www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.announce - Beta D 2.070.0-b1

reply Martin Nowak <code+news.digitalmars dawg.eu> writes:
First beta for the 2.070.0 release.

Still a few things missing from the changelog, there is a new package
std.experimental.ndslice, and native (DWARF based) exception handling on
linux.

http://dlang.org/download.html#dmd_beta
http://dlang.org/changelog/2.070.0.html

Please report any bugs at https://issues.dlang.org

-Martin
Jan 03 2016
next sibling parent Ilya Yaroshenko <ilyayaroshenko gmail.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.

 Still a few things missing from the changelog, there is a new 
 package std.experimental.ndslice, and native (DWARF based) 
 exception handling on linux.

 http://dlang.org/download.html#dmd_beta 
 http://dlang.org/changelog/2.070.0.html

 Please report any bugs at https://issues.dlang.org

 -Martin


Thanks!
Jan 03 2016
prev sibling next sibling parent reply tsbockman <thomas.bockman gmail.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.


Any hope for this?
     
https://github.com/D-Programming-Language/dmd/pull/3407#issuecomment-136974686
It's been bugging a lot of people lately.
Jan 03 2016
next sibling parent Ilya Yaroshenko <ilyayaroshenko gmail.com> writes:
On Sunday, 3 January 2016 at 20:20:19 UTC, tsbockman wrote:

 On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.


 Any hope for this?
     
 https://github.com/D-Programming-Language/dmd/pull/3407#issuecomment-136974686
 It's been bugging a lot of people lately.


+1
Jan 03 2016
prev sibling parent Martin Nowak <code+news.digitalmars dawg.eu> writes:
On 01/03/2016 09:20 PM, tsbockman wrote:

 
 Any hope for this?
    
 https://github.com/D-Programming-Language/dmd/pull/3407#issuecomment-136974686
 
 It's been bugging a lot of people lately.


Well, this still needs a lot of work that nobody was did.
Walter spend almost the whole release cycle on EH, I spend a huge amount
of time on nightlies/install scripts and was away half of the cycle. So
while I still think 313&314 should be one of our main priorities, it
seems I couldn't convince anyone.
Jan 03 2016
prev sibling next sibling parent reply Joakim =?UTF-8?B?QnLDpG5uc3Ryw7Zt?= <notfornow dev.null.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.
 Please report any bugs at https://issues.dlang.org

 -Martin


Regression?
Found when compiling dub-package scriptlike (struct Path).


struct S {
     this(string r = ".") {
     }
}

void main(string[] args) {
     S s;
}

./foo.d(9): Error: variable foo.main.s default construction is 
disabled for type S
Failed: ["dmd", "-v", "-o-", "./foo.d", "-I."]

It works with dmd 2.069.2.
I think this has been discussed and is intentional.
I would suggest clearly mentioning it in the changelog.

// Joakim
Jan 04 2016
parent Martin Nowak <code+news.digitalmars dawg.eu> writes:
On 01/04/2016 04:29 PM, Joakim Brännström wrote:

 Regression?
 Found when compiling dub-package scriptlike (struct Path).


Thanks for reporting. It's not acceptable to break code like that
without a proper deprecation cycle.
https://issues.dlang.org/show_bug.cgi?id=15515
Jan 04 2016
prev sibling next sibling parent reply Martin Nowak <code+news.digitalmars dawg.eu> writes:
On 01/03/2016 08:24 PM, Martin Nowak wrote:

 Please report any bugs at https://issues.dlang.org


Please test the beta.
I immediately found several bugs while testing libraries [¹], one of
which in gfm (affecting a couple of dependent packages) [²] and another
one in ae-graphics.
We've already lost one week w/o fixing those blockers, so again please
help and test the beta.

In case you don't already use Travis-CI or something similar, you should
start doing so. Testing the latest beta in Travis-CI is as simple as
adding dmd-2.070.0-b1 to your .travis.yml [³] (and we'll soon make it
even easier adding support for dmd-beta and dmd-nightly).

[¹]: https://github.com/MartinNowak/project_tester
[²]: https://issues.dlang.org/show_bug.cgi?id=15550
[³]:
https://github.com/p0nce/ae-graphics/blob/12c1f6d4413d7e6978973a9b84152bce3f331e2d/.travis.yml
Jan 10 2016
next sibling parent reply Jacob Carlborg <doob me.com> writes:
On 2016-01-11 00:58, Martin Nowak wrote:


 [¹]: https://github.com/MartinNowak/project_tester


Can one request for adding projects to this?

