Recently i always got this
PG::QueryCanceled exception raised
ERROR: canceling statement due to statement timeout
1) What search were you trying to do? 2) Do you have the "safe mode" or "deleted post filter" account options turned on?
Posted under General
This topic has been locked.
Getting this error when I try to upload an image from twitter using the bookmarklet
NoMethodError exception raised
undefined method `first' for nil:NilClass
app/logical/sources/strategies/twitter.rb:30:in `get'
app/logical/downloads/rewrite_strategies/twitter.rb:33:in `source'
app/logical/downloads/rewrite_strategies/twitter.rb:12:in `rewrite'
app/logical/downloads/file.rb:49:in `block in before_download'
app/logical/downloads/file.rb:48:in `each'
app/logical/downloads/file.rb:48:in `before_download'
app/controllers/uploads_controller.rb:10:in `new'
https://twitter.com/gen020217paku/status/931525783583039493 is the image in question
kudus said:
1) order:comment kancolle and azur lane
2) safe mode off, deleted post filter off
Try kantai_collection order:comment_bumped and azur_lane order:comment_bumped. order:comment_bumped is faster than order:comment and it excludes comments from bots.
DeusExCalamus said:
https://twitter.com/gen020217paku/status/931525783583039493 is the image in question
https://twitter.com/gen020217paku/status/931525783583039493 is just a reply by some guy. The actual image is https://twitter.com/RamchiPixiv/status/931509225070542848.
evazion said:
Try kantai_collection order:comment_bumped and azur_lane order:comment_bumped. order:comment_bumped is faster than order:comment and it excludes comments from bots.
https://twitter.com/gen020217paku/status/931525783583039493 is just a reply by some guy. The actual image is https://twitter.com/RamchiPixiv/status/931509225070542848.
Whoops, gg me.
A minor thing I noticed for the mobile view.
The 'search pools' input is too big and fucks up the page layout - https://my.mixtape.moe/jrnxfe.png
DanmakuShooter said:
A minor thing I noticed for the mobile view.
The 'search pools' input is too big and fucks up the page layout - https://my.mixtape.moe/jrnxfe.png
The following custom CSS code fixes that. It's actually a portion of the mobile layout CSS I developed in topic #14077.
header#top menu form input {
width: 90vw;
}
Chiera said:
I don't know if it's only for me happening but when using the bookmarklet on any image on NicoSeiga, the page loads indefinitely and doesn'T go to the upload page. Eventually, I get a Failbooru.
Me too, just second ago cannot upload from seiga
Edit: i still can upload from seiga but fetch source data doesnt work
Updated
BrokenEagle98 said:
The following custom CSS code fixes that. It's actually a portion of the mobile layout CSS I developed in topic #14077.
header#top menu form input { width: 90vw; }
Thanks. Another thing that bugges me is that into mobile mode there isn't a way to skip pages or go to the last/first page of a forum topic for example. But it's the same for anything that has multiple pages.
Why has that been removed? It was useful.
DanmakuShooter said:
Thanks. Another thing that bugges me is that into mobile mode there isn't a way to skip pages or go to the last/first page of a forum topic for example. But it's the same for anything that has multiple pages.
Why has that been removed? It was useful.
If I were to guess, I'd say preservation of screen real estate. There's not a whole lot of screen width in portrait orientation.
evazion said:
Site update
Features
- Added typo autocorrection during tag autocompletion (issue #3406):
- Example: type "miku_hatsune" and autocomplete will suggest hatsune_miku.
- This only kicks in when you type something that doesn't match any existing tag. Autocomplete is otherwise unchanged.
- The suggestions it gives could be improved. Please report cases where it gives poor results (e.g. "I typed head chain looking for forehead chain but got on head instead").
Can this be changed so that sticking copy: or char: etc. at the start is ignored?
