-
Notifications
You must be signed in to change notification settings - Fork 96
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Use libremesh.mk #729
Use libremesh.mk #729
Conversation
aparcar
commented
Jun 29, 2020
Codecov Report
@@ Coverage Diff @@
## master #729 +/- ##
==========================================
- Coverage 73.90% 73.62% -0.29%
==========================================
Files 34 35 +1
Lines 2817 2972 +155
==========================================
+ Hits 2082 2188 +106
- Misses 735 784 +49
Continue to review full report at Codecov.
|
9c6bdb7
to
ba7b5fd
Compare
It sounds very good to me. I hate writing Makefiles and to have a unified template seems very convenient. Some of the packages could need special stuff (e.g. @aparcar mentioned that tinc package Makefile cannot be simplified this way).
|
This can be seen as a drop in replacement. All other packages with regular Makefiles keep working. Special cases like tinc, where actual C compiling is required, keep their individual Makefiles. |
It is simpler/cleaner. I would prefer something that use composition instead of inheritance, something like:
|
I don't see how this makes anything more difficult. Whoever understands the regular Makefile for packages likely also saw a LuCI Makefile, which is pretty much the same as the LiMe Makefile. Also, as it is still possible to use regular Makefile, existing developers can show the new developer how to convert the Makefile. I did not count but I think the Regarding upstream, I never saw any ambitions to get this done. Even if so, we can also migrate the |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The underlining idea is good, but it seems there are also a bunch of unrelated changes mixed in into this PR, that makes review error prone, please split this PR in smaller pieces taking in account also other comments
Some duplicate files accidentally appeared in the pull request, @aparcar could you delete them? packages/lime-proto-bmx6/src/etc/uci-defaults/85-add-bmx6-addresses-to-hosts |
@aparcar have you seen my other comment on the accidental creation of duplicated files #729 (comment) ? |
e6f111d
to
1eaecba
Compare
Instead of having a complex Makefile per package this commit introduces a Makefile template called `libremesh.mk` which is included in all compatible LibreMesh packages. This follow the approach of LuCI where only very basic information is required per package and the rest is evaluated automatically. The following variables can be set: * LIME_TITLE * LIME_MAINTAINER * LIME_DESCRIPTION * LIME_DEPENDS Everything else is automatically determined.
Signed-off-by: Paul Spooren <mail@aparcar.org>
1eaecba
to
97a74ff
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
An additional comment: you merged the files in the lime-smart-wifi packages into a srv/ directory instead than into a files/ directory.
7bad4c0
to
6962300
Compare
While we wait for @aparcar to change the location of the files inside lime-smart-wifi, can some other dev say if you're ok with this new style of Makefile for LibreMesh specific packages (up to now we had a positive opinion from @G10h4ck and from @dangowrt and a neutral opinion from @spiccinini)? @nicopace @gmarcos87 @germanferrero |
Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Paul Spooren <mail@aparcar.org>
6962300
to
9ad8ea5
Compare
G10h4ck concerns have been addressed by aparcar