www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - From the cycle "Topic of the day" - .gitignore: how big is too big?

reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
I've left a comment recently at 
https://github.com/D-Programming-Language/phobos/pull/3087.

So what's the deal with that? Whenever a new tool leaves some trash, do 
we chalk a circle on the pavement around it?


Andrei
Mar 21 2015
next sibling parent "Jakob Ovrum" <jakobovrum gmail.com> writes:
On Sunday, 22 March 2015 at 01:15:07 UTC, Andrei Alexandrescu 
wrote:

 I've left a comment recently at 
 https://github.com/D-Programming-Language/phobos/pull/3087.

 So what's the deal with that? Whenever a new tool leaves some 
 trash, do we chalk a circle on the pavement around it?


 Andrei


Aye, our tools should be configured to put temporary files and 
output files in uniform directories that can be .gitignore'd 
wholesale. The problem is not in .gitignore but our makefiles.
Mar 21 2015
prev sibling next sibling parent reply "Daniel Murphy" <yebbliesnospam gmail.com> writes:
"Andrei Alexandrescu"  wrote in message 
news:mel52r$252b$1 digitalmars.com...


 I've left a comment recently at 
 https://github.com/D-Programming-Language/phobos/pull/3087.

 So what's the deal with that? Whenever a new tool leaves some trash, do we 
 chalk a circle on the pavement around it?


We add it to the .gitignore, then forget about it forever.  Having a long 
gitignore doesn't cost us anything.
Mar 21 2015
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 3/21/15 8:54 PM, Daniel Murphy wrote:

 "Andrei Alexandrescu"  wrote in message
 news:mel52r$252b$1 digitalmars.com...


 I've left a comment recently at
 https://github.com/D-Programming-Language/phobos/pull/3087.

 So what's the deal with that? Whenever a new tool leaves some trash,
 do we chalk a circle on the pavement around it?


 We add it to the .gitignore, then forget about it forever.  Having a
 long gitignore doesn't cost us anything.


... but complicated/wrong clean rules in makefiles and clutter in ls. We 
should clean our shit. -- Andrei
Mar 21 2015
next sibling parent reply "Vladimir Panteleev" <vladimir thecybershadow.net> writes:
On Sunday, 22 March 2015 at 05:56:56 UTC, Andrei Alexandrescu 
wrote:

 On 3/21/15 8:54 PM, Daniel Murphy wrote:

 We add it to the .gitignore, then forget about it forever.  
 Having a
 long gitignore doesn't cost us anything.


 ... but complicated/wrong clean rules in makefiles and clutter 
 in ls. We should clean our shit. -- Andrei


I thought moving things around was also one of your pet peeves :)

I know for a fact that changing where win32.mak creates 
phobos.lib is going to break at least Digger and my own D build 
scripts which predate Digger. It will probably break others' 
workflow too. Adding lines to .gitignore seems more reasonable 
from that perspective.

posix.mak is already sensible in that it puts everything under 
generated/.
Mar 22 2015
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 3/22/15 3:17 AM, Vladimir Panteleev wrote:

 On Sunday, 22 March 2015 at 05:56:56 UTC, Andrei Alexandrescu wrote:

 On 3/21/15 8:54 PM, Daniel Murphy wrote:

 We add it to the .gitignore, then forget about it forever. Having a
 long gitignore doesn't cost us anything.


 ... but complicated/wrong clean rules in makefiles and clutter in ls.
 We should clean our shit. -- Andrei


 I thought moving things around was also one of your pet peeves :)


Yah but I'm seeing pull requests "yeah there's some more junk out there, 
let's just add it to .gitignore". It seems to me like the wrong way to 
address the matter at hand.


 I know for a fact that changing where win32.mak creates phobos.lib is
 going to break at least Digger and my own D build scripts which predate
 Digger. It will probably break others' workflow too. Adding lines to
 .gitignore seems more reasonable from that perspective.


On Windows I'd speculate junk is even more of a problem - all of it 
appears right there in Explorer and other GUI tools. Doesn't that bother 
people?


 posix.mak is already sensible in that it puts everything under generated/.


Yah guess who did that :o).

Ionno at a higher level this is what bothers me: at work this wouldn't 
be debated - as soon as someone noticed, there'd be agreement and the 
matter would be looked at. I wish we had a better culture on the D team 
as well. It looks that as soon as anything remotely debatable comes 
around, belligerent mode is on.


