From 2cf9ed277dfc9e2b40e70a9653a7292bbf929249 Mon Sep 17 00:00:00 2001 From: Ilya_Shlyakhter Date: Thu, 18 Mar 2021 17:35:22 +0000 Subject: [PATCH] added a forum question about replacing a locked file without explicitly unlocking/relocking --- doc/forum/replacing_a_locked_file_without_unlocking_first.mdwn | 1 + 1 file changed, 1 insertion(+) create mode 100644 doc/forum/replacing_a_locked_file_without_unlocking_first.mdwn diff --git a/doc/forum/replacing_a_locked_file_without_unlocking_first.mdwn b/doc/forum/replacing_a_locked_file_without_unlocking_first.mdwn new file mode 100644 index 0000000000..377f119abb --- /dev/null +++ b/doc/forum/replacing_a_locked_file_without_unlocking_first.mdwn @@ -0,0 +1 @@ +If I want to replace a locked file with new content, rather than edit the existing content, is it ok to skip the [[`git-annex-unlock`|git-annex-unlock]]/[[`git-annex-lock`|git-annex-lock]] steps? If I just remove the symlink to the annex, create a new regular file in its place, then [[`git-annex-add`|git-annex-add]] the new file and commit, things seem to work fine. Does this create some inconsistency that can come back to bite me later, or is this supported usage?