Add-ons Update Should Fetch Compressed JSON Archive

[Random Void User]

Textual formats like JSON compress well. Files going to /tmp/cudatext/addon_man/*.json for Add-ons Update should traverse the Internet archivally compressed. My test below dropped data size from 136K to 15K on Linux.

Code: Select all

7z a -t7z /tmp/footest.7z /tmp/cudatext/addon_man
ls -hal /tmp/footest.7z

Add-ons Update could fetch an archive and decompress to JSON on arrival. Consolidating network requests into one fetch helps by itself. When shared Internet connections slow, small improvements help.

Thanks!

[main Alexey]

I think that changing of .json URLs to URLs of compressed files will give only small gain, only for slow web connections. If you interested, pls give a patch, but with minimal changes.

[Random Void User]

An archive holding all JSON files. Right now, DNS lookup and HTTP request happens N times for N JSON files. Using one archive means 1 lookup and fetch, much quicker.

[main Alexey]

I am not entusiastic with this change, sorry. Current mood is very low. Because we have a war, in Ru.

[main Alexey]

For example.
Today 22 may 2026, Ua runs rockets, powerful rockets from USA (USA's gift). Starobelsk school (EDITED: not school but dormitory) for young students was destroyed with young ppl inside.

I am sure 10-20 young white students were killed.

So called 'russia officials' told none were killed, 39 students wounded.