Removing NO_AUTO_PTR as its a cause of warning generation and not needed#45
Removing NO_AUTO_PTR as its a cause of warning generation and not needed#45
Conversation
|
I guess it is enough to just define it? Does it need the empty value? |
|
It does need the empty no value. If not it gets value of 1 which gives
Warnings
tor. 26. nov. 2020 22.58 skrev Morten V. Pedersen <notifications@github.com
…:
I guess it is enough to just define it? Does it need the empty value?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#45 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AH2XM3JCBZ7OJRQ3M3MPIOLSR3FRTANCNFSM4T7OJ5DA>
.
|
|
Simply defining it should make it work with #ifdef etc., are they doing something else with that marco (like checking it's value)? |
|
The = no value is simply defining it. Cli and preprocessor works
differently. Sent from my phone.
fre. 27. nov. 2020 00.32 skrev Morten V. Pedersen <notifications@github.com
…:
Simply defining it should make it work with #ifdef etc., are they doing
something else with that marco (like checking it's value)?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#45 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AH2XM3KWHZQKJ6WQYT5FK33SR3QPJANCNFSM4T7OJ5DA>
.
|
|
Ok, just saw you wrote that the default value of 1 gives warnings. Where is that happening? In Thrift? |
|
Its seen with boost, win+c++17 combo. The thing is it was 1 on commandline.
And in code just defined.
lør. 28. nov. 2020 10.51 skrev Morten V. Pedersen <notifications@github.com
…:
Ok, just saw you wrote that the default value of 1 gives warnings. Where
is that happening? In Thrift?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#45 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AH2XM3PKMAGURAA6TJKM4NTSSDB2XANCNFSM4T7OJ5DA>
.
|
|
I just to get what kind of warning that produces.. since most code just checks whether it is defined e.g. with |
|
Ok, its called macro redifinition
lør. 28. nov. 2020 15.12 skrev Morten V. Pedersen <notifications@github.com
…:
I just to get what kind of warning that produces.. since most code just
checks whether it is defined e.g. with #ifdef
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#45 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AH2XM3NYJ65KUK64WCBOB5TSSEAL7ANCNFSM4T7OJ5DA>
.
|
|
So this means that we are not the only ones defining it? Where is it being defined (besides by us)? |
|
Within boost
søn. 29. nov. 2020 21.55 skrev Morten V. Pedersen <notifications@github.com
…:
So this means that we are not the only ones defining it? Where is it being
defined (besides by us)?
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#45 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AH2XM3OD2TVYPEMDSLD4XMTSSKYKVANCNFSM4T7OJ5DA>
.
|
|
I see, in that case would the right solution not be to ensure that it isn't defined twice by guarding that |
|
For example here BOOST should probably check whether that macro has already been defined: Because on older compilers they do not define it - but there it will cause deprecation warnings.. So to have have it both on old and new you need to provide it yourself. Unless you want a complicated build script where you check the version of the compiler and then optionally provide it. |
2 participants
No description provided.