Skip to content
TORNLIFE More

Attacksfull

Started by DammitArcher [2036722] on in API Development.

7 replies · 247 views · thread synced · 6 days ago · View on torn.com
About this thread

Posts archived: 8 / 8 posts (100%) · the total is Torn's reply count + the opening post at the last fetch

Counted by TornLife from the archived posts.

Archived posts
8
Discussion span
→
People posting
6
Likes on archived posts
4
Posts by staff, officers and moderators
1
Authority score
72 / 100
Historical score
26 / 100
Story score
33 / 100
Engagement score
54 / 100

Most-liked replies

DammitArcher [2036722]

With the new chaining system, it's important to be able to gather an extended data set in order to calculate the average respect gained by each member (while factoring out bonus hits). Currently, the API calls are limited to 10,000 lines. Would it be possible to extend that so that chains -- which now run to 25k and will eventually run to 100k -- can be effectively tracked and parsed via API?

Alternatively, could a table be created specifically for this purpose and added to the API? The current built-in chain reports imply that the data is already tabled, so adding it to the API would be relatively easy if that's the case.

Thanks! :)
Bug [1455582] Committee Committee

If they do it, it should definitely be separate, otherwise those just wanting what attackingfull correctly provides will have to download 10x the file size when 100k chains are released. In the meantime, why not just have something that pulls from the API once per hour and compiles that into your own list?
DammitArcher [2036722]

Hey, I'm fine with that. Attacksfull was just the closest available table to what I wanted. A new table based on the current chaining reports would be spectacular.
LouBaker [1162207]

I myself save the list in a database by its ID number

when I next download I only collect the latest IDs up until my last save and use the data iv already collected

this is also faster because I have stripped out what I don't need before

when I am looking for attacks from months ago, I can read them from the database


paulwratt [2027970]

sounds like u would benefit fron this API extension:

API server load & bandwidth reduction query option
https://www.torn.com/forums.php?p=threads&f=63&t=16044866&b=0&a=0

in combination with a new db on the Torn side, one that stores the complete list of chains for the last 24-48 hours (how long does it take to do a 10K or 100K chain?)

Even if only the above API extension existed, it would still be a lightweight data pull every 30 min or so, and you would not have to "process out duplicate data", so the actual dtata processing would be quicker too

Mentions: API server load & bandwidth reduction query option

Mafia [610357]

My server side cron every 3 hours to keep attacksfull up to date (and remove duplicate ID) and I remove any defends (win/lose). Only attack will be count on my server. I save up to 50k (will increase up to 100k).. 

When I generating chain report, that will be request 100k(attacks) from my server, not via torn api.

So I dont missed any bonus hit on 10,25,50,100,x,x,x,x on generated chain report .

 

Just now it has 23,229 results (attacks) saved

[image: i.gyazo.com]

cheers..
LouBaker [1162207]

Iv changed my mind about storing 25k lines of data, now I'm just reading and storing each Member's data in an array and throwing the list away, This way only have to store 100+ lines of data

Player data is saved in the same format as my Hitters list 

[image: i87.servimg.com]

so all respect and attacks type data is in one line for each member

XID1234567890=Someone | 45 | 2796.88 | 596 | 539 | | 33| | | 23 | 1 | -4.56 | 14 | 3 | 2 | 5 | | | 4| | | | | | |