-
-
Notifications
You must be signed in to change notification settings - Fork 32
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
WIP: Combine libtinfo
with libncurses
#10
Conversation
Hi! This is the friendly automated conda-forge-linting service. I just wanted to let you know that I linted all conda-recipes in your PR ( |
DYLIB_EXT=dylib | ||
else | ||
DYLIB_EXT=so | ||
fi |
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.
Need to have the right library extension for each platform.
Should add that we welcome other maintainers on board this feedstock (as with the whole ecosystem). As this seems important to bioconda, maybe it would be good to have a few of your people on board. 🚢 Please let us know if you are interested, we are always happy to have other people join us. 😄 |
Rebased on top of this PR ( #11 ) to provide |
This seems like it should work ok, but I would like someone to confirm that it works for them. Also, I have tried very hard to ensure that from a build standpoint everything will still behave correctly whether or not one expects a separate Even though we want this to be fine from a build standpoint, it will be a breaking change from the runtime standpoint. As a result, anything that is already built with Finally, it seems that the That being said, I don't have very strong feelings against this, but am just a little wary. If this is very desirable behavior from the community perspective, I'm willing to discuss this. Hence why I have generated a PR that implements behavior so we can discuss it more. |
libtinfo
with libncurses
libtinfo
with libncurses
I have marked this as WIP so no one accidentally merges this. However, this is really a working, tested change at this point. People are welcome to play with it if they like. We just should not merge it until much more discussion has occurred and we decide we want to go in this direction. Personally, I'm not yet convinced. |
+1 for following the ncurses team lead and not merging these. This might be some crappy maintenance work to add in the extra library in various makefiles, but it will be worth it, IMHO. |
Agreed. Only proposing this for the convenience of downstream. If providing some |
I am fine with keeping the status quo. Patching the Makefile and using |
@@ -20,14 +27,23 @@ do | |||
--enable-symlinks \ | |||
--enable-termcap \ | |||
--enable-pc-files \ | |||
--with-termlib \ |
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.
Dropping this options gets rid of libtinfo
and combines it in with libncurses
.
No reason from me to not do this. 👍 |
Personally, I don't love it. This seems to be opposite all of the Linux distros I have seen and is only really the same as Mac (which has an ancient copy). Mainly, I put this together so we could see it is possible and discuss with downstream if this is something they would want. Thus far it seems like they are happy with the Honestly, I would be in favor of closing. Especially as it looks like they now don't need it. |
Needs #12 and #11 to be merged first.
Fixes #9
Some users expect to find all of
libtinfo
's symbols inlibncurses
. While not all builds oflibncurses
do this and the trend has been to have a separatelibtinfo
, this puts them back together in the same library. However, it does provides symlinks named after thelibtinfo
libraries that point back tolibncurses
. This way programs that expect to link tolibtinfo
still can, but will get those symbols fromlibncurses
. Part of the reason this has been opted for isdefaults
does the same thing. We have included some tests to verify that some of thelibtinfo
symbols are inlibncurses
, which should be sufficient in verifying this works.After this is merged, anything built with
ncurses
support is going to have to be rebuilt as this will break those packages. Currently, that is justerlang
, which needs to be pinned anyways ( conda-forge/erlang-feedstock#5 ). Though it would be good to get this in before our ownpython
is released so that we will have some time to test this new configuration with that build and release it using this new configuration. Thus, this is a good time to make this break if we are going to do it.xref: bioconda/bioconda-recipes#1391
cc @pelson @ocefpaf @msarahan @jjhelmus @scopatz @cel4 @croth1 @johanneskoester