demostf / parser Goto Github PK

First of all, thank you for this piece of software!

I tried to parse a big pile of POV demos (3575 files) recorded over many years using parse_demo binary but many of them failed. Here are the errors:

1. 1006 failures (28%)

   Error: A read game event doesn't contain the expected values, expected type Byte for kills_rank event, got type Long
   

2. 12 failures (0.3%)

   Error: A read game event doesn't contain the expected values, expected type Byte for kills_rank event, got type Short
   

3. 15 failures (0.4%)

   Error: Malformed utf8 while reading string
   caused by: invalid utf-8 sequence of 1 bytes from index 0
   

   In 5 cases the index was not 0.
4. 5 failures (0.1%)

   Error: A read game event doesn't contain the expected values, expected type Long for weapon_def_index event, got type Short
   

That's it so the total number of different errors is surprisingly low :)

And it's also not unlikely that there's been some bit rot.

The demo file is rather large and I couldn't minify it, but I've attached it anyways.

crash.zip

The crash is

The fix here IMO is to do a checked subtraction there, but also not to trust the value we got from the file for the compressed size. That might be a change in bitbuffer too, which should have tests that trying to read huge amounts of data doesn't crash, and instead returns an error, which it doesn't seem to be doing.

It would be nice to fetch out position, velocity or even projectiles (rockets/stickies/pipes).

Working on integrating this with Rocket Jump/Sticky Jump STV demos. Chat logs has helped heaps but if we can get more data that would be awesome

In the README.md, it says to pass the detailed_summary argument to print more information.
However it doesn't work because in main.rs the code looks for the string detailed_summaries instead:

let detailed_summaries = args.contains(&std::string::String::from("detailed_summaries"));

Someone has once asked me if they could confirm in any way if someone possibly had something illegal on their objector, however I wasn't able to help. And in general it'd be interesting to see what people have at any given match on any of their decal applicable items.

Hello, I have been working with this program for quite a bit, modifying it to parse demo files in real time to collect data for machine learning algorithms.

I noticed that when created a DemoParser object with the new_all_with_analyzer function, the team for each user is listed as "Other". However, creating a DemoParser object using the other function new_all properly reports team for each player. I've been trying to fix it but had no luck, fixing this would really help me since I wouldn't have to waste time parsing the demo file twice.

Hi @icewind1991,

First of all, this is an awesome tool and I've used it to ingest chat logs from about 10 years worth of the jump communities STVs! This has allowed us to regain data lost to time (like the old World Records, players, rankings etc) ⭐
(from Tempus Jump Network)

It would be awesome for my archiving and enhancement efforts if we could get the following out of STVs (feel free to say no - I'd also be happy to sponsor this project, but there's no button on GitHub ATM)

• Extract player positions, angle, velocity in a time sequence. (e.g. say I know there is a world record run within some tick range, we can plot this on a website, or visualize the trail, many things to come from this - even a heatmap of all attempts, or attempts over a long history)
  • Extract rocket fired, stickybomb fired/detonated
  • Jump
• Extract chat message colors? (ranks with color etc)
• Get server messages to a single player, through sourcemod/metamod plugins (Is this possible?)
• Player commands, e.g. sourcemod commands, clientside cheats like running thirdperson etc.

I'm not sure if a lot of this is possible, but its a wide wishlist, for discussion.

I look forward to hearing from you soon :)

Here's an example STV demo that fails:

auto-20140701-233440-jump_propel_fixed_v3.zip

I was able to extract chat messages, by just ignoring the failure in the code, but a proper fix would be better.

Here's the commit that allowed me to continue: Hona@852d0aa