-- 
/Jacob Carlborg
Jan 10 2016
parent reply Martin Nowak <code+news.digitalmars dawg.eu> writes:
On 01/11/2016 08:37 AM, Jacob Carlborg wrote:

 On 2016-01-11 00:58, Martin Nowak wrote:
 

 [¹]: https://github.com/MartinNowak/project_tester


 
 Can one request for adding projects to this?
 


If the project is well maintained and relevant, just make a PR.
https://github.com/MartinNowak/project_tester/blob/dd2afea122048dbcb90d0d9b637d31134a6a82ce/projects.xml#L94
Jan 11 2016
parent Jacob Carlborg <doob me.com> writes:
On 2016-01-11 13:58, Martin Nowak wrote:


 If the project is well maintained and relevant, just make a PR.
 https://github.com/MartinNowak/project_tester/blob/dd2afea122048dbcb90d0d9b637d31134a6a82ce/projects.xml#L94


I had Tango in mind. But one might argue that it's not relevant and not 
well maintained. But there's a couple of reasons why I think it would be 
good to test it:

* It's pretty conservatively written. No new fancy feature, or features 
that might be considered experimental. This is a good test base for 
finding breakage for code that has been working for years, the core features

* It doesn't get any new development, which means that the only likely 
reason for the code to not compile is because of a breaking change 
(accidental or intentional)

* Basically every single release of DMD during last couple of years has 
caused at least one breakage in Tango

-- 
/Jacob Carlborg
Jan 11 2016
prev sibling next sibling parent reply Jacob Carlborg <doob me.com> writes:
On 2016-01-11 00:58, Martin Nowak wrote:


 Please test the beta.


The introduction of "message" in Throwable is a breaking change [1], but 
I guess it's not worth reporting an issue for.

[1] Error: class tango.text.xml.SaxParser.SAXException use of 
object.Throwable.message() is hidden by SAXException; use 'alias message 
= Throwable.message;' to introduce base class overload set

-- 
/Jacob Carlborg
Jan 10 2016
parent reply Martin Nowak <code dawg.eu> writes:
On Monday, 11 January 2016 at 07:56:18 UTC, Jacob Carlborg wrote:

 On 2016-01-11 00:58, Martin Nowak wrote:


 Please test the beta.


 The introduction of "message" in Throwable is a breaking change 
 [1], but I guess it's not worth reporting an issue for.


Oh it's well worth to mention any update issues, in particular if 
we're breaking code to introduce problematic designs.
https://github.com/D-Programming-Language/druntime/pull/1445#issuecomment-170530377
Jan 11 2016
parent Jacob Carlborg <doob me.com> writes:
On 2016-01-11 13:40, Martin Nowak wrote:


 Oh it's well worth to mention any update issues, in particular if we're
 breaking code to introduce problematic designs.
 https://github.com/D-Programming-Language/druntime/pull/1445#issuecomment-170530377


Created an issue: https://issues.dlang.org/show_bug.cgi?id=15555

-- 
/Jacob Carlborg
Jan 11 2016
prev sibling next sibling parent reply Basile B. <b2.temp gmx.com> writes:
On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:

 On 01/03/2016 08:24 PM, Martin Nowak wrote:

 Please report any bugs at https://issues.dlang.org


 Please test the beta.


Copy that. I've done what I usually do when a RC is announced and 
found nothing except a small problem in unDead which was actually 
even not a regression.

Maybe there's not much feedback because there's not much problem. 
;p
Jan 11 2016
next sibling parent Rory McGuire via Digitalmars-d-announce writes:
On Mon, Jan 11, 2016 at 10:09 AM, Basile B. via Digitalmars-d-announce <
digitalmars-d-announce puremagic.com> wrote:


 On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:


 On 01/03/2016 08:24 PM, Martin Nowak wrote:


 Please report any bugs at https://issues.dlang.org


 Please test the beta.


 Copy that. I've done what I usually do when a RC is announced and found
 nothing except a small problem in unDead which was actually even not a
 regression.

 Maybe there's not much feedback because there's not much problem. ;p


Yeah, I've been testing and haven't found anything broken.
Jan 11 2016
prev sibling next sibling parent Rory McGuire via Digitalmars-d-announce writes:
On Mon, Jan 11, 2016 at 10:15 AM, Rory McGuire <rjmcguire gmail.com> wrote:


 On Mon, Jan 11, 2016 at 10:09 AM, Basile B. via Digitalmars-d-announce <
 digitalmars-d-announce puremagic.com> wrote:


 On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:


 On 01/03/2016 08:24 PM, Martin Nowak wrote:


 Please report any bugs at https://issues.dlang.org


 Please test the beta.


