SerinApprox | mIRC Update Approximation Script

[Hobbesistan]

SerinApprox 1.2

What is it?

SerinApprox is a system used by Serina - the IRC bot I set up for Lazarus, The East Pacific, and the Lazerene Liberation Army. The system uses NationStates++ API to take Timestamps and calculate them into a useable estimation of updatetime.

How does it work?

Serina uses NationStates++(as previously mentioned), which calculates updatetime based on things like issues answered, delegate switches, ect. This will be far more accurate on more active regions then systems like FriarTuck, who guesstimate based off of the NationStates data dumps, NS++ (And Serina as a result) use hard calculated data. This is going to be far more accurate on more active regions (Serina has been 2-3 minutes more accurate then FT on several situations - which obviously is life and death during a update) but will suffer on more inactive regions because no data has been collected.


Do I need NS++ to make it work?

No - It reads data directly from the NS++ website.


Who to credit?

I would be lieing if i took full credit for the project, I suuuuck with sockets and the like, I'd infact argue most of the credit goes to the scripting help channel in EsperNet, the guys there are exceptionally awesome.

How to use it?

Plug it into the 'Remote' tab of the mIRC script editor, then .approx (region) and if the bot is set up right, it will spit out the region, guesstimated approximation, and even by how much this estimation may be off.

IMPORTANT: Change #YOURCHANNEL in the .approx command to what channel you want it to work in!

<sidenote> You'll need to use underscores (_) for regions that are multiple words (ex: The_East_Pacific instead of The East Pacific), this is really easilly fixable but I am lazy.</sidenote>

Where do I get it?

Here(1.2) (7/29/2015)
Here(1.0) (6/28/2014)
Here(0.97) (4/08/2014)
Here(0.96) (2/20/2014)

Should I expect updates/support?


I'm generally lurking on #hobbeshost on espernet, Please don't TG me for support over this; I get enough telegrams as is.

Cormac: Does it report back to me/the LLA?

Sadly(kidding), no. The code is published and anyone can look it over assured its not reporting back to the hobbeshost spying center.

[Shadow Afforess]

Looks great Hobbes. I wish more of the bots used across NationStates were available too.

[Cerian Quilor]

I'm glad they're not.

[Tano]

I'm glad they're not.

Because...?

[Shadow Afforess]

I'm glad they're not.

I really don't want to send Hobbes thread off the rails, but basically, you're wrong. The secretive culture around so many groups in NationStates is the source of much of the stagnation in the game. I've been apart, am apart of many other communities and the ability to share ideas, technology is what helps them grow and expand. In my experience, much of the culture of long-term players of NationStates is incredibly negative, and actively rejects new ideas and changes. Anyone clearly not "in-the-know" is slammed hard. A clear example of this is the recent R/D Guide that was posted in Gameplay. None of the experienced players offered to help co-write or improve the guide. Most were dismissive or intentionally made negative remarks. That is just a microcosm of what happens all over in NationStates.

I'm not trying to lay blame or say Cerian or others are at fault here. I am just glad Hobbes has made a step towards reversing the stifling culture in NS.

[Hobbesistan]

Having been bugged over TG for help on this, I ask that anyone seeking support goes on #hobbeshost (esper) instead of spamming my TG inbox.

I also will be releasing a version of this that calculates minor update within the next few days.

[Hobbesistan]

0.97
Now with .minor update approximation!

http://pastebin.com/aq4gf6L7


Same install, set #YOURREGION on both .approx and .minor to your appropriate channel.

[Hobbesistan]

Newest NS++ backend update just jacked up SerinApprox, i'll have a fix up soon™

[Mallorea and Riva]

Out of curiosity just how accurate is this?

[Misley]

Out of curiosity just how accurate is this?

As accurate as NS++ is, which makes it useful to check roughly when a region updates but not accurate enough to time jumps from.

[Hobbesistan]

Out of curiosity just how accurate is this?

It uses NS++ backend, this script will give you the rough time a region updates as well as the variance in seconds that the calculation may be off.

I've tested it with most XML scripts and found that for more active regions (because it relies on actual activity) this script is almost always right, but if its a super-inactive region, because no data has been collected, the result will be skewed.

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

[Mallorea and Riva]

Out of curiosity just how accurate is this?

It uses NS++ backend, this script will give you the rough time a region updates as well as the variance in seconds that the calculation may be off.

I've tested it with most XML scripts and found that for more active regions (because it relies on actual activity) this script is almost always right, but if its a super-inactive region, because no data has been collected, the result will be skewed.

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

Right but how wide is the variance window on the calculations?

[Misley]

It uses NS++ backend, this script will give you the rough time a region updates as well as the variance in seconds that the calculation may be off.

I've tested it with most XML scripts and found that for more active regions (because it relies on actual activity) this script is almost always right, but if its a super-inactive region, because no data has been collected, the result will be skewed.

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

Right but how wide is the variance window on the calculations?

