[Issue 15192] DIP25: Nested ref returns are type checked unsoundly
https://issues.dlang.org/show_bug.cgi?id=15192 github-bugzi...@puremagic.com changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --
[Issue 15192] DIP25: Nested ref returns are type checked unsoundly
https://issues.dlang.org/show_bug.cgi?id=15192 --- Comment #3 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/dmd https://github.com/dlang/dmd/commit/ad419f24f498526e112f07bf9fbe26e31de319bc fix Issue 15192 - DIP25: Nested ref returns are type checked unsoundly https://github.com/dlang/dmd/commit/feb06a7548341ab1d794061d66850c611a7fb074 Merge pull request #5893 from WalterBright/fix15192 fix Issue 15192 - DIP25: Nested ref returns are type checked unsoundly --
[Issue 15193] DIP25 (implementation): Lifetimes of temporaries tracked incorrectly
https://issues.dlang.org/show_bug.cgi?id=15193 Walter Bright changed: What|Removed |Added Keywords||pull --- Comment #1 from Walter Bright --- https://github.com/dlang/dmd/pull/5895 --
[Issue 16195] delete should be @system
https://issues.dlang.org/show_bug.cgi?id=16195 github-bugzi...@puremagic.com changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --
[Issue 16195] delete should be @system
https://issues.dlang.org/show_bug.cgi?id=16195 --- Comment #3 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/dmd https://github.com/dlang/dmd/commit/e64ae1d3e5aa078a036242864a68499617c9b278 fix Issue 16195 - delete should be @system https://github.com/dlang/dmd/commit/71ecbed629c830fd10c7923cf6298cfce7f54ee7 Merge pull request #5887 from WalterBright/fix16195 fix Issue 16195 - delete should be @system --
[Issue 15193] DIP25 (implementation): Lifetimes of temporaries tracked incorrectly
https://issues.dlang.org/show_bug.cgi?id=15193 github-bugzi...@puremagic.com changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |FIXED --
[Issue 15193] DIP25 (implementation): Lifetimes of temporaries tracked incorrectly
https://issues.dlang.org/show_bug.cgi?id=15193 --- Comment #2 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/dmd https://github.com/dlang/dmd/commit/f2c50d0ba2fdff00bc58fa6be732e405f2ddf600 fix Issue 15193 - DIP25 (implementation): Lifetimes of temporaries tracked incorrectly https://github.com/dlang/dmd/commit/2b297099a0d8acb241a1032b7783628914d4c5c2 Merge pull request #5895 from WalterBright/fix15193 fix Issue 15193 - DIP25 (implementation): Lifetimes of temporaries tracked incorrectly --
[Issue 16224] New: -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
Issue ID: 16224
Summary: -cov marks the last line of do/while(0); as uncovered
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: enhancement
Priority: P1
Component: dmd
Assignee: nob...@puremagic.com
Reporter: and...@erdani.com
Consider:
int fun(int x)
{
do
{
if (x != 0)
break;
if (x != 1)
break;
if (x != -1)
break;
return x;
}
while (0);
return x * x * x;
}
unittest
{
fun(0);
fun(1);
fun(2);
}
The do/while(0); idiom is a trick used to avoid goto or additional state
variables. It simply allows code to jump to a common point by using "break".
The coverage analyzer marks the line with "while (0);" as uncovered, although
that is an useless tidbit.
--
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224 hst...@quickfur.ath.cx changed: What|Removed |Added CC||hst...@quickfur.ath.cx --- Comment #1 from hst...@quickfur.ath.cx --- So what's the problem here? --
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224 --- Comment #2 from hst...@quickfur.ath.cx --- Nevermind, I get it now. Sorry for the noise. --
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
--- Comment #3 from Andrei Alexandrescu ---
For clarity, I'll paste the code (fixed to do something useful, i.e. compute
the cube of an int) and the listing:
int cube(int x)
{
do
{
if (x == 0)
break;
if (x == 1)
break;
if (x == -1)
break;
return x * x * x;
}
while (0);
return x;
}
unittest
{
cube(0);
cube(1);
cube(-1);
cube(2);
}
=
|int cube(int x)
|{
|do
|{
4|if (x == 0)
1|break;
3|if (x == 1)
1|break;
2|if (x == -1)
1|break;
1|return x * x * x;
|}
000|while (0);
3|return x;
|}
|
|unittest
|{
1|cube(0);
1|cube(1);
1|cube(-1);
1|cube(2);
|}
--
[Issue 16225] New: Internal error cod1.c 1338 with -O
https://issues.dlang.org/show_bug.cgi?id=16225
Issue ID: 16225
Summary: Internal error cod1.c 1338 with -O
Product: D
Version: D2
Hardware: x86_64
OS: Linux
Status: NEW
Severity: enhancement
Priority: P1
Component: dmd
Assignee: nob...@puremagic.com
Reporter: 4bur...@gmail.com
When using -O with dmd fails with the following code (reduced):
```
struct C
{
hash_t foo( )
{
int y;
return ((cast(ubyte*)&y)[1]);
}
}
```
```
nemanjaboric:/home/nemanjaboric/test-clean.reduced $ dmd -O testreduced.d
Internal error: backend/cod1.c 1338
```
```
DMD64 D Compiler v2.071.1
```
--
[Issue 16225] Internal error cod1.c 1338 with -O
https://issues.dlang.org/show_bug.cgi?id=16225 Dicebot changed: What|Removed |Added CC||pub...@dicebot.lv Severity|enhancement |critical --
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
Steven Schveighoffer changed:
What|Removed |Added
CC||schvei...@yahoo.com
--- Comment #4 from Steven Schveighoffer ---
I don't see why while(true) is any worse.