 Copy that. I've done what I usually do when a RC is announced and found
 nothing except a small problem in unDead which was actually even not a
 regression.

 Maybe there's not much feedback because there's not much problem. ;p


 Yeah, I've been testing and haven't found anything broken.


:) had to say it didn't I.

I now have a "out of memory" error from dmd on one of my projects. I have
16GB of ram and 2.069 doesn't do this. I'll try dustmite.
Jan 11 2016
prev sibling parent Rory McGuire via Digitalmars-d-announce writes:
On Mon, Jan 11, 2016 at 11:26 AM, Rory McGuire <rjmcguire gmail.com> wrote:


 On Mon, Jan 11, 2016 at 10:15 AM, Rory McGuire <rjmcguire gmail.com>
 wrote:


 On Mon, Jan 11, 2016 at 10:09 AM, Basile B. via Digitalmars-d-announce <
 digitalmars-d-announce puremagic.com> wrote:


 On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:


 On 01/03/2016 08:24 PM, Martin Nowak wrote:


 Please report any bugs at https://issues.dlang.org


 Please test the beta.


 Copy that. I've done what I usually do when a RC is announced and found
 nothing except a small problem in unDead which was actually even not a
 regression.

 Maybe there's not much feedback because there's not much problem. ;p


 Yeah, I've been testing and haven't found anything broken.


 :) had to say it didn't I.

 I now have a "out of memory" error from dmd on one of my projects. I have
 16GB of ram and 2.069 doesn't do this. I'll try dustmite.


I'm building with dub and this only happened once, I've tried using --force
and I still don't get that error.
Jan 11 2016
prev sibling next sibling parent Jacob Carlborg <doob me.com> writes:
On 2016-01-11 00:58, Martin Nowak wrote:


 Please test the beta.


Looking good so far.

-- 
/Jacob Carlborg
Jan 11 2016
prev sibling parent reply ponce <contact gam3sfrommars.fr> writes:
On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:

 In case you don't already use Travis-CI or something similar, 
 you should start doing so. Testing the latest beta in Travis-CI 
 is as simple as adding dmd-2.070.0-b1 to your .travis.yml [³] 
 (and we'll soon make it even easier adding support for dmd-beta 
 and dmd-nightly).



I fear there are more regressions that people can test for.

Currently we can't "dub test" release builds trough DUB.

If you do:

     dub test mypackage

it builds with flags -g -unittest -w -debug. (like a -b debug 
build + the -unittest flag)

But if you do:

     dub test mypackage -b release

it builds with flags -release -inline -O -w BUT NOT -unittest

Currently we can only "dub test" debug builds without anything 
dangerous.

Since a number of regressions are in the backend, that could be 
useful if DUB would let us test release builds in Travis-CI.
Jan 11 2016
next sibling parent Basile B. <b2.temp gmx.com> writes:
On Monday, 11 January 2016 at 16:24:56 UTC, ponce wrote:

 On Sunday, 10 January 2016 at 23:58:38 UTC, Martin Nowak wrote:

 In case you don't already use Travis-CI or something similar, 
 you should start doing so. Testing the latest beta in 
 Travis-CI is as simple as adding dmd-2.070.0-b1 to your 
 .travis.yml [³] (and we'll soon make it even easier adding 
 support for dmd-beta and dmd-nightly).



 I fear there are more regressions that people can test for.

 Currently we can't "dub test" release builds trough DUB.

 If you do:

     dub test mypackage

 it builds with flags -g -unittest -w -debug. (like a -b debug 
 build + the -unittest flag)

 But if you do:

     dub test mypackage -b release

 it builds with flags -release -inline -O -w BUT NOT -unittest

 Currently we can only "dub test" debug builds without anything 
 dangerous.

 Since a number of regressions are in the backend, that could be 
 useful if DUB would let us test release builds in Travis-CI.


 From my side, when I test a DMD release I just open 1 by 1 the 
things in metad 
(https://github.com/BBasile/metad/tree/master/repos) ,compile 
them, some of the projects have a unittest "config" and I run it 
as well. It's clear that there's a risk of compilation success 
because templates are not instantiated...