Andrei
Mar 22 2015
parent reply "Vladimir Panteleev" <vladimir thecybershadow.net> writes:
On Sunday, 22 March 2015 at 16:08:38 UTC, Andrei Alexandrescu 
wrote:

 On 3/22/15 3:17 AM, Vladimir Panteleev wrote:

 I thought moving things around was also one of your pet peeves 
 :)


 Yah but I'm seeing pull requests "yeah there's some more junk 
 out there, let's just add it to .gitignore". It seems to me 
 like the wrong way to address the matter at hand.


It is much easier to add a few lines to .gitignore than to fix 
the Makefile, then fix Digger, then fix my build scripts, then 
explain why the change was necessary to anyone who then complains 
that this broke their build. It is not a matter of which fix is 
better, but a matter of which is within my threshold of the 
effort I am ready to exert at the moment. If someone with a 
higher threshold and who is also bothered by the .gitignore mess 
comes along, all the better for everyone if they decide to fix 
the problem in a better way. Or do you adhere to the philosophy 
of "good enough is the enemy of better, so let's not fix anything 
unless we can fix it perfectly"?

Furthermore, I'm not sure if a "proper" fix along the lines of 
posix.mak is even possible, given the limitations of DM make.


 On Windows I'd speculate junk is even more of a problem - all 
 of it appears right there in Explorer and other GUI tools. 
 Doesn't that bother people?


My subjective answer is "yes, it is a minor nuisance, but in 
general, that's what .gitignore is for".


 posix.mak is already sensible in that it puts everything under 
 generated/.


 Yah guess who did that :o).

 Ionno at a higher level this is what bothers me: at work this 
 wouldn't be debated - as soon as someone noticed, there'd be 
 agreement and the matter would be looked at. I wish we had a 
 better culture on the D team as well.


One is a controlled environment. Another is an open-source 
project. I had the impression that "move fast and break things" 
does not apply to D any more.


 It looks that as soon as anything remotely debatable comes 
 around, belligerent mode is on.


I was not the one who created this thread. I was the one who 
created a pull request which improved the situation.
Mar 22 2015
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 3/22/15 12:27 PM, Vladimir Panteleev wrote:

 On Sunday, 22 March 2015 at 16:08:38 UTC, Andrei Alexandrescu wrote:

 On 3/22/15 3:17 AM, Vladimir Panteleev wrote:

 I thought moving things around was also one of your pet peeves :)


 Yah but I'm seeing pull requests "yeah there's some more junk out
 there, let's just add it to .gitignore". It seems to me like the wrong
 way to address the matter at hand.


 It is much easier to add a few lines to .gitignore than to fix the
 Makefile, then fix Digger, then fix my build scripts, then explain why
 the change was necessary to anyone who then complains that this broke
 their build. It is not a matter of which fix is better, but a matter of
 which is within my threshold of the effort I am ready to exert at the
 moment. If someone with a higher threshold and who is also bothered by
 the .gitignore mess comes along, all the better for everyone if they
 decide to fix the problem in a better way.


Fair enough. I'll drop this. -- Andrei
Mar 22 2015
parent reply "Mengu" <mengukagan gmail.com> writes:
On Sunday, 22 March 2015 at 19:32:51 UTC, Andrei Alexandrescu 
wrote:

 On 3/22/15 12:27 PM, Vladimir Panteleev wrote:

 On Sunday, 22 March 2015 at 16:08:38 UTC, Andrei Alexandrescu 
 wrote:

 On 3/22/15 3:17 AM, Vladimir Panteleev wrote:

 I thought moving things around was also one of your pet 
 peeves :)


 Yah but I'm seeing pull requests "yeah there's some more junk 
 out
 there, let's just add it to .gitignore". It seems to me like 
 the wrong
 way to address the matter at hand.


 It is much easier to add a few lines to .gitignore than to fix 
 the
 Makefile, then fix Digger, then fix my build scripts, then 
 explain why
 the change was necessary to anyone who then complains that 
 this broke
 their build. It is not a matter of which fix is better, but a 
 matter of
 which is within my threshold of the effort I am ready to exert 
 at the
 moment. If someone with a higher threshold and who is also 
 bothered by
 the .gitignore mess comes along, all the better for everyone 
 if they
 decide to fix the problem in a better way.


