I recently took over a project. It turned out that when someone else managed the server, the code was uploaded to the server through ftp. There is no problem in uploading code for the first time in this way, but if you frequently upload updated code to the server, it is impossible to search for coverage one by one. In the past, I uploaded the code to the server through svn or git, and then obtained the svn or git code in the web directory of the server, and then used the svn or git command to operate the update later.
Now this project is a project temporarily taken over for secondary development, and I don't want to install a Git server on other people's servers, so I still use its ftp upload method. I thought of the difference file packaging function between the two versions of Git, packaged incremental zip files, and then uploaded them to the server using ftp.
git diff 608e120 4abe32e --name-only | xargs zip update.zip
git diff f506693 622b6a6 --name-only | xargs tar -czvf update.tar.gz