Latest two rlz i've found regressions like that but this time 
it's ok.
Regressions for 2.070 must be on very edgy things.
Jan 11 2016
prev sibling parent reply Martin Nowak <code dawg.eu> writes:
On Monday, 11 January 2016 at 16:24:56 UTC, ponce wrote:

 it builds with flags -release -inline -O -w BUT NOT -unittest


Makes sense 
https://github.com/D-Programming-Language/dub/issues/747. The 
compiler does check assertions in unittest blocks even in release 
builds, right?
It's really trivial to add so you might have a try yourself 
https://github.com/D-Programming-Language/dub/blob/3ab683b023bd58dd101e110cf2f6199911eb7477/source/dub/package_.d#L307.
In the meantime it's possible to define a custom build type 
http://code.dlang.org/package-format?lang=json#build-types.
Jan 11 2016
parent reply Jacob Carlborg <doob me.com> writes:
On 2016-01-12 01:54, Martin Nowak wrote:


 Makes sense https://github.com/D-Programming-Language/dub/issues/747.
 The compiler does check assertions in unittest blocks even in release
 builds, right?


Yes, just verified. Actually, it doesn't matter where the assertion is 
placed, as long as -unittest is passed.

-- 
/Jacob Carlborg
Jan 12 2016
parent Guillaume Piolat <name.lastname gmail.com> writes:
On Tuesday, 12 January 2016 at 08:12:58 UTC, Jacob Carlborg wrote:

 Yes, just verified. Actually, it doesn't matter where the 
 assertion is placed, as long as -unittest is passed.


Another surprise.
Jan 12 2016
prev sibling next sibling parent John Colvin <john.loughran.colvin gmail.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.

 Still a few things missing from the changelog, there is a new 
 package std.experimental.ndslice, and native (DWARF based) 
 exception handling on linux.

 http://dlang.org/download.html#dmd_beta 
 http://dlang.org/changelog/2.070.0.html

 Please report any bugs at https://issues.dlang.org

 -Martin


Very pleased to say that all of DlangScience worked with this 
beta, no regressions/breakage.
Jan 11 2016
prev sibling next sibling parent Brian Schott <briancschott gmail.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 First beta for the 2.070.0 release.

 Still a few things missing from the changelog, there is a new 
 package std.experimental.ndslice, and native (DWARF based) 
 exception handling on linux.

 http://dlang.org/download.html#dmd_beta 
 http://dlang.org/changelog/2.070.0.html

 Please report any bugs at https://issues.dlang.org

 -Martin


Both my projects and EMSI's build and pass their tests with this 
beta.
Jan 11 2016
prev sibling next sibling parent reply =?UTF-8?B?Tm9yZGzDtnc=?= <per.nordlow gmail.com> writes:
On Sunday, 3 January 2016 at 19:24:57 UTC, Martin Nowak wrote:

 http://dlang.org/changelog/2.070.0.html


I'm missing changelog entry for

- new algorithm `std.algorithm.comparison.either`
- update for return type of `findSplit*` enabling bool-conversion 
in for instance

if (const hit = haystack.findSplit(needle))
{
     // use hit
}
Jan 14 2016
parent Martin Nowak <code+news.digitalmars dawg.eu> writes:
On 01/14/2016 11:13 AM, Nordlöw wrote:

 I'm missing changelog entry for
 
 - new algorithm `std.algorithm.comparison.either`
 - update for return type of `findSplit*` enabling bool-conversion in for
 instance
 
 if (const hit = haystack.findSplit(needle))
 {
     // use hit
 }


There is a changelog for each project, those entries should be part of PRs.
https://github.com/D-Programming-Language/phobos/blob/7f24ccfeb503445dc56fee0c3e7ee41268ef80a5/changelog.dd
Jan 17 2016
prev sibling next sibling parent reply Adam D. Ruppe <destructionator gmail.com> writes:
Has anyone built a Windows program with the beta? I tried and got 
undefined identifier HWND, but have been unable to minimize the 
test case and it might be just my install not being clean.
Jan 15 2016
parent Rainer Schuetze <r.sagitario gmx.de> writes:
On 16.01.2016 05:09, Adam D. Ruppe wrote:

 Has anyone built a Windows program with the beta? I tried and got
 undefined identifier HWND, but have been unable to minimize the test
 case and it might be just my install not being clean.


Works for me to build Visual D, though it uses it's own windows headers 
most of the time. There are a few tools that use HWND from 
core.sys.windows that seem ok.

I noticed the installer downloads Visual D 0.3.42, this should be 
updated to the latest version: 
https://github.com/D-Programming-Language/installer/pull/164
Jan 16 2016
prev sibling parent Martin Nowak <code+news.digitalmars dawg.eu> writes:
Second and last beta for the 2.070.0 release.

http://dlang.org/download.html#dmd_beta
http://dlang.org/changelog/2.070.0.html

Please report any bugs at https://issues.dlang.org

-Martin
Jan 17 2016