 Fair enough. I'll drop this. -- Andrei


while we're at it, let's add D to this list: 
https://github.com/github/gitignore
Mar 22 2015
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
On 3/22/15 1:37 PM, Mengu wrote:

 while we're at it, let's add D to this list:
 https://github.com/github/gitignore


That's be cool, any takers? -- Andrei
Mar 22 2015
next sibling parent reply Mathias Lang via Digitalmars-d <digitalmars-d puremagic.com> writes:
I was wondering how this could be done this afternoon.
Thanks Mengu.

https://github.com/github/gitignore/pull/1444

2015-03-22 22:08 GMT+01:00 Andrei Alexandrescu via Digitalmars-d <
digitalmars-d puremagic.com>:


 On 3/22/15 1:37 PM, Mengu wrote:


 while we're at it, let's add D to this list:
 https://github.com/github/gitignore


 That's be cool, any takers? -- Andrei
Mar 22 2015
parent "Mengu" <mengukagan gmail.com> writes:
On Sunday, 22 March 2015 at 22:04:53 UTC, Mathias Lang wrote:

 I was wondering how this could be done this afternoon.
 Thanks Mengu.

 https://github.com/github/gitignore/pull/1444

 2015-03-22 22:08 GMT+01:00 Andrei Alexandrescu via 
 Digitalmars-d <
 digitalmars-d puremagic.com>:


 On 3/22/15 1:37 PM, Mengu wrote:


 while we're at it, let's add D to this list:
 https://github.com/github/gitignore


 That's be cool, any takers? -- Andrei




you're welcome :)
Mar 22 2015
prev sibling parent "Mengu" <mengukagan gmail.com> writes:
bOn Sunday, 22 March 2015 at 21:08:27 UTC, Andrei Alexandrescu 
wrote:

 On 3/22/15 1:37 PM, Mengu wrote:

 while we're at it, let's add D to this list:
 https://github.com/github/gitignore


 That's be cool, any takers? -- Andrei


btw, i think it'd be good if dub would automatically include this 
.gitignore file to generated projects.
Mar 22 2015
prev sibling parent Jonathan M Davis via Digitalmars-d <digitalmars-d puremagic.com> writes:
On Saturday, March 21, 2015 22:56:58 Andrei Alexandrescu via Digitalmars-d
wrote:

 On 3/21/15 8:54 PM, Daniel Murphy wrote:

 "Andrei Alexandrescu"  wrote in message
 news:mel52r$252b$1 digitalmars.com...


 I've left a comment recently at
 https://github.com/D-Programming-Language/phobos/pull/3087.

 So what's the deal with that? Whenever a new tool leaves some trash,
 do we chalk a circle on the pavement around it?


 We add it to the .gitignore, then forget about it forever.  Having a
 long gitignore doesn't cost us anything.


 ... but complicated/wrong clean rules in makefiles and clutter in ls. We
 should clean our shit. -- Andrei


Agreed.

- Jonathan M Davis
Mar 22 2015
prev sibling parent reply Jacob Carlborg <doob me.com> writes:
On 2015-03-22 02:15, Andrei Alexandrescu wrote:

 I've left a comment recently at
 https://github.com/D-Programming-Language/phobos/pull/3087.

 So what's the deal with that? Whenever a new tool leaves some trash, do
 we chalk a circle on the pavement around it?


Not sure what you're complaining about. Shouldn't generated files be 
ignored? Or is the problem that they're not put in one single directory?

-- 
/Jacob Carlborg
Mar 23 2015
parent "H. S. Teoh via Digitalmars-d" <digitalmars-d puremagic.com> writes:
On Mon, Mar 23, 2015 at 09:14:15AM +0100, Jacob Carlborg via Digitalmars-d
wrote:

 On 2015-03-22 02:15, Andrei Alexandrescu wrote:

I've left a comment recently at
https://github.com/D-Programming-Language/phobos/pull/3087.

So what's the deal with that? Whenever a new tool leaves some trash,
do we chalk a circle on the pavement around it?


 
 Not sure what you're complaining about. Shouldn't generated files be
 ignored? Or is the problem that they're not put in one single
 directory?


[...]

Word on the street is that the recommended way is to put all generated
files in a dedicated directory rather than mix them in with source
files. Doing so has other side benefits, like being able to parallel
build for multiple targets without stomping over each other (assuming
the generated files get placed in their respective architecture's
subdirectory), avoiding stray orphan generated files causing heisenbugs
(stale object file getting linked into the program overriding symbols
from the real object file -- I've actually run into this at work, and
it's very frustrating).

But for Phobos, meh... just stick it in .gitignore and call it a day. No
need to make such a big deal out of it. Tempest in a teacup.


T

-- 
Never ascribe to malice that which is adequately explained by incompetence. --
Napoleon Bonaparte
Mar 23 2015