Is there a way to make member function NOT callable from constructor?
.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty{ height:90px;width:728px;box-sizing:border-box;
}
I have member function (method) which uses
std::enable_shared_from_this::weak_from_this()
In short: weak_from_this returns weak_ptr to this. One caveat is it can't be used from constructor.
If somebody would use my function from constructor of inherited class, weak_from_this inside it would return expired weak_ptr. I guard against that with assertion checking that it's not expired, but it's a run-time check.
Is there a way to check against it at compile time?
c++ c++17 shared-ptr weak-ptr
edited 12 hours ago
armitus
524114
asked 12 hours ago
KorriKorri
35128
add a comment |
I have member function (method) which uses
std::enable_shared_from_this::weak_from_this()
In short: weak_from_this returns weak_ptr to this. One caveat is it can't be used from constructor.
If somebody would use my function from constructor of inherited class, weak_from_this inside it would return expired weak_ptr. I guard against that with assertion checking that it's not expired, but it's a run-time check.
Is there a way to check against it at compile time?
c++ c++17 shared-ptr weak-ptr
edited 12 hours ago
armitus
524114
asked 12 hours ago
KorriKorri
35128
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
2
Nice question. One way would be to make a dummy class with pure virtual functionweak_from_thisand inherit yours from it. This will make it a hard compile error.
– SergeyA
12 hours ago
1
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago
add a comment |
I have member function (method) which uses
std::enable_shared_from_this::weak_from_this()
In short: weak_from_this returns weak_ptr to this. One caveat is it can't be used from constructor.
If somebody would use my function from constructor of inherited class, weak_from_this inside it would return expired weak_ptr. I guard against that with assertion checking that it's not expired, but it's a run-time check.
Is there a way to check against it at compile time?
c++ c++17 shared-ptr weak-ptr
edited 12 hours ago
armitus
524114
asked 12 hours ago
KorriKorri
35128
I have member function (method) which uses
std::enable_shared_from_this::weak_from_this()
In short: weak_from_this returns weak_ptr to this. One caveat is it can't be used from constructor.
If somebody would use my function from constructor of inherited class, weak_from_this inside it would return expired weak_ptr. I guard against that with assertion checking that it's not expired, but it's a run-time check.
Is there a way to check against it at compile time?
c++ c++17 shared-ptr weak-ptr
c++ c++17 shared-ptr weak-ptr
edited 12 hours ago
armitus
524114
asked 12 hours ago
KorriKorri
35128
edited 12 hours ago
armitus
524114
asked 12 hours ago
KorriKorri
35128
edited 12 hours ago
armitus
524114
edited 12 hours ago
armitus
524114
edited 12 hours ago
armitus
524114
524114
asked 12 hours ago
KorriKorri
35128
asked 12 hours ago
KorriKorri
35128
asked 12 hours ago
KorriKorri
35128
35128
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
2
Nice question. One way would be to make a dummy class with pure virtual functionweak_from_thisand inherit yours from it. This will make it a hard compile error.
– SergeyA
12 hours ago
1
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago
add a comment |
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
2
Nice question. One way would be to make a dummy class with pure virtual functionweak_from_thisand inherit yours from it. This will make it a hard compile error.
– SergeyA
12 hours ago
1
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
2
2
Nice question. One way would be to make a dummy class with pure virtual function
weak_from_this and inherit yours from it. This will make it a hard compile error.– SergeyA
12 hours ago
Nice question. One way would be to make a dummy class with pure virtual function
weak_from_this and inherit yours from it. This will make it a hard compile error.– SergeyA
12 hours ago
1
1
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago
add a comment |
3 Answers
3
active
oldest
votes
I am afraid the answer is "no, it's not possible to protect against this at compile-time." It's always difficult to prove a negative, but consider this: if it was possible to protect a function this way, it would probably have been done for weak_from_this and shared_from_this in the standard library itself.
answered 12 hours ago
AngewAngew
135k11261354
add a comment |
No there is no way. Consider:
void call_me(struct widget*);
struct widget : std::enable_shared_from_this<widget> {
widget() {
call_me(this);
}
void display() {
shared_from_this();
}
};
// later:
void call_me(widget* w) {
w->display(); // crash
}
The thing is there is a reason you want to check for not calling shared_from_this in the constructor. Think about that reason. It's not that shared_from_this cannot be called, it's because it's return value has no way of being assigned yet. It is also not because it will never be assigned. It's because it will be assigned later in the execution of the code. Order of operation is a runtime property of your program. You cannot assert at compile time for order of operation, which is done at runtime.
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
add a comment |
Not as such, but - if performance is not an issue, you could add a flag which indicates construction is complete, and use that to fail at run-time with such calls:
class A {
// ... whatever ...
