WPI format hacking

[roelj]

In an effort to figure out more about the WPI format I'd like to share some samples with you of what I extracted from the format. If it rings any bells, please let me know. If you're interested in hacking on the format yourself, I could push my changes into a separate branch so you can play with it as well.

First the data is expressed in integers, after the ":" the same data is displayed as characters. My sample can be downloaded here: 194_6_output.txt.

[roelj]

Looking at where the data appears it doesn't seem to be important for displaying information.

[xuv]

Sorry. I know the issue is closed, but I'm intrigued by what you think the data could be, if not part of the display info. Is it part of some error mechanism? (just wild guessing)

[roelj]

Maybe I should be more elaborate about what I've tried out so far.. At first I thought it could be an error mechanism indeed: A mechanism to know for sure what comes first and what comes second. The data is increasing with 1 every time it shows up.

Looking at when the data shows up, it's too rare to be a good error mechanism. I could make a separate branch and make a full format-debugging version of the WPI parser, then you'll have as much information as I have gathered so far.

I'll reopen the issue and provide that format-debugging branch.

[xuv]

Thanks for the explanation. Not that I can really help you a lot I think. It was more out of curiosity that I ask all these questions. But I hope someone else can help you out figure out what this data is used for.

[roelj]

See the format-debugging branch.

[roelj]

The format-debugging branch is meant for format hacking. Since there's no further discussion I'm going to close the issue now.