yaml-cpp: update to 0.9

[stephanlachnit]

No description provided.

[stephanlachnit]

I think this is fine to merge:

• Alpine armv7, Alpine x86 and VS cl x86 fails because fp precision tests fail (alright, nothing to be done about it)
• MSYS2 UCRT64 fails because of some locale setting (also alright IMO)

[bgilbert]

For the record, since CI logs age out:

UCRT64:

[  FAILED  ] NodeEmitterTest.RobustAgainstLocale

The 32-bit platforms:

[  FAILED  ] EmitterTest.NumberPrecision
[  FAILED  ] EmitterTest.DefaultPrecision
[  FAILED  ] EmitterTest.SetPrecision

The latter at least seem concerning; those tests aren't new and there are no CI failures in the previous release.

[stephanlachnit]

The 32-bit platforms:

[  FAILED  ] EmitterTest.NumberPrecision
[  FAILED  ] EmitterTest.DefaultPrecision
[  FAILED  ] EmitterTest.SetPrecision

The latter at least seem concerning; those tests aren't new and there are no CI failures in the previous release.

Not new, but the float to string implementation has been changed since 0.8 in jbeder/yaml-cpp@bd070a7.

These are the failures:

  [ RUN      ] EmitterTest.NumberPrecision
  ../subprojects/yaml-cpp-yaml-cpp-0.9.0/test/integration/emitter_test.cpp:30: Failure
  Expected equality of these values:
    expected
      Which is: "- 3.14\n- 54\n- 2.4e+06\n- 2e+06\n- 1e+07"
    out.c_str()
      Which is: "- 3.14\n- 54\n- 2.4e+05\n- 2.8e+05\n- 1.2e+05"


  [ RUN      ] EmitterTest.DefaultPrecision
  ../subprojects/yaml-cpp-yaml-cpp-0.9.0/test/integration/emitter_test.cpp:30: Failure
  Expected equality of these values:
    expected
      Which is: "- 1.3125\n- 1.23455810546875"
    out.c_str()
      Which is: "- 1.3125\n- 1.270590651e-05"


  [ RUN      ] EmitterTest.SetPrecision
  ../subprojects/yaml-cpp-yaml-cpp-0.9.0/test/integration/emitter_test.cpp:30: Failure
  Expected equality of these values:
    expected
      Which is: "- 1.31\n- 1.23456"
    out.c_str()
      Which is: "- 1.31\n- 1.27059e-05"

After searching for a bit, this is jbeder/yaml-cpp#1399.