A() {
// do construction work
constructed = true;
}
foo() {
if (not constructed) {
throw std::logic_error("Cannot call foo() during construction");
}
// the rest of foo
}
bool constructed { false };
}
You could also make these checks only apply when compiling in DEBUG mode (e.g. with conditional compilation using the preprocessor - #ifndef NDEBUG) so that at run time you won't get the performance penalty. Mind the noexcepts though.
An alternative to throwing could be assert()'ing.
answered 10 hours ago
einpoklumeinpoklum
37k28132263
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go withassert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.
– Korri
9 hours ago
@Korri remember thatasserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; firstassertthenthrow, or justthrow.
– Jesper Juhl
8 hours ago
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55576192%2fis-there-a-way-to-make-member-function-not-callable-from-constructor%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
3 Answers
3
active
oldest
votes
3 Answers
3
active
oldest
votes
active
oldest
votes
active
oldest
votes
I am afraid the answer is "no, it's not possible to protect against this at compile-time." It's always difficult to prove a negative, but consider this: if it was possible to protect a function this way, it would probably have been done for weak_from_this and shared_from_this in the standard library itself.
answered 12 hours ago
AngewAngew
135k11261354
add a comment |
I am afraid the answer is "no, it's not possible to protect against this at compile-time." It's always difficult to prove a negative, but consider this: if it was possible to protect a function this way, it would probably have been done for weak_from_this and shared_from_this in the standard library itself.
answered 12 hours ago
AngewAngew
135k11261354
add a comment |
I am afraid the answer is "no, it's not possible to protect against this at compile-time." It's always difficult to prove a negative, but consider this: if it was possible to protect a function this way, it would probably have been done for weak_from_this and shared_from_this in the standard library itself.
answered 12 hours ago
AngewAngew
135k11261354
I am afraid the answer is "no, it's not possible to protect against this at compile-time." It's always difficult to prove a negative, but consider this: if it was possible to protect a function this way, it would probably have been done for weak_from_this and shared_from_this in the standard library itself.
answered 12 hours ago
AngewAngew
135k11261354
answered 12 hours ago
AngewAngew
135k11261354
answered 12 hours ago
AngewAngew
135k11261354
answered 12 hours ago
AngewAngew
135k11261354
135k11261354
add a comment |
add a comment |
No there is no way. Consider:
void call_me(struct widget*);
struct widget : std::enable_shared_from_this<widget> {
widget() {
call_me(this);
}
void display() {
shared_from_this();
}
};
// later:
void call_me(widget* w) {
w->display(); // crash
}
The thing is there is a reason you want to check for not calling shared_from_this in the constructor. Think about that reason. It's not that shared_from_this cannot be called, it's because it's return value has no way of being assigned yet. It is also not because it will never be assigned. It's because it will be assigned later in the execution of the code. Order of operation is a runtime property of your program. You cannot assert at compile time for order of operation, which is done at runtime.
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
add a comment |
No there is no way. Consider:
void call_me(struct widget*);
struct widget : std::enable_shared_from_this<widget> {
widget() {
call_me(this);
}
void display() {
shared_from_this();
}
};
// later:
void call_me(widget* w) {
w->display(); // crash
}
The thing is there is a reason you want to check for not calling shared_from_this in the constructor. Think about that reason. It's not that shared_from_this cannot be called, it's because it's return value has no way of being assigned yet. It is also not because it will never be assigned. It's because it will be assigned later in the execution of the code. Order of operation is a runtime property of your program. You cannot assert at compile time for order of operation, which is done at runtime.
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
add a comment |
No there is no way. Consider:
void call_me(struct widget*);
struct widget : std::enable_shared_from_this<widget> {
widget() {
call_me(this);
}
void display() {
shared_from_this();
}
};
// later:
void call_me(widget* w) {
w->display(); // crash
}
The thing is there is a reason you want to check for not calling shared_from_this in the constructor. Think about that reason. It's not that shared_from_this cannot be called, it's because it's return value has no way of being assigned yet. It is also not because it will never be assigned. It's because it will be assigned later in the execution of the code. Order of operation is a runtime property of your program. You cannot assert at compile time for order of operation, which is done at runtime.
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
No there is no way. Consider:
void call_me(struct widget*);
struct widget : std::enable_shared_from_this<widget> {
widget() {
call_me(this);
}
void display() {
shared_from_this();
}
};
// later:
void call_me(widget* w) {
w->display(); // crash
}
The thing is there is a reason you want to check for not calling shared_from_this in the constructor. Think about that reason. It's not that shared_from_this cannot be called, it's because it's return value has no way of being assigned yet. It is also not because it will never be assigned. It's because it will be assigned later in the execution of the code. Order of operation is a runtime property of your program. You cannot assert at compile time for order of operation, which is done at runtime.
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
answered 11 hours ago
Guillaume RacicotGuillaume Racicot
16.3k53872
16.3k53872
add a comment |
add a comment |
Not as such, but - if performance is not an issue, you could add a flag which indicates construction is complete, and use that to fail at run-time with such calls:
class A {
// ... whatever ...
