On 02/11/2020 11:37, Wenyin Wei wrote: > Dear Git Community, > > Consider the number of large repositories using Git has increased > these years dramatically, is it possible for Git to make a > discontinuous clone way? The required continuity for a git clone has > been hard for large repos, and developers have strived by the somewhat > ugly way, `git fetch –depth=<increasing n>`. > > 1. https://stackoverflow.com/questions/38618885/error-rpc-failed-curl-transfer-closed-with-outstanding-read-data-remaining > 2. https://stackoverflow.com/questions/60878838/git-rpc-failed-curl-18-transfer-closed-with-outstanding-read-data-remaining > > It would be appreciated for many developers to see that git could > support something like broken-point continuingly-cloning. I am not > sure whether the `git clone git@......` way supports this > functionality, but even so, some mirror hubs in China, to accelerate > cloning, only support the `https` way and then become anxious while > cloning. Anticipate your comments and reply. > > Best Wishes, > Wenyin Wei Do have a look at the Partial Clone methods which IIUC have been developed to align with the problem of large repos (such as Windows). This filters the download to just those that have been selected as 'needed. There was also, separately, some patches to IIRC keep alive connections for places with slow internet connections. The list archive is at https://lore.kernel.org/git/ and is relatively easy to search. [e.g. keepalive] https://lore.kernel.org/git/20160715104347.GL19271@xxxxxxxxxxxxxxxxxxxxx/
