These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE Technology Lab

 
  • Topic is locked indefinitely.
 

Importing market prices to Excel

Author
#1 - 2016-08-22 19:28:56 UTC
Hey Guys

So I have a really cool spreadsheet for seeding markets, Works really well.... Apart from the fact that EVE-Centrals market prices API has become so innaccurate that it is now practically unusable. First question... Can someone explain what happened and why? It worked well for so long and now at least 50% of the sell prices I get for Jita are lower than the buy prices i'm getting from the same station.


Because of the above I am now looking at other ways of getting market prices into my excel spreadsheet. Ideally I would love a simple xml api like EC that will give me the prices I need.

I have been told that for the amount of items I'd like to import I should create my own database. Sounds great apart from the fact I am no where near technical enough to know how to even start that.

I know there are ways of doing it on gdocs but I really don't want to migrate a spreadsheet this big over there, and due to the amount of items it would be super slow.


I just want my EVE Central data back......


Thoughts? Ideas?
#2 - 2016-08-22 19:39:08 UTC
From what I know/remember is that eve-central used to get its prices from lots and lots of people using evemon scraping market logs to get the prices as there was not any other way to get price data at that time. Fast forward to now CCP's new CREST endpoint allows the user to get prices from eve servers not from scraped logs, every 5min and ofc far more accurate.

In addition to that market scraping got banned in 2013 which made eve-central kinda useless. After that eve-central started supplying crest market data as well but for whatever reason its just not real accurate.



As to the getting market data into excel check out this link towards the bottom labeled 'power query'
https://market.fuzzwork.co.uk/api/
#3 - 2016-08-23 13:31:56 UTC
Thanks for the explanation. It just seemed to go from mostly reliable data to completely useless over a very quick period of time. I guessed it might be something with Citadels but that doesn't explain why some prices (eg. Thorax and Vexor) were coming in with sell order percentile at least 10-20% below what the apparent buy order percentile.

I've been fiddling with PowerQuery and Fuzzies API. Looks like it might be a better option, just a PITA to go through and redo all of the connections.

Is there any best practice with Power Query? Should I shove as many id's into a query as possible or space them out into smaller groups?
#4 - 2016-08-23 14:04:01 UTC
I have no idea. I don't use excel.

However if it works like a normal GET request, then you can fit a good amount of ids in before you get the the max length for a url. A few big ones would be better than a bunch of smaller ones.
Test Alliance Please Ignore
#5 - 2016-08-24 19:11:04 UTC
Ly'coN wrote:
Is there any best practice with Power Query? Should I shove as many id's into a query as possible or space them out into smaller groups?


I organize calls into tables in groups of 10 to 12 items associated by groups (ie minerals, ores, etc) makes it easier to look at on the page.

Cheers C.

@Caerfinon - Twitter

Forum Jump