news of days long forgotten

the archive

want to look up old infos? find the first news on this site? interested in details about the latest few updates? then read on...

details here as soon as they are available.

monday, march 19th 12:01 2001

æstats⁺⁺ - plan file

i finally finished the obituary files, 100 kb of data. i am sure there are still tons of things to fix in that file, especially the teamplay part, that has not been implemented yet for many of the games. i am finally ready to implement the parser. first step: find out what parameters are possible in what key word lines, and to make sure that the reduced obits syntax, really understands all the ramifications of real world log files.

day 10,11 - tons of obits...

- obits supported:  q3a update, daikatana cleanup,
  new: q2, rocket arena 2, action q2, kingpin, sin,
  heretic ii, team fortress, hexen ii, qw, tribes.
- added new key words: "infos" (to describe the game/mod
  at hand) and "suisrules" to define suicide rules.
- coded a parameter tracker that lists a total of 22
  parameters sofar.

friday, march 16th 14:44 2001

æstats⁺⁺ - plan file

defining a generic syntax for obits seem to be more complicated than i had hoped. e.g. q3a does not have special log line for suicides, if uses frag lines where fragger and fragged are the same or where the fragger is <world>... try to tell this to a general template :)

day 9 - q3a engine obits...

- q3a logs are tracked, along with team arena.
- updated the syntax for star trek voyager - elite force

wednesday, march 14th 11:39 2001

æstats⁺⁺ - plan file

more work done on the obituary syntax. alas this is pretty slow work, since i need to find a concise syntax that is general enough to be useful for other games as well. it would be lot simpler to just hack it together, but this can't be the way to go. to make things worse some logs, especially those of ut have weird 'bugs', like suicides that happen with certain weapons, but the damagetype is not properly set, so that you end up with a redeemerdeath.

day 7,8 - more on obits...

- ngstats log support for ut in there with name tracking.
- team fortress classic - done...

thursday, march 1st 9:51 2001

æstats⁺⁺ - plan file

thanx to maxwell kanat-alexander's latest tfc logs i was able to update the syntax of the tfc obituaries. found one new frag line and several new events.

day 6 - more on tfc...

- team fortress classic - updated...

monday, february 26th 10:16 2001

æstats⁺⁺ - plan file

this weekend i spent mostly on enhancing a few things for good old æstats. it has become totally skinable now. expect a new release in a few days.

for æstats⁺⁺ the obituaries for counter-strike where finished, and those for opposing force and plain half-life where added. i am almost done with team fortress classic support. still i am trying to find a good syntax for obituaries.

day 5 - obituary support for...

- counter-strike
- half-life
- opposing force
- team fortress classic

friday, february 23rd 10:16 2001

æstats⁺⁺ - plan file

started to track special events in the counter-strike obituaries. i am developing a syntax that will work (hopefully) for every game making the parser (par⁺⁺) very flexible, and potentially making it possible to track totally new games without having to code one new line in the parser.

tracking map changes, map names, name tracking via team joins and renames, player id given out by won and ips too, special events/goals are tracked and scores can be assigned to these events. also chat lines will be counted.

day 4 - example obituaries for cs

- MapName "Spawning server \"<mapname>\""
- TeamTracking "<player> is joining the TERRORIST force"
- Connect "\"<player><WON:<playerID>>\" connected,
	address \"<playerIP>\""
- //    Template	"Nice Message"      Score
  Event "<player> has rescued a hostage"
  	"Hostage: Rescued" "+1"
  Event "<player> assasinated the VIP"
  	"Killed the VIP"   "+1"
- Chat "<player> :  "

thursday, february 22nd 9:58 2001

æstats⁺⁺ - plan file

started to put the counter-strike obituaries together. the advantage of external obituaries is... you guessed it... that users can add new weapons or suicides to the list of recognized frag lines, and that on the fly.

the fragtemplate and suistemplate lines define how frags/suicides are presented in the appropriate log file. using keywords that are delimited by < and > æstats⁺⁺ is able to understand those lines. the 'frag' line examples further down lets you 'alias' the found weapons with nice names.

day 3 - example obituaries for cs

- FragTemplate "L <MM>/<DD>/<YYYY> - <hh>:<mm>:<ss>:
               \"<fragger>\" killed \"<fragged>\"
               with <weaponR>"
- SuisTemplate "L <MM>/<DD>/<YYYY> - <hh>:<mm>:<ss>:
               \"<fragged>\" killed <suicide>"
- //   <weaponR> "Nice Weapon Name" "Shortcut"
  Frag "ak47"    "AK-47"            "AK"
- //   <suicide>            "Nice Suicide Name"
  Suis "by world with door" "World: Door"

wednesday, february 21st 10:56 2001

æstats⁺⁺ - plan file

so far so good, but i am already missing c++'s easy way to handle strings and also the memory management... sigh :). from the debugging point of view c is definitely the way to go, since i got the translated code a working in less than 10 minutes.

day 2 - things done and planned

- finished translating c++ parser module code to c.
- starting to develop basic log line recognition
  syntax with general message classes.
- will begin with counter-strike support. this time
  round i will try to track just about all the infos
  possible.

Tuesday, February 20th 9:31 2001

æstats⁺⁺ - start of project

after months of mapping, i have started to get bored with that and will now begin the coding for æstats⁺⁺. the most important info at present is the following: i will not code in ansi c++, i will continue work in ansi c. after looking through the 11 thousand plus lines of aestats code, and seeing how difficult it is in c++ to get the most basic things to work, it would mean months of nerve racking extra debugging work just to get things going. sticking to ansi c will let me use most of my old *working* modules. the two months spent on learning c++ will not be for naught, since this helped me think more clearly in c.

sofar the parser and the stats will be coded in c, i might use some c++ for the brand new output module. these coming weeks i will update this site on a regular basis, giving those of your who are interested a look beind the scenes of the development process (planfiles sorta :). you might spot things that are not going into the right direction and by sending in an email might help to make things better. note: now is the time to send feedback... later on when the groundwork has been laid, changing things will be a lot more difficult.

day 1 - things done and planned

- started off with the parser module, that will be coded
  in ansi c.
- translated the bit of code written in c++ back to c.
- writing code for *external* obituary file, that will
  let you control and recognize most of the games on the
  fly. this will make things a lot easier for admins:
  found a new weapon in your logs? simply add it in the
  obituary file!
- will develop basic log line recognition along with
  general message classes like frags, suicides, events
  (captures, hostage saves, ...).

sunday, november 12th 20:10 2000

æstats⁺⁺ - site launch

welcome to the next generation in log parsing and statistical calculations. i have prepared this site to inform you folks out there, about what the æstats⁺⁺ will be able to do. note that most pages are sorta empty, for the simple reason i have not yet coded the modules.

feel free to check out the support menu items. if you have suggestions to make send in your requests. since i am still in the planning phase any ideas you might have had for æstats, could well fit æstats⁺⁺, so send those in as well.

you can already register for the newsletter.

... end of news archive ...

last updated on monday, may 16th 2005 up since: sunday, november 12th 2000 re-design: monday, may 16th 2005