Pages: [1] :: one page |
|
Author |
Thread Statistics | Show CCP posts - 0 post(s) |
Haguu
Caldari School of Applied Knowledge
|
Posted - 2009.10.09 02:43:00 -
[1]
I started through the EO api doc and did a search here but was unsuccessful so I need to ask.
I thought I read you are only supposed to ask for API information once an hour. Is that per account or per program? E.g., if you have eveboard and eve-sheet external and MEEP and HQ and EFT and EVEMon and a spreadsheet on your client, then if each one keeps a 1-hour cache is it OK even if that is 7 API requests per hour?
If it is not OK, then do you have to host your own website to do the cacheing? A local server would work for programs but not websites or Google Spreadsheets.
What do people with multiple programs do? Recommendations?
Thanks in advance
|
Johnathan Roark
Caldari Quantum Industries RAZOR Alliance
|
Posted - 2009.10.09 03:20:00 -
[2]
Each API returns a cachedUntil time. Your program should store that value and not submit a new request until that timer has expired. Each api varies between 15 mins and 24 hours. I've found that most of the strict timers seem to be tied to IPs.
Quantum Industries is recruiting! |
Elegbara
|
Posted - 2009.10.09 06:39:00 -
[3]
You can ask for information as often as you want but you'll be getting cached version instead (the response does contain field telling you when the data was generated). Thus there's no reason to do it often unless you're using several computers (and each will get the same data this way).
However if you have several characters having access to corporation assets, you can have fresh data on those assets more often then if you had just one character. That is - corporation assets data is cached for 24 hours and cache is identified by what was requested. Since first request was to fetch "Assets of corporation of Character1", you can have another request for "Assets of corporation of Character2" and get up-to-date data even if the first request results are still cached. Open your eyes. And Awaken. |
Vessper
SI Radio
|
Posted - 2009.10.09 07:20:00 -
[4]
EveHQ has it's own in-built API Relay Server which effectively acts as a cache/proxy server for API requests. If any third party application supports it, they can point to this server to obtain their API details. Furthermore, this also works across LAN/WAN so multiple computers running multiple local applications only require a single download of each API to satisfy all requests.
- - - - - - EveHQ Character App |
Hel O'Ween
Men On A Mission
|
Posted - 2009.10.09 12:13:00 -
[5]
Originally by: Johnathan Roark Each API returns a cachedUntil time.
(Mentioning the obvious here, but before anyone scratches his head)
To add to that: the cachedUntil time returned by the API is EVE time. So, make sure to adjust that to your local time.
-- EVEWalletAware - an offline wallet manager |
Haguu
Caldari School of Applied Knowledge
|
Posted - 2009.10.09 19:41:00 -
[6]
Thanks all!
1) I have finally relearned enough to get Ruby,Textmate and EAAL working so I think that covers code I write. I was concerned that if I boot my machine after not doing EVE for over an hour and then start up EveHQ and MEEP and EFT and EVEMon, then my complete guess is that while each are good citizens, and each will hit the eve api server only once per hour, that will be 4 access from my PC/IP address in under a minute. Plus eveboard, evesheet, and any other online things could be hitting it as well. Though in a Google spreadsheet and I am well over my one per hour quota if CCP wants to enforce it. Will they?
Originally by: Elegbara You can ask for information as often as you want but you'll be getting cached version instead (the response does contain field telling you when the data was generated).
Are they clever/nice enough to just allow me to call the API, just that the data will be between 0 and 60 minutes stale depending if another program accessed it previously. An hour stale data is not a problem (and nothing I could do if it was a problem.) I just don't want to be a bad citizen.
2) A way around this would be to run some code (Ruby but perhaps python or java or PHP ) on an external site that I could point all these programs to, at least the ones that support proxies. It can not run locally since the eveboard and similar web aps hopefully can not access things inside my firewall. But should not tax one of the $x/mo low end sites.
Originally by: Vessper EveHQ has it's own in-built API Relay Server which effectively acts as a cache/proxy server for API requests.
The EVEHQ proxy is almost what I want except for two issues. The first is that running the Windows 7 trial in Parallels, I can not get EVEHQ to work reliably. It's a very nice program and after Windows 7 officially launches and EVEHQ officially supports W7 I am sure I will be using it. The second issue is that nothing locally can help for online sites or anything outside the firewall.
3) Being both cheap and spoiled to the "everything on the web should be free" paradigm from the previous millennium is there any free site that does that? Or what I gather is that CCP itself is doing the caching for me so calls from multiple programs should not be an issue and I can just not worry about the number of data consumers I have.
Again, thanks.
|
Sakky Dio
Gallente Blitz Gamers
|
Posted - 2009.10.10 20:02:00 -
[7]
As for 2 from your list, I've developed one that runs purely on PHP, thread about it can be found at: http://www.eveonline.com/ingameboard.asp?a=topic&threadID=1184844
As for 3, the link provided does have access a test one that is up and running all the time on my server, I do use this to handle all my caching, though any API application can cache though this without an issue.
Simple, easy PHP based API Cache, take a look! http://www.eveonline.com/ingameboard.asp?a=topic&threadID=1184844 |
Irongut
H A V O C Against ALL Authorities
|
Posted - 2009.10.12 14:45:00 -
[8]
Originally by: Haguu The second issue is that nothing locally can help for online sites or anything outside the firewall.
You aren't paying attention to the answers you have already been given...
Originally by: Johnathan Roark I've found that most of the strict timers seem to be tied to IPs.
The cache is per method and per IP address. So if you request a skill sheet from your home IP address and a wallet balance from your home IP address these are 2 separate timers and will not interfere with each other. Similarly if you then request a wallet balance from a Google spreadsheet it will come from a diferent IP address (Google's) and will not be effected by the request you made from your home IP address.
-- Please resize your signature to the maximum file size of 24000 bytes. Zymurgist |
|
|
|
Pages: [1] :: one page |
First page | Previous page | Next page | Last page |