49 lines
No EOL
1.8 KiB
Markdown
49 lines
No EOL
1.8 KiB
Markdown
# Data for nerds
|
|
|
|
!!! warning
|
|
Before contributing, read the [license](license.md).
|
|
|
|
Here you have something to get, if you want to participate in the coding and somehow you got included - congrats! You are in.
|
|
Here are the basic knowledge stuff.
|
|
|
|
## Where is the data coming from?
|
|
|
|
In the `connector.py` file, you'll find that we use Redis as our DB of choice.
|
|
|
|
The DB is used for storing everything in stores, as its credentials are currently stored inside the machine itself - with a certain hope to bring it up as a k8s secret, when the chance is given.
|
|
|
|
## API Tokens
|
|
The API tokens are stored in the API hash, used to communicate with Twitter API.
|
|
|
|
!!! warning
|
|
As stated before, the API will have a massive update in February 9th.
|
|
|
|
## The Data Structure
|
|
|
|
The current data structure is going as follows:
|
|
```json
|
|
{
|
|
"id": "gag_name",
|
|
"keywords": ["1", "2"],
|
|
"reply": ["https://www.twitter.com/FromBotYam/status/{{status_id}}/video/1"]
|
|
}
|
|
```
|
|
|
|
Special gags (as being described in the "Specia Gags" section in `BotYamPoster.py`) have different kind of treatment, although the data structure is pretty much the same.<br>
|
|
Unlike the special gags, all the gags are stored in the reply_bank json key under `gags`.
|
|
|
|
## Victims
|
|
|
|
The victims list is built as follows:
|
|
```json
|
|
{
|
|
"stream_filter": "from:Victim1 OR from:Victim2 OR from:Victim3",
|
|
"author_id": ["id1", "id2", "id3"],
|
|
"reply": ["added text 1", "added text 2", "added text 3"]
|
|
}
|
|
```
|
|
`stream_filter` is the actual filter used for the Twitter API Stream, to follow the victims as they post.<br>
|
|
`author_id` is the list of Twitter IDs, representing the victims' users.<br>
|
|
`reply` is the list of the possible added text.
|
|
|
|
The list of the users affected, and the possible replies, are all described in the [User Guide](usage.md). |