def check_iter_locked(db_path, pre_stuff, iter_stuff):
"""Actual implementation of test_errors_locked, so it can be reused."""
# WAL provides more concurrency; some things won't to block with it enabled.
storage = Storage(db_path, wal_enabled=False)
feed = FeedData('one')
entry = EntryData('one', 'entry', datetime(2010, 1, 1), title='entry')
storage.add_feed(feed.url, datetime(2010, 1, 2))
storage.add_or_update_entry(
EntryUpdateIntent(entry, entry.updated, datetime(2010, 1, 1), 0, 0))
storage.add_feed('two', datetime(2010, 1, 1))
storage.add_or_update_entry(
EntryUpdateIntent(entry._replace(feed_url='two'), entry.updated,
datetime(2010, 1, 1), 0, 0))
storage.set_feed_metadata('two', '1', 1)
storage.set_feed_metadata('two', '2', 2)
storage.add_feed_tag('two', '1')
storage.add_feed_tag('two', '2')
if pre_stuff:
pre_stuff(storage)
rv = iter_stuff(storage)
next(rv)
# shouldn't raise an exception
storage = Storage(db_path, timeout=0, wal_enabled=False)
storage.mark_as_read_unread(feed.url, entry.id, 1)
storage = Storage(db_path, timeout=0)
storage.mark_as_read_unread(feed.url, entry.id, 0)
def storage_with_two_entries(storage):
storage.add_feed('feed', datetime(2010, 1, 1))
storage.add_or_update_entry(
EntryUpdateIntent(
EntryData('feed', 'one', datetime(2010, 1, 1)),
datetime(2010, 1, 2),
datetime(2010, 1, 2),
0,
))
storage.add_or_update_entry(
EntryUpdateIntent(
EntryData('feed', 'two', datetime(2010, 1, 1)),
datetime(2010, 1, 2),
datetime(2010, 1, 2),
1,
))
return storage
def test_object_id():
assert Feed('url').object_id == 'url'
assert Entry('entry', 'updated',
feed=Feed('url')).object_id == ('url', 'entry')
assert EntrySearchResult('url', 'entry').object_id == ('url', 'entry')
assert FeedData('url').object_id == 'url'
assert EntryData('url', 'entry', 'updated').object_id == ('url', 'entry')
assert FeedError('url').object_id == 'url'
assert EntryError('url', 'entry').object_id == ('url', 'entry')
def _make_entry(feed_number, number, updated, **kwargs):
return EntryData(
f'{feed_number}',
# evals to tuple
f'{feed_number}, {number}',
updated,
kwargs.pop('title', f'Entry #{number}'),
kwargs.pop('link', f'http://www.example.com/entries/{number}'),
**kwargs,
)
def test_important_entry_remains_important_after_update(storage):
storage.mark_as_important_unimportant('feed', 'one', True)
storage.add_or_update_entry(
EntryUpdateIntent(
EntryData('feed', 'one', datetime(2010, 1, 1)),
datetime(2010, 1, 2),
datetime(2010, 1, 2),
0,
))
assert {
e.id
for e in storage.get_entries(datetime(2010, 1, 1),
EntryFilterOptions(important=True))
} == {'one'}
def test_entry_remains_read_after_update(storage_with_two_entries):
storage = storage_with_two_entries
storage.mark_as_read_unread('feed', 'one', True)
storage.add_or_update_entry(
EntryUpdateIntent(
EntryData('feed', 'one', datetime(2010, 1, 1)),
datetime(2010, 1, 2),
datetime(2010, 1, 2),
0,
))
assert {
e.id
for e in storage.get_entries(datetime(2010, 1, 1),
EntryFilterOptions(read=True))
} == {'one'}
def make_entries(feed_url, url, soup):
for title, fragment, content in extract_text(soup):
try:
updated = datetime.strptime(title.split()[0], '%Y-%m-%d')
except (ValueError, IndexError):
continue
link = urlunparse(urlparse(url)._replace(fragment=fragment))
yield EntryData(
feed_url=feed_url,
id=title,
updated=updated,
title=title,
link=link,
summary=content,
)
def test_get_entries_for_update(storage_cls):
storage = storage_cls(':memory:')
storage.add_feed('feed', datetime(2010, 1, 1))
storage.add_or_update_entry(
EntryUpdateIntent(
EntryData('feed', 'one', datetime(2010, 1, 1)),
datetime(2010, 1, 2),
datetime(2010, 1, 1),
0,
))
assert list(
storage.get_entries_for_update([
('feed', 'one'), ('feed', 'two')
])) == [
EntryForUpdate(datetime(2010, 1, 1)),
None,
]
def check_errors_locked(db_path, pre_stuff, do_stuff, exc_type):
"""Actual implementation of test_errors_locked, so it can be reused."""