Essentially:
while(true)
{
...
break;
}
Is the same as the do...while(0)
--
[Issue 16225] Internal error cod1.c 1338 with -O
https://issues.dlang.org/show_bug.cgi?id=16225 Steven Schveighoffer changed: What|Removed |Added Keywords||ice-on-valid-code --
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
--- Comment #5 from Andrei Alexandrescu ---
(In reply to Steven Schveighoffer from comment #4)
> I don't see why while(true) is any worse.
>
> Essentially:
>
> while(true)
> {
> ...
> break;
> }
>
> Is the same as the do...while(0)
Yah, switch is also fine:
switch (0)
{
default:
}
It just seems to me do/while(0) is the clearest, most obvious, and the least
bug-prone.
--
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
--- Comment #6 from Johan Engelen ---
> The coverage analyzer marks the line with "while (0);" as uncovered,
> although that is an useless tidbit.
I don't think it is useless. In the OP example, while(0) is uncovered because
there is never a "fallthrough exit" of that scope. Consider something like this
where coverage of while(0) is useful.
int fun(int x)
{
do
{
if (x != 0)
break;
if (x != 1)
break;
if (x != -1)
break;
mightThrow(x);
}
while (0);
return x * x * x;
}
--
[Issue 16225] Internal error cod1.c 1338 with -O
https://issues.dlang.org/show_bug.cgi?id=16225 --- Comment #1 from Nemanja Boric <4bur...@gmail.com> --- Looks like it got introduced in 2.068 --
[Issue 16224] -cov marks the last line of do/while(0); as uncovered
https://issues.dlang.org/show_bug.cgi?id=16224
--- Comment #7 from Andrei Alexandrescu ---
(In reply to Johan Engelen from comment #6)
> > The coverage analyzer marks the line with "while (0);" as uncovered,
> > although that is an useless tidbit.
>
> I don't think it is useless. In the OP example, while(0) is uncovered
> because there is never a "fallthrough exit" of that scope. Consider
> something like this where coverage of while(0) is useful.
>
> int fun(int x)
> {
> do
> {
> if (x != 0)
> break;
> if (x != 1)
> break;
> if (x != -1)
> break;
> mightThrow(x);
> }
> while (0);
> return x * x * x;
> }
In this case I agree coverage should be as usual. The presence of "return"
changes things.
--
[Issue 16225] [REG 2.068] Internal error cod1.c 1338 with -O
https://issues.dlang.org/show_bug.cgi?id=16225 Steven Schveighoffer changed: What|Removed |Added CC||schvei...@yahoo.com Summary|Internal error cod1.c 1338 |[REG 2.068] Internal error |with -O |cod1.c 1338 with -O OS|Linux |All Severity|critical|regression --- Comment #2 from Steven Schveighoffer --- Confirms it fails also on MacOS --
[Issue 16142] Adding a dtor / postblit (even disabled) forces opAssign
https://issues.dlang.org/show_bug.cgi?id=16142 --- Comment #3 from github-bugzi...@puremagic.com --- Commits pushed to master at https://github.com/dlang/dmd https://github.com/dlang/dmd/commit/517f14a949a2998e51517cbef444f2b3208a3789 Fix issue 16142: Disabled opEquals is overriden when a dtor/postblit is present In order to pick up disabled opEquals, we need to process the members of the struct anyway even if there is a dtor / postblit. https://github.com/dlang/dmd/commit/113cf3f4f8b4a786d7d7dca6bfc62ed024d12af9 Merge pull request #5854 from mathias-lang-sociomantic/opassign-2 Fix issue 16142: Disabled opAssign is overriden when a dtor/postblit is present --
[Issue 16226] New: -dip25 doesn't work if the return type is not explicit
https://issues.dlang.org/show_bug.cgi?id=16226
Issue ID: 16226
Summary: -dip25 doesn't work if the return type is not explicit
Product: D
Version: D2
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: P1
Component: dmd
Assignee: nob...@puremagic.com
Reporter: issues.dl...@jmdavisprog.com
This code results in an error with -dip25:
void main() @safe
{
int i;
foo(i);
}
ref int foo(ref int bar) @safe
{
return bar;
}
but this code
void main() @safe
{
int i;
foo(i);
}
ref foo(ref int bar) @safe
{
return bar;
}
and this code
void main() @safe
{
int i;
foo(i);
}
ref auto foo(ref int bar) @safe
{
return bar;
}
do not result in an error. Note that unless the return type actually includes
int explictly rather than letting it be inferred, -dip25 is bypassed
completely. Interestingly enough, it does catch this case though and error out
correctly:
void main() @safe
{
foo();
}
ref foo() @safe
{
int bar = 42;
return bar;
}
--
[Issue 16226] -dip25 doesn't work if the return type is not explicit
https://issues.dlang.org/show_bug.cgi?id=16226 Jonathan M Davis changed: What|Removed |Added Keywords||safe --
