+from typing import Optional
 
 from profiles.models import UserProfile
+from webdav3.client import Client
+
+logger = logging.getLogger("__name__")
+
 
-def get_webdav_client(user_id):
-    client = None
+def get_webdav_client(user_id) -> Optional[Client]:
     profile = UserProfile.objects.filter(user_id=user_id).first()
 
     if not profile:
-        logger.info("[get_webdav_client] no profile for user", extra={"user_id": user_id})
+        logger.info(
+            "[get_webdav_client] no profile for user",
+            extra={"user_id": user_id},
+        )
         return
 
     if not profile.webdav_user:
-        logger.info("[get_webdav_client] no webdave user for profile", extra={"user_id": user_id})
+        logger.info(
+            "[get_webdav_client] no webdave user for profile",
+            extra={"user_id": user_id},
+        )
         return
 
     return Client(
         {
-            'webdav_hostname': profile.webdav_url,
-            'webdav_login':    profile.webdav_user,
-            'webdav_password': profile.webdav_pass,
+            "webdav_hostname": profile.webdav_url,
+            "webdav_login": profile.webdav_user,
+            "webdav_password": profile.webdav_pass,
         }
     )