# WAL provides more concurrency; some things won't to block with it enabled.
storage = Storage(db_path, wal_enabled=False)
storage.db.execute("PRAGMA busy_timeout = 0;")
feed = FeedData('one')
entry = EntryData('one', 'entry', datetime(2010, 1, 2))
storage.add_feed(feed.url, datetime(2010, 1, 1))
storage.add_or_update_entry(
EntryUpdateIntent(entry, entry.updated, datetime(2010, 1, 1), 0, 0))
in_transaction = threading.Event()
can_return_from_transaction = threading.Event()
def target():
storage = Storage(db_path, wal_enabled=False)
storage.db.isolation_level = None
storage.db.execute("BEGIN EXCLUSIVE;")
in_transaction.set()
can_return_from_transaction.wait()
storage.db.execute("ROLLBACK;")
if pre_stuff:
pre_stuff(storage, feed, entry)
thread = threading.Thread(target=target)
thread.start()
in_transaction.wait()
try:
with pytest.raises(exc_type) as excinfo:
do_stuff(storage, feed, entry)
assert 'locked' in str(excinfo.value.__cause__)
finally:
can_return_from_transaction.set()
thread.join()
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(url='{}invalid.json'.format(url_base), )
entries = [
EntryData(
feed_url=feed.url,
id='2',
updated=None,
enclosures=(
Enclosure(href='control'),
Enclosure(href='float size', length=100),
Enclosure(href='non-number size'),
),
),
EntryData(
feed_url=feed.url,
id='3.1415',
title='float id',
updated=None,
),
EntryData(
feed_url=feed.url,
id='author name',
updated=None,
),
title='RSS Title',
link='http://www.example.com/main.html',
author='Example editor ([email protected])',
)
entries = [
EntryData(
feed_url=feed.url,
id='7bd204c6-1655-4c27-aeee-53f933c5395f',
updated=datetime.datetime(2009, 9, 6, 16, 20),
title='Example entry',
link='http://www.example.com/blog/post/1',
author='Example editor',
published=None,
summary='Here is some text containing an interesting description.',
content=(
# the text/plain type comes from feedparser
Content(value='Example content', type='text/plain'),
),
enclosures=(
Enclosure(href='http://example.com/enclosure'),
Enclosure(href='http://example.com/enclosure-with-type', type='image/jpeg'),
Enclosure(href='http://example.com/enclosure-with-length', length=100000),
Enclosure(href='http://example.com/enclosure-with-bad-length'),
),
),
EntryData(
feed_url=feed.url,
id='00000000-1655-4c27-aeee-00000000',
updated=datetime.datetime(2009, 9, 6, 0, 0, 0),
title='Example entry, again',
),
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(url='{}empty.atom'.format(url_base))
entries = [
EntryData(
feed_url=feed.url,
id='urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a',
updated=None,
# added by feedparser
link='urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a',
)
]
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(url='{}relative.rss'.format(url_base),
link='{}file.html'.format(rel_base))
entries = [
EntryData(
feed_url=feed.url,
id='7bd204c6-1655-4c27-aeee-53f933c5395f',
updated=None,
link='{}blog/post/1'.format(rel_base),
summary='one <a href="{}target">two</a> three'.format(rel_base),
content=(
Content(value='<script>evil</script> content',
type='text/plain',
language=None),
Content(value='content', type='text/html', language=None),
),
enclosures=(
# for RSS feedparser doesn't make relative links absolute
# (it does for Atom)
Enclosure(href='enclosure?q=a#fragment'), ),
)
]
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(url='{}relative.atom'.format(url_base),
link='{}file.html'.format(rel_base))
entries = [
EntryData(
feed_url=feed.url,
id='urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a',
updated=None,
link='{}entries/entry.html'.format(rel_base),
summary='one <a href="{}target">two</a> three'.format(rel_base),
content=(
Content(value='<script>evil</script> content',
type='text/plain',
language=None),
Content(value='content', type='text/html', language=None),
),
enclosures=(
# the text/html type comes from feedparser
Enclosure(href='{}enclosure?q=a#fragment'.format(rel_base),
type='text/html'), ),
)
]
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(url='{}empty.rss'.format(url_base))
entries = [
EntryData(feed_url=feed.url,
id='7bd204c6-1655-4c27-aeee-53f933c5395f',
updated=None)
]
import datetime
from reader import Content
from reader import Enclosure
from reader._types import EntryData
from reader._types import FeedData
feed = FeedData(
url='{}empty.json'.format(url_base),
)
entries = [
EntryData(
feed_url=feed.url,
id='1',
updated=None,
content=(
Content(
value='content',
type='text/plain',
),
),
),
]
EntryData(
feed_url=feed.url,
id='urn:uuid:1225c695-cfb8-4ebb-aaaa-80da344efa6a',
updated=datetime.datetime(2003, 12, 13, 18, 30, 2),
title='Atom-Powered Robots Run Amok',
link='http://example.org/2003/12/13/atom03',
author='John Doe',
published=datetime.datetime(2003, 12, 13, 17, 17, 51),
summary='Some text.',
content=(
# the text/plain type comes from feedparser
Content(value='content', type='text/plain'),
Content(value='content with type', type='text/whatever'),
Content(value='content with lang',
type='text/plain',
language='en'),
),
enclosures=(
# the text/html type comes from feedparser
Enclosure(href='http://example.org/enclosure', type='text/html'),
Enclosure(href='http://example.org/enclosure-with-type',
type='text/whatever'),
Enclosure(
href='http://example.org/enclosure-with-length',
type='text/html',
length=1000,
),
Enclosure(href='http://example.org/enclosure-with-bad-length',
type='text/html'),
),
),