| # |
Jul 17th 2018, 12:15 |
 admad |
@birdy247 @neon1024 endless scroll is the most annoying thing frontend devs have come up wiht |
| # |
Jul 17th 2018, 12:03 |
snake-venom |
can anyone here help me in this ? |
| # |
Jul 17th 2018, 12:02 |
snake-venom |
here is my nginx file |
| # |
Jul 17th 2018, 12:02 |
snake-venom |
https://pastebin.com/tSe7gdfP |
| # |
Jul 17th 2018, 12:02 |
snake-venom |
i have a folder /plugins and in this i want to run 5 cakephp projects.. but i am not getting what to add in config nginx file... |
| # |
Jul 17th 2018, 11:50 |
neon1024 |
Make sure you lint your JS, because if that breaks you’ll never get page 2 ;) |
| # |
Jul 17th 2018, 11:49 |
neon1024 |
You’d have to disable JS to use the footer :rolling_on_the_floor_laughing: |
| # |
Jul 17th 2018, 11:49 |
neon1024 |
I can’t imagine the horror if you have endless scroll, and only put certain links in the footer |
| # |
Jul 17th 2018, 11:48 |
neon1024 |
It’s like the kitty who wants to be stroked, but when you get close just wanders out of hands reach |
| # |
Jul 17th 2018, 11:48 |
neon1024 |
For the love all things holy, don’t have a footer. Please don’t have a footer. |
| # |
Jul 17th 2018, 11:47 |
neon1024 |
Hah, hate those things. |
| # |
Jul 17th 2018, 11:47 |
birdy247 |
FYI, we are using a endless scroll on the frontend |
| # |
Jul 17th 2018, 11:47 |
birdy247 |
just wondered if there was a less expensive way of quering all books every time |
| # |
Jul 17th 2018, 11:47 |
birdy247 |
so the plan is to query the DB for ALL books, pop into a collection and then take the necessary |
| # |
Jul 17th 2018, 11:47 |
neon1024 |
Yeah, so you’ll need to get all teh books, calculate the score, setup the collection and then paginate that new collection I guess right? |
| # |
Jul 17th 2018, 11:46 |
neon1024 |
Books right? So you’re getting all the books anyway |
| # |
Jul 17th 2018, 11:46 |
birdy247 |
1 User would want to see books with a score > 8 |
| # |
Jul 17th 2018, 11:46 |
neon1024 |
Well you either paginate Books, or you paginate Users. |
| # |
Jul 17th 2018, 11:45 |
neon1024 |
Okay a trait, not an interface, but I was almost right! |
| # |
Jul 17th 2018, 11:45 |
neon1024 |
https://github.com/cakephp/cakephp/blob/master/src/ORM/ResultSet.php#L34 |
| # |
Jul 17th 2018, 11:45 |
birdy247 |
@neon1024 If we paginate a collection, thats all good. but we still need to retrieve all books from the sql. |
| # |
Jul 17th 2018, 11:44 |
neon1024 |
Shouldn’t be too hard, as I think ResultSet implements CollectionInterface iirc |
| # |
Jul 17th 2018, 11:43 |
matt |
thanks, yes think we will go with paginating the collection manually but with caching |
| # |
Jul 17th 2018, 11:42 |
neon1024 |
At least then you alleviate some of the pressure, in exchange for slightly stale data |
| # |
Jul 17th 2018, 11:42 |
neon1024 |
Keyed by user id or something |
| # |
Jul 17th 2018, 11:42 |
neon1024 |
I guess you could cache the calculated user results |
| # |
Jul 17th 2018, 11:41 |
neon1024 |
It works. Is it perfect? Not by half ;) |
| # |
Jul 17th 2018, 11:41 |
neon1024 |
Meaning my request / response time was still snappy |
| # |
Jul 17th 2018, 11:40 |
neon1024 |
Which is why I ended up with a shell, and I could “front-load” the data |
| # |
Jul 17th 2018, 11:40 |
birdy247 |
I think we will see how big the problem is |
| # |
Jul 17th 2018, 11:40 |
neon1024 |
Away from the user request is #1 |
| # |
Jul 17th 2018, 11:40 |
neon1024 |
It’s expensive, it just a matter of choosing where you want to put the expense |
| # |
Jul 17th 2018, 11:39 |
neon1024 |
Yep, and the aggregate table will be huge |
| # |
Jul 17th 2018, 11:39 |
birdy247 |
If we have 10,000 books and 20,000 users, could be quite a cron! |
| # |
Jul 17th 2018, 11:39 |
neon1024 |
Worked for me when I had to calculate bespoke pricing for every API consumer and every package |
| # |
Jul 17th 2018, 11:38 |
neon1024 |
Or on a cron, or both |
| # |
Jul 17th 2018, 11:38 |
neon1024 |
Perhaps either in a worker queue kicked off when the user logs in |
| # |
Jul 17th 2018, 11:38 |
neon1024 |
Then I can update that aggregates table on a per-user basis |
| # |
Jul 17th 2018, 11:38 |
neon1024 |
Then I’d have a shell which processes all the books for every user |
| # |
Jul 17th 2018, 11:38 |
neon1024 |
If it’s really expensive I’d probably do an aggregate table |
| # |
Jul 17th 2018, 11:37 |
neon1024 |
I think I’ve drunk too much tea and eaten too much pick and mix today :shocked_face_with_exploding_head: |