A() {
// do construction work
constructed = true;
}
foo() {
if (not constructed) {
throw std::logic_error("Cannot call foo() during construction");
}
// the rest of foo
}
bool constructed { false };
}
You could also make these checks only apply when compiling in DEBUG mode (e.g. with conditional compilation using the preprocessor - #ifndef NDEBUG) so that at run time you won't get the performance penalty. Mind the noexcepts though.
An alternative to throwing could be assert()'ing.
answered 10 hours ago
einpoklumeinpoklum
37k28132263
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go withassert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.
– Korri
9 hours ago
@Korri remember thatasserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; firstassertthenthrow, or justthrow.
– Jesper Juhl
8 hours ago
add a comment |
Not as such, but - if performance is not an issue, you could add a flag which indicates construction is complete, and use that to fail at run-time with such calls:
class A {
// ... whatever ...
A() {
// do construction work
constructed = true;
}
foo() {
if (not constructed) {
throw std::logic_error("Cannot call foo() during construction");
}
// the rest of foo
}
bool constructed { false };
}
You could also make these checks only apply when compiling in DEBUG mode (e.g. with conditional compilation using the preprocessor - #ifndef NDEBUG) so that at run time you won't get the performance penalty. Mind the noexcepts though.
An alternative to throwing could be assert()'ing.
answered 10 hours ago
einpoklumeinpoklum
37k28132263
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go withassert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.
– Korri
9 hours ago
@Korri remember thatasserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; firstassertthenthrow, or justthrow.
– Jesper Juhl
8 hours ago
add a comment |
Not as such, but - if performance is not an issue, you could add a flag which indicates construction is complete, and use that to fail at run-time with such calls:
class A {
// ... whatever ...
A() {
// do construction work
constructed = true;
}
foo() {
if (not constructed) {
throw std::logic_error("Cannot call foo() during construction");
}
// the rest of foo
}
bool constructed { false };
}
You could also make these checks only apply when compiling in DEBUG mode (e.g. with conditional compilation using the preprocessor - #ifndef NDEBUG) so that at run time you won't get the performance penalty. Mind the noexcepts though.
An alternative to throwing could be assert()'ing.
answered 10 hours ago
einpoklumeinpoklum
37k28132263
Not as such, but - if performance is not an issue, you could add a flag which indicates construction is complete, and use that to fail at run-time with such calls:
class A {
// ... whatever ...
A() {
// do construction work
constructed = true;
}
foo() {
if (not constructed) {
throw std::logic_error("Cannot call foo() during construction");
}
// the rest of foo
}
bool constructed { false };
}
You could also make these checks only apply when compiling in DEBUG mode (e.g. with conditional compilation using the preprocessor - #ifndef NDEBUG) so that at run time you won't get the performance penalty. Mind the noexcepts though.
An alternative to throwing could be assert()'ing.
answered 10 hours ago
einpoklumeinpoklum
37k28132263
answered 10 hours ago
einpoklumeinpoklum
37k28132263
answered 10 hours ago
einpoklumeinpoklum
37k28132263
answered 10 hours ago
einpoklumeinpoklum
37k28132263
37k28132263
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go withassert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.
– Korri
9 hours ago
@Korri remember thatasserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; firstassertthenthrow, or justthrow.
– Jesper Juhl
8 hours ago
add a comment |
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go withassert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.
– Korri
9 hours ago
@Korri remember thatasserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; firstassertthenthrow, or justthrow.
– Jesper Juhl
8 hours ago
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go with
assert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.– Korri
9 hours ago
Since apparently there isn't compile-time solution which doesn't make code less readable, I decided to go with
assert(!wptr.expired()). I think it's a little bit more fitting than exception, because there is no way to recover from this situation.– Korri
9 hours ago
@Korri remember that
asserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; first assert then throw, or just throw.– Jesper Juhl
8 hours ago
@Korri remember that
asserts are usually compiled out in release builds, so nothing will happen. An exception however is not, so it would still terminate the program (if not caught and swallowed) in a release build. You could do both; first assert then throw, or just throw.– Jesper Juhl
8 hours ago
add a comment |
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55576192%2fis-there-a-way-to-make-member-function-not-callable-from-constructor%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Note there is a difference in scope between a child class constructor body and the parent class constructor: the latter has been executed completely before you even start initializing the child class's members (if any), let alone enter the child class constructor body.
– rubenvb
12 hours ago
2
Nice question. One way would be to make a dummy class with pure virtual function
weak_from_thisand inherit yours from it. This will make it a hard compile error.– SergeyA
12 hours ago
1
@SergeyA Why didn't you post that as an answer? All other people here seem to conclude that it's not possible so either your comment is wrong and misleading or they are wrong and you should show how it can be achieved.
– Bakuriu
6 hours ago