Pagwin/website
1
0
Fork 0
mirror of https://github.com/Pagwin-Fedora/website.git synced 2025-07-18 07:25:43 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

final changes to the new blog article

Browse source
This commit is contained in:
Pagwin 2022-01-31 12:21:43 -05:00
parent 8759383b3f
commit d89025bef1
1 changed files with 6 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

11
content/blog/pogo_so_far.md
View file

@ -26,16 +26,17 @@ I put it this all toegether with copious usage of FFI to connect the Rust, Go an
## Problems
But recently I've decided that all of that is dumb and unnecessary. Initially it made sense to use the somewhat unwieldy FFI setup because the backend was going to take a struct instead of a string and structs cannot be passed via stdin. Once I realized how hard and unnecessary that was due to FFI things I dropped it. With that my architecture was just passing around strings and I realized that my setup is overly complicated for that. So instead of FFI and TCP spaghetti instead I'm going to have ~~subprocess spaghetti~~ each component of the application be separate executables of some kind or another.
But recently I've decided that all of that is dumb and unnecessary. Initially it made sense to use the somewhat unwieldy FFI setup because the backend was going to take a struct instead of a string and structs cannot be passed via stdin<sup>[citation needed]</sup>. Once I realized how hard and unnecessary that was due to FFI things I dropped it. With that my architecture was just passing around strings and I realized that my setup is overly complicated for that. So instead of FFI and TCP spaghetti instead I'm going to have ~~subprocess spaghetti~~ each component of the application be separate executables of some kind or another.
## The New Architecture
This description is going to suck so I recommend clicking off if you get bored. For my new architecture there will be 4 components
This description is going to suck so I recommend clicking off if you get bored. For my new architecture there will be 5 components
1. The init script
2. The frontend
3. the middleware
4. and The backend.
3. the front-to-back middleware
4. the back-to-front middleware
5. and The backend.
The original repository that held all the code will have it all removed as the frontend, middleware and backend are developed in separate repositories which are had as git submodules in the main repository which will be where an init script and a build script will be held.
@ -49,7 +50,7 @@ The frontend will accept user input and print it to stdout in a manner understoo
### The middleware
The middleware will take messages from the frontend and parse them for the backend to understand and vice versa. In order to check functionality of the middleware to ensure it is passed messages from the end(s) it wants the init process will do a brief feature check of the middleware by running it with the `--features` cli option.
The front to back middleware will take messages from the frontend and parse them for the backend to understand and vice versa for back to front.
### The backend