def patched_base_post_serializer(mocker):
"""Fixture that patches the base serializer class for ESPostSerializer"""
article_content = {"text": "hello world"}
base_serialized_data = {
"author_id": 1,
"author_name": "Author Name",
"author_headline": "Author Headline",
"article_content": article_content,
"plain_text": render_article_text(article_content),
"profile_image": "/media/profile/1/208c7d959608417eb13bc87392cb5f77-2018-09-21T163449_small.jpg",
"channel_title": "channel 1",
"channel_name": "channel_1",
"channel_type": "restricted",
"post_type": LINK_TYPE_SELF,
"text": "Post Text",
"score": 1,
"created": 123,
"num_comments": 0,
"removed": False,
"deleted": False,
"id": 1,
"title": "post_title",
"url": None,
"thumbnail": None,
"slug": "post-title",
}
yield mocker.patch(
"search.serializers.ESPostSerializer.base_serializer",
return_value=mocker.Mock(data=base_serialized_data, _get_user=mocker.Mock()),
)
def get_plain_text(self, instance):
"""Return plain text content"""
if instance.article is not None:
return render_article_text(instance.article.content)
elif instance.is_self:
return markdown_to_plain_text(instance.selftext)
else:
return instance.preview_text
def plain_text(self):
"""Returns a plaintext represention of the post"""
if getattr(self, "article", None) is not None:
return render_article_text(self.article.content)
elif self.post_type == LINK_TYPE_SELF:
return markdown_to_plain_text(self.text)
elif self.post_type == LINK_TYPE_LINK:
return self.preview_text
return None
def default_post_response_data(channel, post, user):
"""
Helper function. Returns a dict containing some of the data that we expect from the API given
a channel, post, and user.
"""
# For some reason, the default values are different for staff and non-staff users
if user.is_staff:
user_dependent_defaults = {"upvoted": False, "num_reports": 0}
else:
user_dependent_defaults = {"upvoted": True, "num_reports": None}
post_obj = Post.objects.get(post_id=post.id)
article = Article.objects.filter(post=post_obj).first()
text = post.text
if not text and not post.url:
text = ""
if article:
plain_text = render_article_text(article.content)
elif text:
plain_text = markdown_to_plain_text(text)
else:
plain_text = None
return {
"url": post.url,
"url_domain": urlparse(post.url).hostname if post.url else None,
"cover_image": None,
"thumbnail": None,
"text": text,
"article_content": article.content if article is not None else None,
"plain_text": plain_text,
"post_type": post_obj.post_type,
"title": post.title,
"removed": False,
"deleted": False,
"subscribed": False,
"score": 1,
"author_id": user.username,
"id": post.id,
"slug": get_reddit_slug(post.permalink),
"created": post.created,
"num_comments": 0,
"channel_name": channel.name,
"channel_title": channel.title,
"channel_type": channel.channel_type,
"profile_image": image_uri(user.profile),
"author_name": user.profile.name,
"author_headline": user.profile.headline,
"edited": False,
"stickied": False,
**user_dependent_defaults,
}
def test_update_post_text(mocker, reddit_submission_obj):
"""
Test that update_post_text calls the indexing task with the right parameters
"""
patched_task = mocker.patch("search.task_helpers.update_document_with_partial")
update_post_text(reddit_submission_obj)
assert patched_task.delay.called is True
assert patched_task.delay.call_args[0] == (
gen_post_id(reddit_submission_obj.id),
{
"text": reddit_submission_obj.selftext,
"plain_text": render_article_text(reddit_submission_obj.article_content),
},
POST_TYPE,
)
def update_post_text(post_obj):
"""
Serializes post object text and runs a task to update the text for the associated ES document.
Args:
post_obj (praw.models.reddit.submission.Submission): A PRAW post ('submission') object
"""
update_document_with_partial.delay(
gen_post_id(post_obj.id),
{
"plain_text": render_article_text(post_obj.article_content),
"text": post_obj.selftext,
},
POST_TYPE,
)
def reddit_submission_obj():
"""A dummy Post object"""
article_content = {"text": "some text"}
return SimpleNamespace(
author=SimpleNamespace(name="testuser"),
article_content=article_content,
plain_text=render_article_text(article_content),
subreddit=SimpleNamespace(
display_name="channel_1", title="Channel", subreddit_type="public"
),
selftext="Body text",
score=1,
created=12345,
id="a",
title="Post Title",
num_comments=1,
is_self=True,
likes=1,
banned_by=None,
edited=False,
permalink="http://reddit.local/r/channel_1/a/post-title",
)
def test_render_article_text(input_node, output_text):
"""render_article_text should extract the text from any arbitrary article node tree"""
assert render_article_text(input_node) == output_text