moviewyrm/bookwyrm/goodreads_import.py

122 lines
3.7 KiB
Python
Raw Normal View History

2020-03-29 07:05:09 +00:00
''' handle reading a csv from goodreads '''
import csv
2020-11-13 17:47:35 +00:00
import logging
2021-01-13 21:36:01 +00:00
from bookwyrm import models
from bookwyrm.models import ImportJob, ImportItem
2021-01-13 21:36:01 +00:00
from bookwyrm.tasks import app
2020-03-29 07:05:09 +00:00
2020-11-13 17:47:35 +00:00
logger = logging.getLogger(__name__)
2020-03-25 12:58:27 +00:00
2020-03-29 07:05:09 +00:00
def create_job(user, csv_file, include_reviews, privacy):
2020-05-09 21:26:27 +00:00
''' check over a csv and creates a database entry for the job'''
job = ImportJob.objects.create(
user=user,
include_reviews=include_reviews,
privacy=privacy
)
2021-01-02 17:42:33 +00:00
for index, entry in enumerate(list(csv.DictReader(csv_file))):
if not all(x in entry for x in ('ISBN13', 'Title', 'Author')):
raise ValueError('Author, title, and isbn must be in data.')
ImportItem(job=job, index=index, data=entry).save()
return job
2021-01-02 19:29:50 +00:00
2020-11-13 17:02:41 +00:00
def create_retry_job(user, original_job, items):
''' retry items that didn't import '''
job = ImportJob.objects.create(
user=user,
include_reviews=original_job.include_reviews,
privacy=original_job.privacy,
retry=True
)
for item in items:
ImportItem(job=job, index=item.index, data=item.data).save()
return job
2020-05-09 21:26:27 +00:00
2021-01-02 19:29:50 +00:00
def start_import(job):
2020-05-09 21:26:27 +00:00
''' initalizes a csv import job '''
result = import_data.delay(job.id)
job.task_id = result.id
job.save()
2020-04-20 16:10:19 +00:00
2020-05-09 21:26:27 +00:00
2020-04-20 16:10:19 +00:00
@app.task
def import_data(job_id):
2020-05-09 21:26:27 +00:00
''' does the actual lookup work in a celery task '''
job = ImportJob.objects.get(id=job_id)
2020-04-22 11:43:10 +00:00
try:
for item in job.items.all():
try:
item.resolve()
2020-12-13 02:13:00 +00:00
except Exception as e:# pylint: disable=broad-except
2020-11-13 17:47:35 +00:00
logger.exception(e)
item.fail_reason = 'Error loading book'
item.save()
continue
2020-04-22 11:43:10 +00:00
if item.book:
item.save()
# shelves book and handles reviews
2021-01-13 21:36:01 +00:00
handle_imported_book(
job.user, item, job.include_reviews, job.privacy)
2020-04-22 11:43:10 +00:00
else:
item.fail_reason = 'Could not find a match for book'
2020-04-22 11:43:10 +00:00
item.save()
finally:
job.complete = True
job.save()
2021-01-13 21:36:01 +00:00
def handle_imported_book(user, item, include_reviews, privacy):
''' process a goodreads csv and then post about it '''
if isinstance(item.book, models.Work):
item.book = item.book.default_edition
if not item.book:
return
existing_shelf = models.ShelfBook.objects.filter(
book=item.book, user=user).exists()
2021-01-13 21:36:01 +00:00
# shelve the book if it hasn't been shelved already
if item.shelf and not existing_shelf:
desired_shelf = models.Shelf.objects.get(
identifier=item.shelf,
user=user
)
2021-02-04 21:21:55 +00:00
models.ShelfBook.objects.create(
book=item.book, shelf=desired_shelf, user=user)
2021-01-13 21:36:01 +00:00
for read in item.reads:
# check for an existing readthrough with the same dates
if models.ReadThrough.objects.filter(
user=user, book=item.book,
start_date=read.start_date,
finish_date=read.finish_date
).exists():
continue
read.book = item.book
read.user = user
read.save()
if include_reviews and (item.rating or item.review):
review_title = 'Review of {!r} on Goodreads'.format(
item.book.title,
) if item.review else ''
# we don't know the publication date of the review,
# but "now" is a bad guess
published_date_guess = item.date_read or item.date_added
2021-02-04 21:21:55 +00:00
models.Review.objects.create(
2021-01-13 21:36:01 +00:00
user=user,
book=item.book,
name=review_title,
content=item.review,
rating=item.rating,
published_date=published_date_guess,
privacy=privacy,
)