psi-jack/yadm
1
0
Fork 0
mirror of synced 2025-01-05 12:42:58 -05:00
Code Issues Releases Wiki Activity

Add new detail to HOOKS documentation

Browse source
This commit is contained in:
Tim Byrne 2017-07-06 21:53:23 -05:00
parent 557ce93fab
commit bf61ad662d
No known key found for this signature in database
GPG key ID: 14DB4FC2465A4B12
1 changed files with 56 additions and 15 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

71
yadm.1
View file

@ -400,21 +400,6 @@ Override the HOSTNAME for the purpose of symlinking alternate files.
.TP
.B local.user
Override the USER for the purpose of symlinking alternate files.
.SH HOOKS
.B yadm
has the capability to execute scripts before or after any operation that
.B yadm
can perform. To utilize this functionality, create a directory to store the hook
scripts at
.BR $HOME/.yadm/hooks.
Then, create scripts inside this directory for whatever operation you want to
hook into. For instance, if you'd like a script to run after
.B yadm pull,
your hook script should be executable and located at
.BR $HOME/.yadm/hooks/post_pull.sh.
Any of the
.B yadm
subcommands can utilize this functionality.
.SH ALTERNATES
When managing a set of files across different systems, it can be useful to have
an automated way of choosing an alternate version of a file for a different
@ -644,6 +629,62 @@ Even if disabled, permissions can be manually updated by running
The SSH directory processing can be disabled using the
.I yadm.ssh-perms
configuration.
.SH HOOKS
For every command
.B yadm
supports, a program can be provided to run before or after that command. These
are referred to as "hooks".
.B yadm
looks for
hooks in the directory
.IR $HOME/.yadm/hooks .
Each hook is named using a prefix of
.I pre_
or
.IR post_ ,
followed by the command which should trigger the hook. For
example, to create a hook which is run after every
.I yadm pull
command, create a hook named
.IR post_pull.
Hooks must have the executable file permission set.
If a
.I pre_
hook is defined, and the hook terminates with a non-zero exit status,
.B yadm
will refuse to run the
.B yadm
command. For example, if a
.I pre_commit
hook is defined, but that command ends with a non-zero exit status, the
.I yadm commit
will never be run. This allows one to "short-circuit" any operation using a
.I pre_
hook.
Hooks have the following environment variables available to them at runtime:
.TP
.B YADM_HOOK_COMMAND
The command which triggered the hook
.TP
.B YADM_HOOK_EXIT
The exit status of the
.B yadm
command
.TP
.B YADM_HOOK_FULL_COMMAND
The
.B yadm
command with all command line arguments
.TP
.B YADM_HOOK_REPO
The path to the
.B yadm
repository
.TP
.B YADM_HOOK_WORK
The path to the work-tree
.SH FILES
The following are the default paths
.B yadm