www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - core.sys.posix.sys.ioctl

reply sanjayss <dummy dummy.dummy> writes:
Is there any reason that this module is not complete for 
platforms other than Linux -- the ioctl() system call is common 
across all Unix-like OSes, so it doesn't make sense that this is 
only partially supported. (I am using the latest DMD).
Jan 15
parent reply Adam D. Ruppe <destructionator gmail.com> writes:
On Friday, 15 January 2016 at 18:32:22 UTC, sanjayss wrote:
 Is there any reason that this module is not complete for 
 platforms other than Linux
Nobody has written it up, except the parts they use.
Jan 15
parent reply sanjayss <dummy dummy.dummy> writes:
On Friday, 15 January 2016 at 18:34:14 UTC, Adam D. Ruppe wrote:
 On Friday, 15 January 2016 at 18:32:22 UTC, sanjayss wrote:
 Is there any reason that this module is not complete for 
 platforms other than Linux
Nobody has written it up, except the parts they use.
Is the contribution process straightforward. Since I am mucking around in this area on OSX, I could attempt creating a more complete ioctl module for OSX...unless ofcourse someone else has already tried it and there are too many gotchas or there already have been too many discussions on the right way to do this and there is no consensus (I don't want to get into too much of these types of discussions -- not enough time).
Jan 15
parent reply Adam D. Ruppe <destructionator gmail.com> writes:
On Friday, 15 January 2016 at 21:21:26 UTC, sanjayss wrote:
 Is the contribution process straightforward.
For this, yes. Should be able to just fork druntime and edit the ioctl.d that exists to flesh it out to be more complete. Make sure it matches the original C names, values, etc., and there should be no real friction in getting it in (just remember to use spaces rather than tabs lol)
Jan 15
parent reply sanjayss <dummy dummy.dummy> writes:
On Friday, 15 January 2016 at 21:49:38 UTC, Adam D. Ruppe wrote:
 On Friday, 15 January 2016 at 21:21:26 UTC, sanjayss wrote:
 Is the contribution process straightforward.
For this, yes. Should be able to just fork druntime and edit the ioctl.d that exists to flesh it out to be more complete. Make sure it matches the original C names, values, etc., and there should be no real friction in getting it in (just remember to use spaces rather than tabs lol)
So I got round to doing this -- https://github.com/D-Programming-Language/druntime/pull/1485 -- it's not clear what happens now -- is there a set of people that reviews pull requests and comment on it or is it open to all?
Jan 27
next sibling parent tsbockman <thomas.bockman gmail.com> writes:
On Thursday, 28 January 2016 at 06:15:54 UTC, sanjayss wrote:
 So I got round to doing this -- 
 https://github.com/D-Programming-Language/druntime/pull/1485 -- 
 it's not clear what happens now -- is there a set of people 
 that reviews pull requests and comment on it or is it open to 
 all?
Since your pull request (PR) is new, it is at the top of druntime's list. This pretty much guarantees that several contributors (like myself) will look at it, at least briefly. Anyone who feels both sufficiently qualified (not me; I don't own a Mac) and interested to do a real review will probably leave some sort of comment, although initially this may be about some tedious triviality like your source code formatting or choice of name for a new symbol. Work through whatever issues they bring up, and sooner or later you will probably get their approval for the PR. This doesn't yet mean that your PR will definitely be accepted into the project, but it should attract the attention of one of the senior members of the project, people who actually have the authority to merge stuff: Andrei Alexandrescu, Martin Nowak, etc. One of them will either request additional changes to the PR, or tell you why it probably won't be accepted. Once you have satisfied their requirements, your PR should be merged pretty quickly. For something uncontroversial like your PR, the whole process will probably take a few days or a few weeks, depending on how busy the potential reviewers are with other things.
Jan 27
prev sibling parent tsbockman <thomas.bockman gmail.com> writes:
On Thursday, 28 January 2016 at 06:15:54 UTC, sanjayss wrote:
 So I got round to doing this -- 
 https://github.com/D-Programming-Language/druntime/pull/1485 -- 
 it's not clear what happens now -- is there a set of people 
 that reviews pull requests and comment on it or is it open to 
 all?
I should also say that the process I described is typical of non-trivial Phobos and DMD submissions. Fewer people seem to be interested in druntime, though, so it's quite likely that the first person to seriously review your PR actually *will* be someone with the authority to merge it.
Jan 28