It depends on the regional activity level of things that change at update (influence changes, issues answered, delegate changes, etc.). Large, active regions will be fairly accurate (within ~1-2 minutes) but small, inactive regions can be off by several minutes, if there's enough data to calculate an estimate at all.

[Mallorea and Riva]

Right but how wide is the variance window on the calculations?

It depends on the regional activity level of things that change at update (influence changes, issues answered, delegate changes, etc.). Large, active regions will be fairly accurate (within ~1-2 minutes) but small, inactive regions can be off by several minutes, if there's enough data to calculate an estimate at all.

That's more or less what I figured. Any approx that isn't accurate to within ten seconds really isn't useful.

[Hobbesistan]

It uses NS++ backend, this script will give you the rough time a region updates as well as the variance in seconds that the calculation may be off.

I've tested it with most XML scripts and found that for more active regions (because it relies on actual activity) this script is almost always right, but if its a super-inactive region, because no data has been collected, the result will be skewed.

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

Right but how wide is the variance window on the calculations?

I've seen 20-30seconds on more active regions, 15s or so at most.

[Misley]

It depends on the regional activity level of things that change at update (influence changes, issues answered, delegate changes, etc.). Large, active regions will be fairly accurate (within ~1-2 minutes) but small, inactive regions can be off by several minutes, if there's enough data to calculate an estimate at all.

That's more or less what I figured. Any approx that isn't accurate to within ten seconds really isn't useful.

It's not useful to time jumps or anything terribly serious, but it's nice to have a quick glance at a region and be able to tell roughly where in the update order it falls without having to refer to the dumps or a tool that pulls data from the dumps.

[Shadow Afforess]

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

What exactly broke, because I didn't change any existing fields in the backend...

[Hobbesistan]

It's not quite working right now due to a update in NS++ backend, i'll have a patch out soonish.

What exactly broke, because I didn't change any existing fields in the backend...

I have it set up in a numbered system (it reads the variables as numbers, i.e 1- for variance), because of the new update_order, its reading the wrong variables which spews out a stupid number (i.e <@Aiden> Major Update Calculation:  194:56:58 (+/-  16747  seconds ) )

A pretty easy fix when I have time.

[Shadow Afforess]

Ah, gotcha. The update was part of a few changes that allow NS++ to give updates for all regions, even regions with no data available.

[Misley]

So I'm working on a Python version of this to use in XChat (a clearly superior IRC client ) and thought I'd share the fixed version of the mIRC script for anyone who wants to use it but doesn't care enough to figure out how to do it themselves:

Code: Select all

SerinApprox 0.98
~~~~~~~~~~~~~Minor~~~~~~~~~~~~~~~~~~~~
on *:text:.minor &:#YOURREGION:{
  if !$sock(pluspluss) { sockopen pluspluss nationstatesplusplus.net 80 | set %regionplus $2 }
  set %yourchannel #YOURREGION
  else notice $nick Wait!
}
on *:sockopen:pluspluss:{
  sockwrite -n $sockname GET /api/region/updatetime/?region= $+ %regionplus HTTP/1.0
  sockwrite -n $sockname $crlf
}

on *:sockread:pluspluss: {
  sockread &plus
  if $regex($bvar(&plus,1-).text,/"([^"]+)":("[^"]*"|[^"{][^,}]+)/g) {
    var %time = $calc($regml(12) / 1000 - 18000)
    if (%time < 0) %time = $calc(%time + 86400)
    msg %yourchannel  Minor Update Calculation:  $duration(%time,3) (+/-  $floor($calc($regml(10) /1000))  seconds )
  }
  unset %regionplus
}
~~~~~~~~~~~~~~Major~~~~~~~~~~~~~~~~~~~~~
on *:text:.approx &:#YOURREGION:{
  if !$sock(plusplus) { sockopen plusplus nationstatesplusplus.net 80 | set %regionplus $2 }
  set %yourchannel #YOURREGION
  else notice $nick Wait!
}
on *:sockopen:plusplus:{
  sockwrite -n $sockname GET /api/region/updatetime/?region= $+ %regionplus HTTP/1.0
  sockwrite -n $sockname $crlf
}

on *:sockread:plusplus: {
  sockread &plus
  if $regex($bvar(&plus,1-).text,/"([^"]+)":("[^"]*"|[^"{][^,}]+)/g) {
    var %time = $calc($regml(28) / 1000 - 18000)
    if (%time < 0) %time = $calc(%time + 86400)
    msg %yourchannel  Major Update Calculation:  $duration(%time,3) (+/-  $floor($calc($regml(26) /1000))  seconds )
  }
  unset %regionplus
}


[Hobbesistan]

Heres 1.0
http://pastebin.com/CQ3eqdxD

[Hobbesistan]

Version 1.2 is now live:

• Adjusted socket system now that nationstatesplusplus.net forces HTTPS
• Fixed what variables it calls upon as the NS++ backend now uses different numbers for different veriables
• Fixed a annoying bug with bolding.

http://pastebin.com/eU1GHRRF