I noticed when typing copy:tensei the top results are:
megami_tensei
tenseiani
mushoku tensei
copy x
edo tensei
comic mugen tensei
tense
when there are 13 tags beginning with "tensei" (probably 14 by the time you read this), of which 11 (12) are copyright ones (none of which appear in the dropdown at all, although both the non-copyright ones do) which the person might actually be attempting to enter. Tensei shitara slime datta ken is the top result by the time you reach "tensei_" but when doing copy: at the start it doesn't even appear on the dropdown until "copy:tensei_shita" and doesn't get above shin megami tensei until you reach "copy:tensei_shitara_sl"
kuuderes_shadow said:
Can this be changed so that sticking copy: or char: etc. at the start is ignored?
I noticed when typing copy:tensei the top results are:
megami_tensei
tenseiani
mushoku tensei
copy x
edo tensei
comic mugen tensei
tensewhen there are 13 tags beginning with "tensei" (probably 14 by the time you read this), of which 11 (12) are copyright ones (none of which appear in the dropdown at all, although both the non-copyright ones do) which the person might actually be attempting to enter. Tensei shitara slime datta ken is the top result by the time you reach "tensei_" but when doing copy: at the start it doesn't even appear on the dropdown until "copy:tensei_shita" and doesn't get above shin megami tensei until you reach "copy:tensei_shitara_sl"
Try using copyright instead of the shorthand version for now. I have verified that the full form does work. Regardless, I'll submit an issue for the above.
Created issue #3455.
search[id], search[created_at], and search[updated_at] parameters to all index endpoints. These use the same syntax as the id: and date: metatags, so you can do things like https://danbooru.donmai.us/tags?search[created_at]=2017-11-01..2017-12-01 (issue #3430).i) (issue #3424).copy:tensei (issue #3455).Autocomplete takes a lot of time to load.
Manually typing is faster, even tags like eyebrows_visible_through_hair.
albert said:
First off, I want to apologize for neglecting the forum during the migration. I wanted to focus on getting all the critical bugs fixed.
I think we're now at the point where things have stabilized. The issues with uploads earlier were server/database related and were not caused by code bugs, and I've set up some monitoring tools to make sure I get alerted when start failing. I've been keeping a close eye on Github to keep track of issues. I also want to thank everyone who took the initiative to sign up and submit issues. You've all proven invaluable in me getting the site up and running smoothly.
That being said, I'm sure people still have some issues they want me to hear about and I'm now at a point where I can start listening. I just have one request: I want to focus on two things: (1) bugs and (2) inconsistencies with the old Danbooru. Please save any major enhancements or suggestions for the future. I'll get to those in due time.
Even if you don't have a Github account I encourage you to visit the issue list to see if there's anything that stands out, whether you strongly agree or disagree with an issue.
Im getting the nobody here but us chickens treatment when searching multiple tags, please fix
SnoopDoge420 said:
Im getting the nobody here but us chickens treatment when searching multiple tags, please fix
What tags are you searching for? Are you searching for explicit tags on https://safebooru.donmai.us ?
If that's the case, you need to switch to the main site at https://danbooru.donmai.us, as safebooru will hide all content rated under questionable or explicit.
When an image from a new artist was uploaded, and the creator doesn't aware of the way to create new tag, the creator of the tag might simply create the artist name as a general tag. Afterward, when someone correct it by adding the prefix artist: in front of the artist name tag, upon saving, the danbooru would display an error saying "Unable to delete artist_name. Please check for implications." which I think this error shouldn't occur as the tag is still intended to be there.
c933103 said:
When an image from a new artist was uploaded, and the creator doesn't aware of the way to create new tag, the creator of the tag might simply create the artist name as a general tag. Afterward, when someone correct it by adding the prefix artist: in front of the artist name tag, upon saving, the danbooru would display an error saying "Unable to delete artist_name. Please check for implications." which I think this error shouldn't occur as the tag is still intended to be there.
What tags were you trying to use? The above error should only occur if an implication relation exists. In other words, there should be an implication ??? -> artist_name. I can't really say more than that without knowing what post you were trying to change, and what tags you were trying to change.