Tons of warnings about comparison between signed and unsigned integers

[TheQuantumPhysicist]

There are over 10 warnings on comparison between signed and unsigned integers. My compiler sees them by default; if you can't, use the flag -Wsign-compare. Some times it's because of auto, and others just direct comparison of int with sizes.

It's quite a good idea to explicitly define these types and avoid these warnings. I always try to make sure that my warnings list is empty, so that I can see serious things when they happen.

[TheQuantumPhysicist]

OK. It turned out not to be in lots of places, but two places that are repeated in many warnings. Here's the patch to fix this (I ensured the tests still pass):

diff --git a/include/jwt/base64.hpp b/include/jwt/base64.hpp
index 2f2a646..fcf99d6 100644
--- a/include/jwt/base64.hpp
+++ b/include/jwt/base64.hpp
@@ -94,7 +94,7 @@ std::string base64_encode(const char* in, size_t len)

   int i = 0;
   int j = 0;
-  for (; i < len - 2; i += 3) {
+  for (; i < static_cast<int>(len) - 2; i += 3) {
     const auto first  = in[i];
     const auto second = in[i+1];
     const auto third  = in[i+2];
diff --git a/include/jwt/impl/algorithm.ipp b/include/jwt/impl/algorithm.ipp
index c9fe51a..55d6591 100644
--- a/include/jwt/impl/algorithm.ipp
+++ b/include/jwt/impl/algorithm.ipp
@@ -306,9 +306,9 @@ std::string PEMSign<Hasher>::public_key_ser(

   ECDSA_SIG_get0(ec_sig.get(), &ec_sig_r, &ec_sig_s);

-  auto r_len = BN_num_bytes(ec_sig_r);
-  auto s_len = BN_num_bytes(ec_sig_s);
-  auto bn_len = (degree + 7) / 8;
+  int r_len = BN_num_bytes(ec_sig_r);
+  int s_len = BN_num_bytes(ec_sig_s);
+  int bn_len = static_cast<int>((degree + 7) / 8);

   if ((r_len > bn_len) || (s_len > bn_len)) {
     ec = AlgorithmErrc::SigningErr;

[arun11299]

Thanks again!!

Would you like to create a pull request for this patch so that I can commit your changes directly ?

[TheQuantumPhysicist]

Done!

Cheers!

[TheQuantumPhysicist]

I have found more similar warnings. I'll make another pull request.