Closed gjost closed 4 years ago
Problem reoccurred on stage site. Here is a sample of the affected area:
<div class="authors row">
<div class="span3">
<a href="">mw_api_url</a><br/>
<a href="">title_sort</a><br/>
<a href="">public</a><br/>
<a href="">url_title</a><br/>
<a href="">article_titles</a><br/>
<a href="">modified</a><br/>
<a href="">published</a><br/>
<a href="">title</a><br/>
<a href="">body</a><br/>
</div>
It's printing just the field names. View code:
def authors(request, template_name='wikiprox/authors.html'):
return render(request, template_name, {
'authors': models.Author.authors(num_columns=4),
})
Running this in python shell returns a list of Author objects:
>>> from wikiprox import models
>>> models.Author.authors(num_columns=4)
[<wikiprox.models.Author "Kaori Akiyama">, <wikiprox.models.Author "Jasmine Alinder">, <wikiprox.models.Author "Emily Anderson">, <wikiprox.models.Author "Megan Asaka">, <wikiprox.models.Author "Gil Asakawa">, <wikiprox.models.Author "Allan W. Austin">, <wikiprox.models.Author "Eiichiro Azuma">, <wikiprox.models.Author "Anne Blankenship">, <wikiprox.models.Author "Matthew M. Briones">, <wikiprox.models.Author "Kayla Canelo">, <wikiprox.models.Author "John Christgau">, <wikiprox.models.Author "Sheila H. Chun">, <wikiprox.models.Author "Bonnie J. Clark">, <wikiprox.models.Author "Tom Coffman">, <wikiprox.models.Author "Jennifer Cronk">, <wikiprox.models.Author "Klancy de Ne
It appears that models.Author.authors
is not returning the answers in columns as requested.
The cache key in models.Author.authors
did not include the num_columns
arg.
def authors(num_columns=None):
KEY = 'encyc-front:authors'
data = cache.get(KEY)
if not data:
searcher = search.Searcher()
If the function was called without the arg then the authors page would break. If the function was called with the arg i.e. by the authors list page, then something else probably broke.
Fixed in encycfront
commit f11455c
for package encycfront-develop_4.0.0-rc2~deb9
.
Manual fix made to encyc-front in production since I expect to upgrade to 4.0.0 tomorrow.
Running
redis-cli flushall
on the server fixes the problem, but it only re-occurs later.