I’m trialing newsblur internally for my company, with my limited knowledge in python i’m unable to figure out why addsite does not bring any match as it does in newsblur.com site.
For example typing “hindu” on newsblur site brings up a list of sites that matches that term with the amount of subscribers using them, where as in the internal installation it does not.
Any pointers on where to look and what to configure will be a big help.
Thanks in Advance,
Parthi
Not sure if this is the reason in your case, but there’s a minimum number of subscribers that a feed needs to have (or have had at some point) before it will appear in the autocomplete, to prevent single-user private feeds from showing up.
Sure, you need to have elasticsearch running. Once running, it will not auto-populate. I assume you’ve setup celeryd to run the beat events. Well, I didn’t make the search index for feed title/address autocomplete a celerybeat task. I run it manually. Since the feeds that would match a phrase don’t change very often, I just run a single command when I want to populate it.
./manage.py update_index -b 1000 -w 4
Thanks guys,
was away for the past couple of day’s thats why replying to your comments late.
Good to know that you can run this task and populate this.
it would be beneficial if we can get a list of documents administrators can run and do certain task which are meant to improve the user experience of newsblur.
Thanks for your help.
That’s the only one that isn’t automated. I suppose i should automate it to run weekly (along with the dozen other tasks which are already on celerybeat to run regularly, including feed fetching every minute and clearing old accounts once a day).
Hi Sam,
We manage to run this task, but no joy.
./manage.py update_index -b 1000 -w 4
Unknown command: ‘update_index’
I’m running this from NewsBlur directory where i had cloned from github.
Am i missing anything?
Thanks for your help.