Just now started loading for me.
Yup loading now… im surprised that between the time you posted that and now that another 18 planets rolled…
1196 planets found… as of 9:35am est
The data is cached for one hour so there’s that but it’s still a ton
According to the update notes, type selection was added to the world builder today. I’m sure people are playing with that.
Hi, I am really fond of three data extraction and display function!
There are (as always) three questions 
1/ Is is possible to add a "select planet(s) by name(-mask/pattern)?
2/ The “is editable” property is shown as “no” even when I have activated the “allows to modify blocks…” in the world controller. Is this property something different?
3/ A sovereign / creative planets are orbiting around a planet with 1 blinksecs. Which planet this is would be of some help setting up portals. Is this info somewhere obtainable?
Good work!
- I had a filter for planet name there but it wasn’t getting used much so I removed it. I can add it back if you’d like.
- You have the right property but your planet likely needs a new scan from someone with the BoundlessProxyUI - Help track Sovereign World Colors + More! tool.
- Closest planet (assigned planet) was displayed beneath the planet name but it appears I must have accidentally removed it. Adding it back now.
1 Like
FYI the buttons in the top-right of each planet are styled a little wonky right now everyone. I’ll be fixing it shortly.
Since getting a 504 gateway error from the dump endpoint seems to be a common thing, would it be possible to get some kind of error handling for that?
I’m getting an error today
“Unfortunately, something went wrong. Click below to try again or refresh the page…”
I tried a hard refresh.
Taking a look right now.
Edit: It loads up for me:
One thing worth mentioning is that the data I’m using is now up to 18MB in size (I’m going to have to solve this problem sooner than later because 18MB is already way too much to transfer for most folks). That data gets cached on the API side once every hour and it takes almost 2 minutes to DO the caching. During those 2 minutes, people will see the error message and there’s nothing I can really do about that. Just try again in a few minutes.
Yup, I think its time to cut down on the reliance of Javascript a bit and process/cache the 18MB of data server-side and paginate it server-side before serving it to users.
Pagination is already there server-side afaik but the API isn’t set up for the kind of filtering I’m doing client-side. The filtering and sorting I’m providing folks is purely client-side and requires that I have ALL of the data. There are two ways forward as far as I see it:
- Add the filtering and sorting options that I’m currently doing client-side and offer them up as part of the API instead.
- Add a PS Planet Explorer-specific endpoint for retrieval of all worlds + blocks colors + latest poll and trim out everything that is absolutely not needed to cut down the data size as much as we can.
1 is probably the better approach but it will require @Angellus to do a significant amount on adding the necessary filters and sorting. It would also tightly couple @Angellus and I going forward such that i I want to add new features to the Planet Explorer, it would always have to be done in tandem with a new feature added to Boundlexx.
So fetching the data from the api and processing it the way you need it server side (on portalseekers.com) is out of the question? Reason why I say that is then you can format the data the way YOU need it, and not require @Angellus to continue changing the api.
Processing it and caching it server side (on portalseekers.com) would GREATLY reduce the stress on @Angellus servers as well.
2 Likes
Sure I can cache it server side on my end but then you’ve got double caching and you’re going to have greatly out of date data. So why even have a tool and why even have an API?
1 Like
define double caching. I don’t think the API itself caches its data.
Angellus is already caching the data I’m using for an hour. The data I retrieve is anywhere from 1-60 minutes old. This is why active exos don’t show up right away in my tool. Now I can get around that particular problem by making a separate API call on load to check explicitly for exos but you get the point.
If I add server-side caching on my end (i.e. stand up a MySql database to store the data and then cache stuff there and use it to pull data into my app) then we have double-caching.
All of the filters you are currently using already exists:
- Type:
is_perm,is_exo,is_sovereign,is_creative,show_inactiveandactivefilters on the Worlds API. ex: https://api.boundlexx.app/api/v1/worlds/simple/?is_sovereign=true&is_creative=False&active=false - Tier: the
tierfilter on the Worlds API. Ex: https://api.boundlexx.app/api/v1/worlds/simple/?tier=4 - Atmosphere:
world_typefilter on Worlds API. Ex: https://api.boundlexx.app/api/v1/worlds/simple/?world_type=LUSH - Region:
regionfilter on Worlds API. Ex: https://api.boundlexx.app/api/v1/worlds/simple/?region=use - Can Visit:
is_publicfilter on Worlds API. Ex: https://api.boundlexx.app/api/v1/worlds/simple/?is_public=false - Filter By Color:
- If you want to let the user search their color.
searchfilter on Colors API +blocks_urlin results. Ex: https://api.boundlexx.app/api/v1/colors/?search=red → https://api.boundlexx.app/api/v1/colors/78/blocks/ - If you want to let users select color from list. Skip straight to Color/Blocks URL. Ex: https://api.boundlexx.app/api/v1/colors/78/blocks/
- If you want to let the user search their color.
- Check for Colors: Sovereign Colors on Items API. Ex: https://api.boundlexx.app/api/v1/items/9555/sovereign-colors/
- Sort by resource:
orderingfilter on the Resource Counts endpoint. Ex: https://api.boundlexx.app/api/v1/items/32803/resource-counts/?ordering=-percentage
You just have to stop trying to show all of the data for all of the worlds at once. The Boundless Information Station has all of the filters mentioned above and more, but there are separate views for different types of data. There are over 20k World Block Colors combinations for over 1k worlds. You cannot easily show that all in a single API request or even show that in a browser window. Planet Explorer lags my browser hardcode and it takes almost a full minute for the page to re-render after I resize the window.
3 Likes
Same here, and I have an i9 9900k with 32GB of DDR4 RAM. The excessive use of Javascript to clientside everything doesn’t help.
You sound like someone who just hates js. You probably love php though and php, just like jsp, is garbage. I’m full stack btw. I am equally experienced with both Java as well as html, css and js. The reason I don’t have a “server side” to my app is because for hosting, you don’t get a Java or Node (either would be fine for me) install unless you upgrade to the premium tier.
I can refactor the app as most of what I need from a filtering perspective is already there (not the sorting though) but I’m inclined to just walk away and shut it down as there are clearly some members of the community who don’t appreciate it.
Node is better than PHP, but how dare you
Jokes aside, my opinion, as someone who uses your app, your design is significantly more powerful than the other tools available, and I would love to see this design applied with a more efficient setup.
I’m full stack too, working with PHP/React professionally, but preferring node for back-end as well, I just don’t think browsers, especially mobile browsers, are built to handle this amount of data. Yes, JS is fantastic, and I know I’m not who you made that comment to, but browsers suck at holding that much data.
I hope everyone’s feedback on your implementation doesn’t discourage you from continuing the project, as I would love to see it continued and improved.