Align format of override config.yaml with base cfg

setup.py

@@ -783,7 +783,7 @@ def parse_config_vars(conf: Dict[str, str]) -> Dict[str, str]:
     for k, v in conf.items():
         if isinstance(v, str) and (v.startswith("0x") or k == 'PRESET_BASE' or k == 'CONFIG_NAME'):
             # Represent byte data with string, to avoid misinterpretation as big-endian int.
-            # Everything is either byte data or an integer, with PRESET_BASE as one exception.
+            # Everything except PRESET_BASE and CONFIG_NAME is either byte data or an integer.
             out[k] = f"'{v}'"
         else:
             out[k] = str(int(v))

tests/core/pyspec/eth2spec/gen_helpers/gen_base/gen_runner.py

@@ -1,3 +1,4 @@
+from eth_utils import encode_hex
 import os
 import time
 import shutil
@@ -95,6 +96,20 @@ def run_generator(generator_name, test_providers: Iterable[TestProvider]):
     yaml = YAML(pure=True)
     yaml.default_flow_style = None
 
+    # Spec config is using a YAML subset
+    cfg_yaml = YAML(pure=True)
+    cfg_yaml.default_flow_style = False  # Emit separate line for each key
+
+    def cfg_represent_bytes(self, data):
+        return self.represent_int(encode_hex(data))
+
+    cfg_yaml.representer.add_representer(bytes, cfg_represent_bytes)
+
+    def cfg_represent_quoted_str(self, data):
+        return self.represent_scalar(u'tag:yaml.org,2002:str', data, style="'")
+
+    cfg_yaml.representer.add_representer(context.quoted_str, cfg_represent_quoted_str)
+
     log_file = Path(output_dir) / 'testgen_error_log.txt'
 
     print(f"Generating tests into {output_dir}")
@@ -167,10 +182,14 @@ def run_generator(generator_name, test_providers: Iterable[TestProvider]):
                         written_part = True
                         if out_kind == "meta":
                             meta[name] = data
-                        if out_kind == "data":
-                            output_part("data", name, dump_yaml_fn(data, name, file_mode, yaml))
-                        if out_kind == "ssz":
-                            output_part("ssz", name, dump_ssz_fn(data, name, file_mode))
+                        elif out_kind == "cfg":
+                            output_part(out_kind, name, dump_yaml_fn(data, name, file_mode, cfg_yaml))
+                        elif out_kind == "data":
+                            output_part(out_kind, name, dump_yaml_fn(data, name, file_mode, yaml))
+                        elif out_kind == "ssz":
+                            output_part(out_kind, name, dump_ssz_fn(data, name, file_mode))
+                        else:
+                            assert False  # Unknown kind
                 except SkippedTest as e:
                     print(e)
                     skipped_test_count += 1

tests/core/pyspec/eth2spec/gen_helpers/gen_base/gen_typing.py

@@ -10,9 +10,10 @@ from dataclasses import dataclass
 # Elements: name, out_kind, data
 #
 # out_kind is the type of data:
+#  - "meta" for generic data to collect into a meta data dict
+#  - "cfg" for a spec config dictionary
 #  - "data" for generic
 #  - "ssz" for SSZ encoded bytes
-#  - "meta" for generic data to collect into a meta data dict.
 TestCasePart = NewType("TestCasePart", Tuple[str, str, Any])
 
 

tests/core/pyspec/eth2spec/test/context.py

@@ -1,7 +1,6 @@
 import pytest
 from dataclasses import dataclass
 import importlib
-from eth_utils import encode_hex
 
 from eth2spec.phase0 import mainnet as spec_phase0_mainnet, minimal as spec_phase0_minimal
 from eth2spec.altair import mainnet as spec_altair_mainnet, minimal as spec_altair_minimal
@@ -506,18 +505,22 @@ def with_presets(preset_bases, reason=None):
     return decorator
 
 
+class quoted_str(str):
+    pass
+
+
 def _get_basic_dict(ssz_dict: Dict[str, Any]) -> Dict[str, Any]:
     """
-    Get dict of Python built-in types from a dict of SSZ objects.
+    Get dict of basic types from a dict of SSZ objects.
     """
     result = {}
     for k, v in ssz_dict.items():
         if isinstance(v, int):
             value = int(v)
         elif isinstance(v, bytes):
-            value = encode_hex(v)
+            value = bytes(bytearray(v))
         else:
-            value = str(v)
+            value = quoted_str(v)
         result[k] = value
     return result
 
@@ -553,7 +556,7 @@ def with_config_overrides(config_overrides):
             # To output the changed config to could be serialized with yaml test vectors,
             # the dict SSZ objects have to be converted into Python built-in types.
             output_config = _get_basic_dict(modified_config)
-            yield 'config', 'data', output_config
+            yield 'config', 'cfg', output_config
 
             spec.config = spec.Configuration(**modified_config)
 

tests/core/pyspec/eth2spec/test/utils/utils.py

@@ -17,6 +17,7 @@ def vector_test(description: str = None):
         # this wraps the function, to yield type-annotated entries of data.
         # Valid types are:
         #   - "meta": all key-values with this type can be collected by the generator, to put somewhere together.
+        #   - "cfg": spec config dictionary
         #   - "ssz": raw SSZ bytes
         #   - "data": a python structure to be encoded by the user.
         def entry(*args, **kw):