[clang-tidy] create a check that looks for constructing a string from a nullptr

[LegalizeAdulthood]

Consider the following code:

std::string foo()
{
    return nullptr;
}

See compiler explorer example

This code compiles just fine with no warnings or errors, but crashes when you run it. We should be able to detect cases where a nullptr is used to construct a string, at least in the simple case above. I had a real-world bug that was exactly the above scenario, although I have simplified the example to remove irrelevant code. Obviously more sophisticated data flow analysis can propagate nullptr through usages until it hits a std::basic_string constructor, but even detecting the above simple case would be a useful start.

Note: there are several constructor overloads that take a pointer to a character as an argument and they are all invalid if the pointer is nullptr.

[llvmbot]

@llvm/issue-subscribers-clang-tidy

Author: Richard Thomson (LegalizeAdulthood)

Consider the following code:

std::string foo()
{
    return nullptr;
}

See compiler explorer example

This code compiles just fine with no warnings or errors, but crashes when you run it. We should be able to detect cases where a nullptr is used to construct a string, at least in the simple case above. I had a real-world bug that was exactly the above scenario, although I have simplified the example to remove irrelevant code. Obviously more sophisticated data flow analysis can propagate nullptr through usages until it hits a std::basic_string constructor, but even detecting the above simple case would be a useful start.

[llvmbot]

Hi!

This issue may be a good introductory issue for people new to working on LLVM. If you would like to work on this issue, your first steps are:

1. Check that no other contributor has already been assigned to this issue. If you believe that no one is actually working on it despite an assignment, ping the person. After one week without a response, the assignee may be changed.
2. In the comments of this issue, request for it to be assigned to you, or just create a pull request after following the steps below. Mention this issue in the description of the pull request.
3. Fix the issue locally.
4. Run the test suite locally. Remember that the subdirectories under test/ create fine-grained testing targets, so you can e.g. use make check-clang-ast to only run Clang's AST tests.
5. Create a Git commit.
6. Run git clang-format HEAD~1 to format your changes.
7. Open a pull request to the upstream repository on GitHub. Detailed instructions can be found in GitHub's documentation. Mention this issue in the description of the pull request.

If you have any further questions about this issue, don't hesitate to ask via a comment in the thread below.

[llvmbot]

@llvm/issue-subscribers-good-first-issue

Author: Richard Thomson (LegalizeAdulthood)

Consider the following code:

std::string foo()
{
    return nullptr;
}

See compiler explorer example

This code compiles just fine with no warnings or errors, but crashes when you run it. We should be able to detect cases where a nullptr is used to construct a string, at least in the simple case above. I had a real-world bug that was exactly the above scenario, although I have simplified the example to remove irrelevant code. Obviously more sophisticated data flow analysis can propagate nullptr through usages until it hits a std::basic_string constructor, but even detecting the above simple case would be a useful start.

[chrchr-github]

Caught by bugprone-string-constructor and clang-analyzer-cplusplus.StringChecker:

<source>:9:12: warning: constructing string from nullptr is undefined behaviour [bugprone-string-constructor]
    9 |     return nullptr;
      |            ^
<source>:9:12: warning: The parameter must not be null [clang-analyzer-cplusplus.StringChecker]

[vaisakh-prod]

Hi @chrchr-github,
are you looking into this??
@LegalizeAdulthood
I'd like to give have a look at this.

[chrchr-github]

Not sure what there is to do, with two checks already reporting the issue...

[carlosgalvezp]

Then I would say we don't need a third check ;)

[LegalizeAdulthood]

which two checks report the issue?

[LegalizeAdulthood]

oic now github mobile wasn't showing all the comments (how unhelpful!)