purpose of ELLIPSIS_REPO
Opened this issue · 1 comments
Dear all,
I noticed some (minor) inconsistencies in how $ELLIPSIS_REPO
is defined:
Line 18 in 63846f1
and
Line 21 in 63846f1
From what I see, $ELLIPSIS_REPO
is only used when installing via curl ellipsis.sh | bash
, and in the test suite.
- Not sure how many people actually use their own fork of ellipsis itself, but wouldn't it make sense to add
$ELLIPSIS_USER
to the line ininstall.bash
? And, the other way round, add$ELLIPSIS_PROTO
to the definition insrc/init.bash
? - Why is
$ELLIPSIS_REPO
defined insrc/init.bash
? From what I see, when installing ellipsis, the order of execution is- (
scripts/shim.sh
) scripts/install.bash
- define
$url
based on$ELLIPSIS_REPO
(see above) - clone
$url
- source
src/init.bash
- define
$ELLIPSIS_REPO
. At that point it is never used any more.
- (
Or did I miss something?
Looking forward to hearing from you!
Awesome that you took the time to dive into the code!
I think you are correct, and $ELLIPSIS_REPO
does not seem to be used/needed after the init.bash
file is loaded. Probably a remnant from the past. (So I guess it's safe to just remove it there)
Could be that someone actually used ellipsis/ellipsis
to then bootstrap username/ellipsis
but I think it's safe to ignore this edge case.
In the installer the username is not replaced as that would result in the installer defaulting to a custom repo url, while 99% of the users actually use ellipsis/ellipsis
. So for me it makes most sense not to have the username included in the url there.