The delang is using merge instead of rebase/squash

[deadalnix via Digitalmars-d]

On Monday, 20 March 2017 at 05:10:04 UTC, Martin Nowak wrote:
> On Wednesday, 15 March 2017 at 13:14:31 UTC, deadalnix wrote:
>> This is making the history very spaghettified. Is that 
>> possible to have the bot rebase/squash commits and then 
>> pushing ?
>
> I don't really agree with the argument. A merge commit is a 
> clear way to integrate changes from a PR/branch.
> Just rebasing a PR on top of master removes a lot of 
> information from git, only leaving references to github.
>
> Can you be more specific, what you mean w/ spaghetti? The fact 
> that reciew fixes are added to PRs.
>
> Also github's commit view misleadingly shows commits from 
> merged PRs/branches, which aren't actually in master.

Because a picture is clearer than a thousand words:

| | | | | | | |
* | | | | | | |   08ae52d8 The Dlang Bot
|\ \ \ \ \ \ \ \      Merge pull request #5231 from 
RazvanN7/Update_generated
| |_|_|_|_|/ / /
|/| | | | | | |
| | | | | | | |
| * | | | | | | c6480976 RazvanN7
|/ / / / / / /      Updated posix.mak makefile to use 
../tools/checkwhitespace.d
| | | | | | |
* | | | | | |   1181fcf7 The Dlang Bot
|\ \ \ \ \ \ \      Merge pull request #5239 from 
sprinkle131313/ignore-vscode-lib
| | | | | | | |
| * | | | | | | f1b8d0d4 sprinkle131313
| | | | | | | |     Add temp/tmp folder to gitignore.
| | | | | | | |
| * | | | | | | b67bf9d1 sprinkle131313
| | |_|/ / / /      Add vscode folder and lib files to gitignore.
| |/| | | | |
| | | | | | |
* | | | | | |   0b41c996 The Dlang Bot
|\ \ \ \ \ \ \      Merge pull request #5242 from 
wilzbach/fix-lref-links
| | | | | | | |
| * | | | | | | 090d5164 Sebastian Wilzbach
|/ / / / / / /      Fix links from $(LREF $(D ...)) -> $(LREF ...)
| | | | | | |
* | | | | | |   f2a019df The Dlang Bot
|\ \ \ \ \ \ \      Merge pull request #5241 from 
MartinNowak/merge_stable
| | | | | | | |
| * | | | | | | a6cb85b8 Sebastian Wilzbach
| | | | | | | |     Add @safe to std.regex unittest
| | | | | | | |
| * | | | | | |   ad70b082 Martin Nowak
| |\ \ \ \ \ \ \      Merge remote-tracking branch 
'upstream/stable' into merge_stable
|/ / / / / / / /
| | | | | | | |
| * | | | | | |   694dd174 Stefan Koch
| |\ \ \ \ \ \ \      Merge pull request #5167 from 
DmitryOlshansky/fix-freeform-regex
| | | | | | | | |
| | * | | | | | | 62cf615d Dmitry Olshansky
| |/ / / / / / /      Fix issue 17212 std.regex doesn't ignore 
whitespace after character classes
| | | | | | | |
* | | | | | | | 5b07bd59 Sebastian Wilzbach
| |_|_|_|/ / /      [BOOKTABLES]: Add BOOKTABLE to 
stdx.checkedint (#5238)
|/| | | | | |
| | | | | | |
* | | | | | |   75059373 Jack Stouffer
|\ \ \ \ \ \ \      Merge pull request #5225 from 
wilzbach/booktable-std-utf
| |_|_|_|_|_|/
|/| | | | | |

What the hell is going on in there ?

In addition there are a bunch of practical issues with this way 
of doing things. First there is no given that any intermediate 
state is sound, or even builds at all. That makes it very hard to 
bissect anything. There are also a lot of errands and correction 
that are made during review that are not that interesting to keep 
in the project history. Knowing that someone did thing the A way 
and then changed it the B way after review is more noise than 
useful infos in the general case, and in the rare case when 
someone actually wants to know, the github PR is still out there 
(on that note, yes GH PR kind fo sucks, but that's another topic).