Closed yllekz closed 4 years ago
1) Yup. The reason is pretty simple: avoiding GBs of junk inside the history (effectively forcing regular users to pull and store unnecessary data). For example, each day there are more than 250k IP addresses being pushed to the 1.txt
list. As a result, history diffs become huge pretty fast.
2) This was not a decision made in haste. At the beginning, history has been preserved, which resulted with almost unusable git
operations over the repository
3) Maybe something like https://github.com/desktop/desktop/issues/1484#issuecomment-484938631 could be of a help to you (i.e. switch --allow-unrelated-histories
)
As a lurking watcher of this repo, might I suggest using Git LFS to manage the pain of large files in the repo?
@jonathanhoskin switched all *.txt
to Git LFS :)
Nice one 👍
@jonathanhoskin will need to switch back :(
It seems that ipsum
is being pulled/downloaded by many people, thus the problem with bandwidth.
Gah! Sorry to hear that. LFS does seem like the way forward while you are choosing to host files on Github.
While I'm not in a position to provide ongoing financial support, I am in a position to provide this project with free LFS hosting, if you want it.
Let me know if you are interested.
@jonathanhoskin let it be like this for now. People are already used to old links for direct access (Note: Git LFS required changes to raw access from raw
to media
files).
p.s. Anyhow, if this will still be bothering you in a month or so, please come back and we'll could figure something out
Hi there, I'm using GitHub Desktop and trying to pull the changes when this list updates, however I get a "Unable to merge unrelated histories in this repository" error every time, resulting in the need for me to re-clone the repo any time there is an update.
Is this repo being fully republished with every update making it incompatible with pulls?