<table border="1" cellspacing="0" cellpadding="8">
    <tr>
        <th>Issue</th>
        <td>
            <a href=https://github.com/llvm/llvm-project/issues/69319>69319</a>
        </td>
    </tr>

    <tr>
        <th>Summary</th>
        <td>
            clang-tidy: complains about const_cast adding const [cppcoreguidelines-pro-type-const-cast]
        </td>
    </tr>

    <tr>
      <th>Labels</th>
      <td>
            clang-tidy
      </td>
    </tr>

    <tr>
      <th>Assignees</th>
      <td>
      </td>
    </tr>

    <tr>
      <th>Reporter</th>
      <td>
          BikingGlobetrotter
      </td>
    </tr>
</table>

<pre>
    For the following code, clang-tidy raises a "cppcoreguidelines-pro-type-const-cast" issue.

`
#include <algorithm>
#include <array>
#include <charconv>
#include <cstdio>
#include <string>
#include <type_traits>

template <std::size_t N>
void AppendToCharsResult(
        std::wstring& str, const char (&buffer)[N], std::to_chars_result result)
{
        // simple conversion from char to wchar_t (as no critical characters are expected)
        wchar_t wbuffer[N];
        auto wbufferend =
                std::copy(static_cast<const char*>(buffer), const_cast<const char*>(result.ptr), wbuffer);
        str.insert(str.end(), wbuffer, wbufferend);
}

template <typename T>
std::enable_if_t<std::is_integral_v<T>> AppendValue(std::wstring& str, const T& value)
{
        char buffer[512];
        AppendToCharsResult(str, buffer, std::to_chars(buffer, buffer + sizeof(buffer), value));
}


int main()
{
    std::wstring s;
 AppendValue(s, 5);
    return 0;
}
`

It recommends to not use the const_cast, as "Modifying a variable that was declared const is undefined behavior, even with const_cast.". Actually, we are "adding" constness, so this cannot result in undefined behavior. 

The calling code, cannot remove the const_cast completely, since this results in a compiler that not know how to call std::copy with having a `const char*` and `mutable char*`. Replacing const_cast with static_cast would be possible, but that is less specific.

For me, it looks like the cppcoreguidelines-pro-type-const-cast test is too aggressive not differentiating between casts that add constness and casts that remove constness.

See godbolt for reference.
https://godbolt.org/z/nd6hfKhb4
</pre>
<img width="1px" height="1px" alt="" src="http://email.email.llvm.org/o/eJyMVk1v4zgS_TX0pRBDpvwRHXxwnPZisdg-9ARzNSiyJHFCkQJZssf96wekZFvOJDN9UYzUx3v1qlikCEHXFnHLVi9s9ToTPTXOb1_0u7b1f4wrkbwjQj8rnbpsD84DNQiVM8adta1BOoWM70EaYesn0uoCXuiAAQQwzmXXSeex7rVCoy2Gp867J7p0-CSdDfQkRSDGOegQepyz7JVlu_G7zsYfPNdWml4hsHwvTO28pqZl-bdP7d6Lyxc22QgvnT19ZQ6ktPvCGMhrW39hjAUdyQtN4e6RvoRtZwSNKRTLdyzfBf0TjwTfb74npxXsug6tenP7RvjwA0NviPHna7LiFn0eqfA1BPJJ_CglxOogRvB12VcVesYLtnr5zlav0ekWT-4YXcPRJwzwI1QxQm1ebpiMHxg_QNBtZzDCnNAH7SxU3rUDIDk4xx9HitgigHUgvSYthUkeQhL6AMIj4J8dSkJ1x8qKa_B55DwSzu8kRB8xBjNaBSx_vROc6CJdd2H8OZAgLY9psPL9XRrGd1Fv_nwX5yrdPzgP4sw7ugac7-ETjoH8XNuAnhIDP0erUiseY_aTOh4ybF6_Gpo4Wla0CG-3ablVjFaUBo-6OtJ0vHQ4aktYe2GOJ5bvU2T-bRyw34XpMbH8l3l6i_85Dd6fzEbq_q1pqwX_0LbPx3lEuAvyt7GcdOjqB4y_QDw0rvrYvxu_r9UcvtoStELbsSuP5QAAfNQDwi3hR-Ui8GqKGOM9Uu8tZJ_QuG2y9P1vPHLStS1aFeL5sY6gD5g262Qe-R5EiEv0_07p6hIpCTgJr2PTgRpBcBYBFEojPKqxazpAbxVW2qKCEhtx0i5JiSe0cNbUTDDmjPM57CT1wphLmk9MJ5VxLpRKU8EHf4sh1R0cUKMDSGEj73GHaPsJ6hymZb_F6oQxD7fGNUnrTh_rB-ni1iEciAVtJQ7QA2aIoCJ5aYN-ECRme7fuDI07R2kjIDxsiEGCSDDpydbZ46lfZyDikllnbU9J6btlDj-wM0IOJdyIpoyTtQNn15soA3QuBF0aHEaZBo46gMEQIHQodaXlw60Xb9g2-WsC49x7AKPfR21-5TIFwmEKyDkQde0xBH3CpIzSw-4hLSjWUCKdES3EuDCQE0rd-52UmBjHPt3sD8x_Q4TaqdIZgsp58Jiw5PVWb4i6EPuQ7pTRc-58zfjhJ-MHq9ZN9b-mXM7UNldFXogZbhfrYrNaZsvn51mzLRZLtcSsWi9RFZvNphTLTVYt82qRZ4USYqa3POP5IltsFotssdjMs0qqVbFalqrKVVEiW2bYCm3mxpzaiD1L747tusgXxcyIEk1ITyHO7w8axuNqm_ltDHoq-zqwZWZ0oHBPQ5oMbicx-W6YX6FtAFG6nqYTM5yt8cSy1cuvPZNWr7Pem-0HITU1fTmXrmX8EPmMf2KGP1AS44dUY2D8kMr8KwAA___-kDEq">