Exporting Aeon Timeline 3 to Obsidian

Templates and Showcase
#61

Do you mean this:

Remember that aeon3obsidian is an organized open source project that has been around for quite a long time. When I decided to stick with AT2, I summarized some open points in issues in the GitHub repository.

The first version was a bit of a hack, by repackaging the JSON part of the .aeon file into Python dictionaries, which were a bit easier to handle, and from which the Markdown pages were then knitted together.

In the meantime, I am in the process of structuring the code so that it is easier for contributors to maintain it. Reading data from .aeon files and writing to .md files can now be optimized independently of each other. However, if you want to continue working on it, you should understand and use this structure. Simply letting a LLM loose on it and seeing what happens would make my efforts seem in vain.

Back to your suggestion: This leads us to another issue, namely the user-defined calendar. Would the Obsidian tool you mentioned even support such a thing?
At the moment, aeon3obsidian also only supports the Gregorian calendar, more specifically, the “AD” era. But that can easily be changed, as I have also moved the date/time/duration processing into its own class, which can be developed separately from the rest.
See:

#62

Sounds like you’ve made some great progress. I know when the LLM got it functional it wasn’t linking things together, so I wasn’t sure what your original output looked like. A dynamic timeline note, like a MOC I suppose, sounds pretty good. Though if we have AT3 we’ll be using AT3 so it’s important to keep in mind precisely what the benefit is of porting the data over to Obsidian. Personally, I think the biggest benefit to me would be the ability to do my universal searches within Obsidian and have those come back with timeline event data. The search/filter feature in Obsidian is incredibly useful for visualizing specific aspects of a complex saga like mine.

Regarding Gregorian A.D. only, the LLM did figure out how to get the Eras working, so like you said, that should be pretty easy. My timeline has six separate eras and about 25,000 years. So that was the very first thing to break once the LLM got the code to simply run. I’m sure a properly implemented solution would be way better.

If I were to take a crack at maintaining or modifying the code, I’d first need to fully understand it. I’m sure you’re using techniques I’ve never encountered before, so I’d have to get familiar with those. Most of my coding experience has to do with inventory management, financial transactions, and general database operations. My experience with Python is mostly from 2002-2003 where I wrote an entity engine before the Entity Framework was even a twinkle in Microsoft’s eye. I still have the code. :smiley: But I haven’t done any real Python work since then. More c# and SQL Server mainly.

#63

It’s not so much about the Python language as it is about classes and methods. Anyway, you can always take a look at the code in the GitHub repository (the src folder).

My script is almost finished, only the “narrative” part is missing. I will be releasing version 2.0 soon.

For now, I’ve packed the Obsidian files (click on the “Download raw file” symbol) as they are generated at the current stage of development and put them online.
If you want, you can play around with them. The example project of “Murder on the Orient Express” is so complex that it is really worth exploring the possibilities of Obsidian.

Enjoy!

1 Like
#64

I just released version 2.0.1. Most of the code is rewritten from scratch. This took me about 50 hours of work. That’s enough for now.
However, the result is pleasing and the outstanding issues from the first version have been resolved.

Screenshot
Screenshot1164×804 69.4 KB

Here is the project home page: