ethspecify 0.2.5__py3-none-any.whl → 0.3.2__py3-none-any.whl

ethspecify/core.py

@@ -11,6 +11,91 @@ import tokenize
 import yaml
 
 
+def validate_exception_items(exceptions, version):
+    """
+    Validate that exception items actually exist in the spec.
+    Raises an exception if any item doesn't exist.
+    """
+    if not exceptions:
+        return
+
+    # Get the pyspec data
+    try:
+        pyspec = get_pyspec(version)
+    except Exception as e:
+        print(f"Warning: Could not validate exceptions - failed to load pyspec: {e}")
+        return
+
+    # Map exception keys to pyspec keys
+    exception_to_pyspec_map = {
+        'functions': 'functions',
+        'fn': 'functions',
+        'constants': 'constant_vars',
+        'constant_variables': 'constant_vars',
+        'constant_var': 'constant_vars',
+        'configs': 'config_vars',
+        'config_variables': 'config_vars',
+        'config_var': 'config_vars',
+        'presets': 'preset_vars',
+        'preset_variables': 'preset_vars',
+        'preset_var': 'preset_vars',
+        'ssz_objects': 'ssz_objects',
+        'containers': 'ssz_objects',
+        'container': 'ssz_objects',
+        'dataclasses': 'dataclasses',
+        'dataclass': 'dataclasses',
+        'custom_types': 'custom_types',
+        'custom_type': 'custom_types'
+    }
+
+    errors = []
+
+    for exception_key, exception_items in exceptions.items():
+        # Get the corresponding pyspec key
+        pyspec_key = exception_to_pyspec_map.get(exception_key)
+        if not pyspec_key:
+            errors.append(f"Unknown exception type: '{exception_key}'")
+            continue
+
+        # Ensure exception_items is a list
+        if not isinstance(exception_items, list):
+            exception_items = [exception_items]
+
+        for item in exception_items:
+            # Parse item#fork format
+            if '#' in item:
+                item_name, fork = item.split('#', 1)
+            else:
+                # If no fork specified, we'll check if it exists in any fork
+                item_name = item
+                fork = None
+
+            # Check if the item exists
+            item_found = False
+            if fork:
+                # Check specific fork
+                if ('mainnet' in pyspec and
+                    fork in pyspec['mainnet'] and
+                    pyspec_key in pyspec['mainnet'][fork] and
+                    item_name in pyspec['mainnet'][fork][pyspec_key]):
+                    item_found = True
+            else:
+                # Check if item exists in any fork
+                if 'mainnet' in pyspec:
+                    for check_fork in pyspec['mainnet']:
+                        if (pyspec_key in pyspec['mainnet'][check_fork] and
+                            item_name in pyspec['mainnet'][check_fork][pyspec_key]):
+                            item_found = True
+                            break
+
+            if not item_found:
+                fork_suffix = f"#{fork}" if fork else ""
+                errors.append(f"invalid key: {exception_key}.{item_name}{fork_suffix}")
+
+    if errors:
+        error_msg = "Invalid exception items in configuration:\n" + "\n".join(f"  - {e}" for e in errors)
+        raise Exception(error_msg)
+
 def load_config(directory=None):
     """
     Load configuration from .ethspecify.yml file in the specified directory.
@@ -25,7 +110,22 @@ def load_config(directory=None):
         try:
             with open(config_path, 'r') as f:
                 config = yaml.safe_load(f)
-                return config if config else {}
+                if not config:
+                    return {}
+
+                # Get version from config, default to 'nightly'
+                version = config.get('version', 'nightly')
+
+                # Validate exceptions in root config
+                if 'exceptions' in config:
+                    validate_exception_items(config['exceptions'], version)
+
+                # Also validate exceptions in specrefs section if present
+                if 'specrefs' in config and isinstance(config['specrefs'], dict):
+                    if 'exceptions' in config['specrefs']:
+                        validate_exception_items(config['specrefs']['exceptions'], version)
+
+                return config
         except (yaml.YAMLError, IOError) as e:
             print(f"Warning: Error reading .ethspecify.yml file: {e}")
             return {}
@@ -232,10 +332,16 @@ def get_spec(attributes, preset, fork, version="nightly"):
             + " = "
             + pyspec[preset][fork]["custom_types"][attributes["custom_type"]]
         )
-    elif "ssz_object" in attributes:
+    elif "ssz_object" in attributes or "container" in attributes:
         if spec is not None:
             raise Exception(f"Tag can only specify one spec item")
-        spec = pyspec[preset][fork]["ssz_objects"][attributes["ssz_object"]]
+        if "ssz_object" in attributes and "container" in attributes:
+            raise Exception(f"cannot contain 'ssz_object' and 'container'")
+        if "ssz_object" in attributes:
+            object_name = attributes["ssz_object"]
+        else:
+            object_name = attributes["container"]
+        spec = pyspec[preset][fork]["ssz_objects"][object_name]
     elif "dataclass" in attributes:
         if spec is not None:
             raise Exception(f"Tag can only specify one spec item")
@@ -471,6 +577,140 @@ def extract_attributes(tag):
     return dict(attr_pattern.findall(tag))
 
 
+def sort_specref_yaml(yaml_file):
+    """
+    Sort specref entries in a YAML file by their name field.
+    Preserves formatting and adds single blank lines between entries.
+    """
+    if not os.path.exists(yaml_file):
+        return False
+
+    try:
+        with open(yaml_file, 'r') as f:
+            content_str = f.read()
+
+        # Try to fix common YAML issues with unquoted search strings containing colons
+        # This is the same fix used in check_source_files
+        content_str = re.sub(r'(\s+search:\s+)([^"\n]+:)(\s*$)', r'\1"\2"\3', content_str, flags=re.MULTILINE)
+
+        try:
+            content = yaml.safe_load(content_str)
+        except yaml.YAMLError:
+            # Fall back to FullLoader if safe_load fails
+            content = yaml.load(content_str, Loader=yaml.FullLoader)
+
+        if not content:
+            return False
+
+        # Handle both array of objects and single object formats
+        if isinstance(content, list):
+            # Sort the list by 'name' field if it exists
+            # Special handling for fork ordering within the same item
+            def sort_key(x):
+                name = x.get('name', '') if isinstance(x, dict) else str(x)
+
+                # Known fork names for ordering
+                forks = ['phase0', 'altair', 'bellatrix', 'capella', 'deneb', 'electra']
+
+                # Check if name contains a # separator (like "slash_validator#phase0")
+                if '#' in name:
+                    base_name, fork = name.rsplit('#', 1)
+                    # Define fork order based on known forks list
+                    fork_lower = fork.lower()
+                    if fork_lower in forks:
+                        fork_order = forks.index(fork_lower)
+                    else:
+                        # Unknown forks go after known ones, sorted alphabetically
+                        fork_order = len(forks)
+                    return (base_name, fork_order, fork)
+                else:
+                    # Check if name ends with a fork name (like "BeaconStatePhase0")
+                    name_lower = name.lower()
+                    for i, fork in enumerate(forks):
+                        if name_lower.endswith(fork):
+                            # Extract base name
+                            base_name = name[:-len(fork)]
+                            return (base_name, i, name)
+
+                    # No fork pattern found, just sort by name
+                    return (name, 0, '')
+
+            sorted_content = sorted(content, key=sort_key)
+
+            # Custom YAML writing to preserve formatting
+            output_lines = []
+            for i, item in enumerate(sorted_content):
+                if i > 0:
+                    # Add a single blank line between entries
+                    output_lines.append("")
+
+                # Start each entry with a dash
+                first_line = True
+                for key, value in item.items():
+                    if first_line:
+                        prefix = "- "
+                        first_line = False
+                    else:
+                        prefix = "  "
+
+                    if key == 'spec':
+                        # Preserve spec content as-is, using literal style
+                        output_lines.append(f"{prefix}{key}: |")
+                        # Indent the spec content - preserve it exactly as-is
+                        spec_lines = value.rstrip().split('\n') if isinstance(value, str) else str(value).rstrip().split('\n')
+                        for spec_line in spec_lines:
+                            output_lines.append(f"    {spec_line}")
+                    elif key == 'sources':
+                        if isinstance(value, list) and len(value) == 0:
+                            # Keep empty lists on the same line for clarity
+                            output_lines.append(f"{prefix}{key}: []")
+                        else:
+                            output_lines.append(f"{prefix}{key}:")
+                            if isinstance(value, list):
+                                for source in value:
+                                    if isinstance(source, dict):
+                                        output_lines.append(f"    - file: {source.get('file', '')}")
+                                        if 'search' in source:
+                                            search_val = source['search']
+                                            # Only quote if:
+                                            # 1. Contains a colon followed by space or end of string (YAML mapping issue)
+                                            # 2. Not already quoted
+                                            # 3. Doesn't contain internal quotes (like regex patterns)
+                                            if ((':' in search_val or search_val.endswith(':')) and
+                                                not (search_val.startswith('"') and search_val.endswith('"')) and
+                                                '"' not in search_val):
+                                                # Only quote simple strings with colons, not complex patterns
+                                                if not any(char in search_val for char in ['\\', '*', '|', '[', ']', '(', ')']):
+                                                    search_val = f'"{search_val}"'
+                                            output_lines.append(f"      search: {search_val}")
+                                        if 'regex' in source:
+                                            # Keep boolean values lowercase for consistency
+                                            regex_val = str(source['regex']).lower()
+                                            output_lines.append(f"      regex: {regex_val}")
+                                    else:
+                                        output_lines.append(f"    - {source}")
+                    else:
+                        # Handle other fields - don't escape or modify the value
+                        output_lines.append(f"{prefix}{key}: {value}")
+
+            # Strip trailing whitespace from all lines
+            output_lines = [line.rstrip() for line in output_lines]
+
+            # Write back the sorted content
+            with open(yaml_file, 'w') as f:
+                f.write('\n'.join(output_lines))
+                f.write('\n')  # End file with newline
+
+            return True
+        elif isinstance(content, dict):
+            # If it's a single dict, we can't sort it
+            return False
+    except (yaml.YAMLError, IOError) as e:
+        print(f"Error sorting {yaml_file}: {e}")
+        return False
+
+    return False
+
 def replace_spec_tags(file_path, config=None):
     with open(file_path, 'r') as file:
         content = file.read()
@@ -485,6 +725,9 @@ def replace_spec_tags(file_path, config=None):
         re.DOTALL
     )
 
+    # Collect processed spec items for potential YAML updates
+    processed_items = []
+
     def rebuild_opening_tag(attributes, hash_value):
         # Rebuild a fresh opening tag from attributes, overriding any existing hash.
         new_opening = "<spec"
@@ -523,6 +766,17 @@ def replace_spec_tags(file_path, config=None):
         spec = get_spec(attributes, preset, fork, version)
         hash_value = hashlib.sha256(spec.encode('utf-8')).hexdigest()[:8]
 
+        # Collect this item for potential YAML updates
+        processed_items.append({
+            'attributes': attributes,
+            'preset': preset,
+            'fork': fork,
+            'style': style,
+            'version': version,
+            'spec': spec,
+            'hash': hash_value
+        })
+
         if style == "hash":
             # Rebuild a fresh self-closing tag.
             updated_tag = rebuild_self_closing_tag(attributes, hash_value)
@@ -534,7 +788,7 @@ def replace_spec_tags(file_path, config=None):
             prefix = content[:match.start()].splitlines()[-1]
             prefixed_spec = "\n".join(
                 f"{prefix}{line}" if line.rstrip() else prefix.rstrip()
-                for line in spec_content.rstrip().split("\n")
+                for line in spec_content.split("\n")
             )
             updated_tag = f"{new_opening}\n{prefixed_spec}\n{prefix}</spec>"
             return updated_tag
@@ -547,6 +801,151 @@ def replace_spec_tags(file_path, config=None):
     with open(file_path, 'w') as file:
         file.write(updated_content)
 
+    # Return processed items for potential YAML updates
+    return processed_items
+
+
+def get_yaml_filename_for_spec_attr(spec_attr):
+    """Map spec attribute to YAML filename."""
+    attr_to_file = {
+        'fn': 'functions.yml',
+        'function': 'functions.yml',
+        'constant_var': 'constants.yml',
+        'config_var': 'configs.yml',
+        'preset_var': 'presets.yml',
+        'container': 'containers.yml',
+        'ssz_object': 'containers.yml',
+        'dataclass': 'dataclasses.yml',
+        'custom_type': 'types.yml',
+    }
+    return attr_to_file.get(spec_attr)
+
+
+def get_spec_attr_and_name(attributes):
+    """Extract the spec attribute and item name from tag attributes."""
+    spec_attrs = ['fn', 'function', 'constant_var', 'config_var', 'preset_var',
+                  'container', 'ssz_object', 'dataclass', 'custom_type']
+    for attr in spec_attrs:
+        if attr in attributes:
+            return attr, attributes[attr]
+    return None, None
+
+
+def load_yaml_entries(yaml_file):
+    """Load existing entries from a YAML file."""
+    if not os.path.exists(yaml_file):
+        return []
+
+    try:
+        with open(yaml_file, 'r') as f:
+            content_str = f.read()
+
+        # Try to fix common YAML issues with unquoted search strings containing colons
+        content_str = re.sub(r'(\s+search:\s+)([^"\n]+:)(\s*$)', r'\1"\2"\3', content_str, flags=re.MULTILINE)
+
+        try:
+            content = yaml.safe_load(content_str)
+        except yaml.YAMLError:
+            content = yaml.load(content_str, Loader=yaml.FullLoader)
+
+        if isinstance(content, list):
+            return content
+        return []
+    except (yaml.YAMLError, IOError):
+        return []
+
+
+def extract_spec_tag_key(spec_content):
+    """Extract a unique key from spec tag to identify duplicates."""
+    if not spec_content:
+        return None
+
+    # Extract the opening spec tag
+    match = re.search(r'<spec\b([^>]*)>', spec_content)
+    if not match:
+        return None
+
+    # Extract attributes from the tag
+    attributes = extract_attributes(match.group(0))
+
+    # Build a key from the spec attribute and fork
+    # e.g., "constant_var:DOMAIN_PTC_ATTESTER:fork:gloas"
+    key_parts = []
+    for attr in ['fn', 'function', 'constant_var', 'config_var', 'preset_var',
+                 'container', 'ssz_object', 'dataclass', 'custom_type']:
+        if attr in attributes:
+            key_parts.append(f"{attr}:{attributes[attr]}")
+            break
+
+    if 'fork' in attributes:
+        key_parts.append(f"fork:{attributes['fork']}")
+
+    return ':'.join(key_parts) if key_parts else None
+
+
+def add_missing_entries_to_yaml(yaml_file, new_entries):
+    """Add new entries to a YAML file and sort it."""
+    if not new_entries:
+        return
+
+    # Load existing entries
+    existing_entries = load_yaml_entries(yaml_file)
+
+    # Build a set of existing spec tag keys
+    existing_spec_keys = set()
+    for entry in existing_entries:
+        if isinstance(entry, dict) and 'spec' in entry:
+            spec_key = extract_spec_tag_key(entry['spec'])
+            if spec_key:
+                existing_spec_keys.add(spec_key)
+
+    # Filter out entries that already exist (based on spec tag, not name)
+    entries_to_add = []
+    for entry in new_entries:
+        spec_key = extract_spec_tag_key(entry.get('spec', ''))
+        if spec_key and spec_key not in existing_spec_keys:
+            entries_to_add.append(entry)
+            existing_spec_keys.add(spec_key)  # Avoid duplicates within new entries
+
+    if not entries_to_add:
+        return
+
+    # Combine and write
+    all_entries = existing_entries + entries_to_add
+
+    # Ensure directory exists
+    os.makedirs(os.path.dirname(yaml_file) if os.path.dirname(yaml_file) else '.', exist_ok=True)
+
+    # Write combined entries using the same format as generate_specref_files
+    with open(yaml_file, 'w') as f:
+        for i, entry in enumerate(all_entries):
+            if i > 0:
+                f.write('\n')
+            f.write(f'- name: {entry["name"]}\n')
+            if 'sources' in entry:
+                if isinstance(entry['sources'], list) and len(entry['sources']) == 0:
+                    f.write('  sources: []\n')
+                else:
+                    f.write('  sources:\n')
+                    for source in entry['sources']:
+                        if isinstance(source, dict):
+                            f.write(f'    - file: {source.get("file", "")}\n')
+                            if 'search' in source:
+                                f.write(f'      search: {source["search"]}\n')
+                            if 'regex' in source:
+                                f.write(f'      regex: {source["regex"]}\n')
+                        else:
+                            f.write(f'    - {source}\n')
+            if 'spec' in entry:
+                f.write('  spec: |\n')
+                for line in entry['spec'].split('\n'):
+                    f.write(f'    {line}\n')
+
+    # Sort the file
+    sort_specref_yaml(yaml_file)
+
+    print(f"Added {len(entries_to_add)} new entries to {yaml_file}")
+
 
 def check_source_files(yaml_file, project_root, exceptions=None):
     """
@@ -614,6 +1013,7 @@ def check_source_files(yaml_file, project_root, exceptions=None):
                                 'config_var': 'configs',
                                 'preset_var': 'presets',
                                 'ssz_object': 'ssz_objects',
+                                'container': 'ssz_objects',
                                 'dataclass': 'dataclasses',
                                 'custom_type': 'custom_types'
                             }
@@ -625,19 +1025,23 @@ def check_source_files(yaml_file, project_root, exceptions=None):
         if not spec_ref and 'name' in item:
             spec_ref = item['name']
 
+        # Extract item name and fork for exception checking
+        item_name_for_exception = None
+        fork_for_exception = None
+        if spec_ref and '#' in spec_ref and '.' in spec_ref:
+            # Format: "functions.item_name#fork"
+            _, item_with_fork = spec_ref.split('.', 1)
+            if '#' in item_with_fork:
+                item_name_for_exception, fork_for_exception = item_with_fork.split('#', 1)
+
         # Check if sources list is empty
         if not item['sources']:
             if spec_ref:
-                # Extract item name and fork from spec_ref for exception checking
-                if '#' in spec_ref and '.' in spec_ref:
-                    # Format: "functions.item_name#fork"
-                    _, item_with_fork = spec_ref.split('.', 1)
-                    if '#' in item_with_fork:
-                        item_name, fork = item_with_fork.split('#', 1)
-                        # Check if this item is in exceptions
-                        if is_excepted(item_name, fork, exceptions):
-                            total_count += 1
-                            continue
+                # Check if this item is in exceptions
+                if item_name_for_exception and fork_for_exception:
+                    if is_excepted(item_name_for_exception, fork_for_exception, exceptions):
+                        total_count += 1
+                        continue
 
                 errors.append(f"EMPTY SOURCES: {spec_ref}")
             else:
@@ -647,6 +1051,13 @@ def check_source_files(yaml_file, project_root, exceptions=None):
             total_count += 1
             continue
 
+        # Check if item has non-empty sources but is in exceptions
+        if item_name_for_exception and fork_for_exception:
+            if is_excepted(item_name_for_exception, fork_for_exception, exceptions):
+                errors.append(f"EXCEPTION CONFLICT: {spec_ref} has a specref")
+                total_count += 1
+                continue
+
         for source in item['sources']:
             # All sources now use the standardized dict format with file and optional search
             if not isinstance(source, dict) or 'file' not in source:
@@ -758,7 +1169,7 @@ def extract_spec_tags_from_yaml(yaml_file, tag_type=None):
 
     # Known tag type attributes
     tag_attributes = ['fn', 'function', 'constant_var', 'config_var', 'preset_var',
-                      'ssz_object', 'dataclass', 'custom_type']
+                      'ssz_object', 'container', 'dataclass', 'custom_type']
 
     try:
         with open(yaml_file, 'r') as f:
@@ -807,6 +1218,9 @@ def extract_spec_tags_from_yaml(yaml_file, tag_type=None):
                         # Normalize function to fn
                         if found_tag_type == 'function':
                             found_tag_type = 'fn'
+                        # Normalize container to ssz_object
+                        if found_tag_type == 'container':
+                            found_tag_type = 'ssz_object'
                         break
 
                 if found_tag_type and 'fork' in attrs:
@@ -822,6 +1236,209 @@ def extract_spec_tags_from_yaml(yaml_file, tag_type=None):
     return tag_types_found, pairs
 
 
+def generate_specrefs_from_files(files_with_spec_tags, project_dir):
+    """
+    Generate specrefs data from files containing spec tags.
+    Returns a dict with spec tag info and their source locations.
+    """
+    specrefs = {}
+
+    for file_path in files_with_spec_tags:
+        try:
+            with open(file_path, 'r', encoding='utf-8') as f:
+                content = f.read()
+
+            # Find all spec tags in the file
+            spec_tag_pattern = r'<spec\s+([^>]+?)(?:\s*/>|>)'
+            matches = re.finditer(spec_tag_pattern, content)
+
+            for match in matches:
+                tag_attrs_str = match.group(1)
+                attrs = extract_attributes(f"<spec {tag_attrs_str}>")
+
+                # Determine the spec type and name
+                spec_type = None
+                spec_name = None
+                fork = attrs.get('fork', None)
+
+                # Check each possible spec attribute
+                for attr_name in ['fn', 'function', 'constant_var', 'config_var',
+                                  'preset_var', 'ssz_object', 'container', 'dataclass', 'custom_type']:
+                    if attr_name in attrs:
+                        spec_type = attr_name
+                        spec_name = attrs[attr_name]
+                        break
+
+                if spec_type and spec_name:
+                    # Normalize container to ssz_object for consistency
+                    if spec_type == 'container':
+                        spec_type = 'ssz_object'
+                    # Create a unique key for this spec reference
+                    key = f"{spec_type}.{spec_name}"
+                    if fork:
+                        key += f"#{fork}"
+
+                    if key not in specrefs:
+                        specrefs[key] = {
+                            'name': spec_name,
+                            'type': spec_type,
+                            'fork': fork,
+                            'sources': []
+                        }
+
+                    # Add this source location
+                    rel_path = os.path.relpath(file_path, project_dir)
+
+                    # Get line number of the match
+                    lines_before = content[:match.start()].count('\n')
+                    line_num = lines_before + 1
+
+                    specrefs[key]['sources'].append({
+                        'file': rel_path,
+                        'line': line_num
+                    })
+
+        except (IOError, UnicodeDecodeError):
+            continue
+
+    return specrefs
+
+
+def process_generated_specrefs(specrefs, exceptions, version):
+    """
+    Process the generated specrefs and check coverage.
+    Returns (success, results)
+    """
+    results = {}
+    overall_success = True
+
+    # Group specrefs by type for coverage checking
+    specrefs_by_type = {}
+    for _, data in specrefs.items():
+        spec_type = data['type']
+        if spec_type not in specrefs_by_type:
+            specrefs_by_type[spec_type] = []
+        specrefs_by_type[spec_type].append(data)
+
+    # Map spec types to history keys
+    type_to_history_key = {
+        'fn': 'functions',
+        'function': 'functions',
+        'constant_var': 'constant_vars',
+        'config_var': 'config_vars',
+        'preset_var': 'preset_vars',
+        'ssz_object': 'ssz_objects',
+        'container': 'ssz_objects',
+        'dataclass': 'dataclasses',
+        'custom_type': 'custom_types'
+    }
+
+    # Map to exception keys
+    type_to_exception_key = {
+        'fn': 'functions',
+        'function': 'functions',
+        'constant_var': 'constants',
+        'config_var': 'configs',
+        'preset_var': 'presets',
+        'ssz_object': 'ssz_objects',
+        'container': 'ssz_objects',
+        'dataclass': 'dataclasses',
+        'custom_type': 'custom_types'
+    }
+
+    # Get spec history for coverage checking
+    history = get_spec_item_history("mainnet", version)
+
+    # Check coverage for each type
+    total_found = 0
+    total_expected = 0
+    all_missing = []
+
+    for spec_type, items in specrefs_by_type.items():
+        history_key = type_to_history_key.get(spec_type, spec_type)
+        exception_key = type_to_exception_key.get(spec_type, spec_type)
+
+        # Get exceptions for this type - handle both singular and plural keys
+        type_exceptions = []
+        if exception_key in exceptions:
+            type_exceptions = exceptions[exception_key]
+        # Also check plural forms
+        elif exception_key + 's' in exceptions:
+            type_exceptions = exceptions[exception_key + 's']
+        # Check if singular form exists when we have plural
+        elif exception_key.endswith('s') and exception_key[:-1] in exceptions:
+            type_exceptions = exceptions[exception_key[:-1]]
+
+        # Special handling for ssz_objects/containers
+        if spec_type in ['ssz_object', 'container'] and not type_exceptions:
+            # Check for 'containers' as an alternative key
+            if 'containers' in exceptions:
+                type_exceptions = exceptions['containers']
+            elif 'container' in exceptions:
+                type_exceptions = exceptions['container']
+
+        # Build set of what we found
+        found_items = set()
+        for item in items:
+            if item['fork']:
+                found_items.add(f"{item['name']}#{item['fork']}")
+            else:
+                # If no fork specified, we need to check all forks
+                if history_key in history and item['name'] in history[history_key]:
+                    for fork in history[history_key][item['name']]:
+                        found_items.add(f"{item['name']}#{fork}")
+
+        # Check what's expected
+        if history_key in history:
+            for item_name, forks in history[history_key].items():
+                for fork in forks:
+                    expected_key = f"{item_name}#{fork}"
+                    total_expected += 1
+
+                    # Check if excepted
+                    if is_excepted(item_name, fork, type_exceptions):
+                        total_found += 1
+                        continue
+
+                    if expected_key in found_items:
+                        total_found += 1
+                    else:
+                        # Use the proper type prefix for the missing item
+                        type_prefix_map = {
+                            'functions': 'functions',
+                            'constant_vars': 'constants',
+                            'config_vars': 'configs',
+                            'preset_vars': 'presets',
+                            'ssz_objects': 'ssz_objects',
+                            'dataclasses': 'dataclasses',
+                            'custom_types': 'custom_types'
+                        }
+                        prefix = type_prefix_map.get(history_key, history_key)
+                        all_missing.append(f"{prefix}.{expected_key}")
+
+    # Count total spec references found
+    total_refs = len(specrefs)
+
+    # Store results
+    results['Project Coverage'] = {
+        'source_files': {
+            'valid': total_refs,
+            'total': total_refs,
+            'errors': []
+        },
+        'coverage': {
+            'found': total_found,
+            'expected': total_expected,
+            'missing': all_missing
+        }
+    }
+
+    if all_missing:
+        overall_success = False
+
+    return overall_success, results
+
+
 def check_coverage(yaml_file, tag_type, exceptions, preset="mainnet", version="nightly"):
     """
     Check that all spec items from ethspecify have corresponding tags in the YAML file.
@@ -830,6 +1447,7 @@ def check_coverage(yaml_file, tag_type, exceptions, preset="mainnet", version="n
     # Map tag types to history keys
     history_key_map = {
         'ssz_object': 'ssz_objects',
+        'container': 'ssz_objects',
         'config_var': 'config_vars',
         'preset_var': 'preset_vars',
         'dataclass': 'dataclasses',
@@ -890,16 +1508,53 @@ def run_checks(project_dir, config):
     else:
         # New format: specrefs: { files: [...], exceptions: {...} }
         specrefs_files = specrefs_config.get('files', [])
-        exceptions = specrefs_config.get('exceptions', {})
 
+        # Support exceptions in either specrefs section or root, but not both
+        specrefs_exceptions = specrefs_config.get('exceptions', {})
+        root_exceptions = config.get('exceptions', {})
+
+        if specrefs_exceptions and root_exceptions:
+            print("Warning: Exceptions found in both root and specrefs sections. Using specrefs exceptions.")
+            exceptions = specrefs_exceptions
+        elif specrefs_exceptions:
+            exceptions = specrefs_exceptions
+        else:
+            exceptions = root_exceptions
+
+    # If no files specified, search the whole project for spec tags
     if not specrefs_files:
-        print("Error: No specrefs files specified in .ethspecify.yml")
-        print("Please add a 'specrefs:' section with 'files:' listing the files to check")
-        return False, {}
+        print("No specific files configured, searching entire project for spec tags...")
+
+        # Determine search root - configurable in specrefs section
+        if 'search_root' in specrefs_config:
+            # Use configured search_root (relative to project_dir)
+            search_root_rel = specrefs_config['search_root']
+            search_root = os.path.join(project_dir, search_root_rel) if not os.path.isabs(search_root_rel) else search_root_rel
+            search_root = os.path.abspath(search_root)
+        else:
+            # Default behavior: if we're in a specrefs directory, search in the parent directory
+            search_root = os.path.dirname(project_dir) if os.path.basename(project_dir) == 'specrefs' else project_dir
+
+        print(f"Searching for spec tags in: {search_root}")
+
+        # Use grep to find all files containing spec tags
+        files_with_spec_tags = grep(search_root, r'<spec\b[^>]*>', [])
+
+        if not files_with_spec_tags:
+            print(f"No files with spec tags found in the project")
+            return True, {}
+
+        # Generate in-memory specrefs data from the found spec tags
+        all_specrefs = generate_specrefs_from_files(files_with_spec_tags, search_root)
+
+        # Process the generated specrefs
+        return process_generated_specrefs(all_specrefs, exceptions, version)
+
 
     # Map tag types to exception keys (support both singular and plural)
     exception_key_map = {
-        'ssz_object': ['ssz_objects', 'ssz_object'],
+        'ssz_object': ['ssz_objects', 'ssz_object', 'containers', 'container'],
+        'container': ['ssz_objects', 'ssz_object', 'containers', 'container'],
         'config_var': ['configs', 'config_variables', 'config_var'],
         'preset_var': ['presets', 'preset_variables', 'preset_var'],
         'dataclass': ['dataclasses', 'dataclass'],
@@ -928,7 +1583,16 @@ def run_checks(project_dir, config):
         # Process each tag type found in the file
         if not tag_types_found:
             # No spec tags found, still check source files
-            valid_count, total_count, source_errors = check_source_files(yaml_path, os.path.dirname(project_dir), [])
+            # Determine source root - use search_root if configured, otherwise use default behavior
+            if 'search_root' in specrefs_config:
+                search_root_rel = specrefs_config['search_root']
+                source_root = os.path.join(project_dir, search_root_rel) if not os.path.isabs(search_root_rel) else search_root_rel
+                source_root = os.path.abspath(source_root)
+            else:
+                # Default behavior: parent directory
+                source_root = os.path.dirname(project_dir)
+
+            valid_count, total_count, source_errors = check_source_files(yaml_path, source_root, [])
 
             # Store results using filename as section name
             section_name = filename.replace('.yml', '').replace('-', ' ').title()
@@ -980,7 +1644,16 @@ def run_checks(project_dir, config):
                         if key in exceptions:
                             all_exceptions.extend(exceptions[key])
 
-            valid_count, total_count, source_errors = check_source_files(yaml_path, os.path.dirname(project_dir), all_exceptions)
+            # Determine source root - use search_root if configured, otherwise use default behavior
+            if 'search_root' in specrefs_config:
+                search_root_rel = specrefs_config['search_root']
+                source_root = os.path.join(project_dir, search_root_rel) if not os.path.isabs(search_root_rel) else search_root_rel
+                source_root = os.path.abspath(source_root)
+            else:
+                # Default behavior: parent directory
+                source_root = os.path.dirname(project_dir)
+
+            valid_count, total_count, source_errors = check_source_files(yaml_path, source_root, all_exceptions)
 
             # Store results using filename as section name
             section_name = filename.replace('.yml', '').replace('-', ' ').title()
@@ -1005,3 +1678,375 @@ def run_checks(project_dir, config):
                 overall_success = False
 
     return overall_success, results
+
+
+def update_entry_names_in_yaml_files(project_dir, specrefs_files):
+    """
+    Update all entry names to use the format <spec_item>#<fork>.
+    """
+    for yaml_file in specrefs_files:
+        yaml_path = os.path.join(project_dir, yaml_file)
+
+        if not os.path.exists(yaml_path):
+            continue
+
+        # Load existing entries
+        existing_entries = load_yaml_entries(yaml_path)
+        if not existing_entries:
+            continue
+
+        updated = False
+        for entry in existing_entries:
+            if not isinstance(entry, dict) or 'spec' not in entry:
+                continue
+
+            # Extract spec tag attributes
+            spec_content = entry['spec']
+            match = re.search(r'<spec\b([^>]*)>', spec_content)
+            if not match:
+                continue
+
+            attributes = extract_attributes(match.group(0))
+
+            # Get the spec item name and fork
+            spec_attr, item_name = get_spec_attr_and_name(attributes)
+            fork = attributes.get('fork')
+
+            if item_name and fork:
+                # Build the expected name
+                expected_name = f'{item_name}#{fork}'
+
+                # Update if different
+                if entry.get('name') != expected_name:
+                    entry['name'] = expected_name
+                    updated = True
+
+        # Write back if updated
+        if updated:
+            with open(yaml_path, 'w') as f:
+                for i, entry in enumerate(existing_entries):
+                    if i > 0:
+                        f.write('\n')
+                    f.write(f'- name: {entry["name"]}\n')
+                    if 'sources' in entry:
+                        if isinstance(entry['sources'], list) and len(entry['sources']) == 0:
+                            f.write('  sources: []\n')
+                        else:
+                            f.write('  sources:\n')
+                            for source in entry['sources']:
+                                if isinstance(source, dict):
+                                    f.write(f'    - file: {source.get("file", "")}\n')
+                                    if 'search' in source:
+                                        f.write(f'      search: {source["search"]}\n')
+                                    if 'regex' in source:
+                                        f.write(f'      regex: {source["regex"]}\n')
+                                else:
+                                    f.write(f'    - {source}\n')
+                    if 'spec' in entry:
+                        f.write('  spec: |\n')
+                        for line in entry['spec'].split('\n'):
+                            f.write(f'    {line}\n')
+
+            # Sort the file
+            sort_specref_yaml(yaml_path)
+            print(f"Updated entry names in {yaml_file}")
+
+
+def add_missing_spec_items_to_yaml_files(project_dir, config, specrefs_files):
+    """
+    Add missing spec items to existing YAML files.
+    Ensures all spec items from the specification exist in YAML files with sources: []
+    """
+    version = config.get('version', 'nightly')
+    preset = 'mainnet'  # Could make this configurable
+
+    # Get all spec items
+    pyspec = get_pyspec(version)
+    if preset not in pyspec:
+        print(f"Error: Preset '{preset}' not found")
+        return
+
+    # Get all forks in chronological order, excluding EIP forks
+    all_forks = sorted(
+        [fork for fork in pyspec[preset].keys() if not fork.startswith("eip")],
+        key=lambda x: (x != "phase0", x)
+    )
+
+    # Map YAML filenames to category keys and spec attribute names
+    filename_to_category = {
+        'constants.yml': ('constant_vars', 'constant_var'),
+        'configs.yml': ('config_vars', 'config_var'),
+        'presets.yml': ('preset_vars', 'preset_var'),
+        'functions.yml': ('functions', 'fn'),
+        'containers.yml': ('ssz_objects', 'container'),
+        'dataclasses.yml': ('dataclasses', 'dataclass'),
+        'types.yml': ('custom_types', 'custom_type'),
+    }
+
+    for yaml_file in specrefs_files:
+        yaml_path = os.path.join(project_dir, yaml_file)
+        yaml_basename = os.path.basename(yaml_file)
+
+        if yaml_basename not in filename_to_category:
+            continue
+
+        category, spec_attr = filename_to_category[yaml_basename]
+
+        # Collect all items in this category organized by name and fork
+        items_by_name = {}
+        for fork in all_forks:
+            if fork not in pyspec[preset]:
+                continue
+            fork_data = pyspec[preset][fork]
+
+            if category not in fork_data:
+                continue
+
+            for item_name, item_data in fork_data[category].items():
+                if item_name not in items_by_name:
+                    items_by_name[item_name] = []
+                items_by_name[item_name].append((fork, item_data))
+
+        # Build entries for missing items
+        new_entries = []
+        for item_name in sorted(items_by_name.keys()):
+            forks_data = items_by_name[item_name]
+
+            # Find all unique versions of this item (where content differs)
+            versions = []  # List of (fork, item_data, spec_content)
+            prev_content = None
+
+            for fork, item_data in forks_data:
+                # Build the spec content based on category
+                if category == 'functions':
+                    spec_content = item_data
+                elif category in ['constant_vars', 'config_vars', 'preset_vars']:
+                    # item_data is a list: [type, value, ...]
+                    if isinstance(item_data, (list, tuple)) and len(item_data) >= 2:
+                        type_info = item_data[0]
+                        value = item_data[1]
+                        if type_info:
+                            spec_content = f"{item_name}: {type_info} = {value}"
+                        else:
+                            spec_content = f"{item_name} = {value}"
+                    else:
+                        spec_content = str(item_data)
+                elif category == 'ssz_objects':
+                    spec_content = item_data
+                elif category == 'dataclasses':
+                    spec_content = item_data.replace("@dataclass\n", "")
+                elif category == 'custom_types':
+                    # custom_types are simple type aliases: TypeName = SomeType
+                    spec_content = f"{item_name} = {item_data}"
+                else:
+                    spec_content = str(item_data)
+
+                # Only add this version if it's different from the previous one
+                if prev_content is None or spec_content != prev_content:
+                    versions.append((fork, item_data, spec_content))
+                    prev_content = spec_content
+
+            # Create entries based on number of unique versions
+            use_fork_suffix = len(versions) > 1
+
+            for idx, (fork, item_data, spec_content) in enumerate(versions):
+                # Calculate hash of current version
+                hash_value = hashlib.sha256(spec_content.encode('utf-8')).hexdigest()[:8]
+
+                # For multiple versions after the first, use diff style
+                if use_fork_suffix and idx > 0:
+                    # Get previous version for diff
+                    prev_fork, _, prev_spec_content = versions[idx - 1]
+
+                    # Generate diff
+                    diff_content = diff(prev_fork, strip_comments(prev_spec_content), fork, strip_comments(spec_content))
+
+                    # Build spec tag with style="diff"
+                    spec_tag = f'<spec {spec_attr}="{item_name}" fork="{fork}" style="diff" hash="{hash_value}">'
+                    content = diff_content
+                else:
+                    # Build spec tag without style="diff"
+                    spec_tag = f'<spec {spec_attr}="{item_name}" fork="{fork}" hash="{hash_value}">'
+                    content = spec_content
+
+                # Create entry
+                entry_name = f'{item_name}#{fork}' if use_fork_suffix else item_name
+                entry = {
+                    'name': entry_name,
+                    'sources': [],
+                    'spec': f'{spec_tag}\n{content}\n</spec>'
+                }
+                new_entries.append(entry)
+
+        # Add missing entries to the YAML file
+        if new_entries:
+            add_missing_entries_to_yaml(yaml_path, new_entries)
+
+
+def generate_specref_files(output_dir, version="nightly", preset="mainnet"):
+    """
+    Generate specref YAML files without sources for manual mapping.
+    Creates a basic directory structure with empty sources.
+    """
+    # Create output directory if it doesn't exist
+    os.makedirs(output_dir, exist_ok=True)
+
+    # Get all spec items
+    pyspec = get_pyspec(version)
+    if preset not in pyspec:
+        raise ValueError(f"Preset '{preset}' not found")
+
+    # Get all forks in chronological order, excluding EIP forks
+    all_forks = sorted(
+        [fork for fork in pyspec[preset].keys() if not fork.startswith("eip")],
+        key=lambda x: (x != "phase0", x)
+    )
+
+    # Map history keys to file names and spec attribute names
+    category_map = {
+        'constant_vars': ('constants.yml', 'constant_var'),
+        'config_vars': ('configs.yml', 'config_var'),
+        'preset_vars': ('presets.yml', 'preset_var'),
+        'functions': ('functions.yml', 'fn'),
+        'ssz_objects': ('containers.yml', 'container'),
+        'dataclasses': ('dataclasses.yml', 'dataclass'),
+        'custom_types': ('types.yml', 'custom_type'),
+    }
+
+    # Collect all items organized by category
+    items_by_category = {cat: {} for cat in category_map.keys()}
+
+    for fork in all_forks:
+        if fork not in pyspec[preset]:
+            continue
+        fork_data = pyspec[preset][fork]
+
+        for category in items_by_category.keys():
+            if category not in fork_data:
+                continue
+
+            for item_name, item_data in fork_data[category].items():
+                # Track which forks have this item
+                if item_name not in items_by_category[category]:
+                    items_by_category[category][item_name] = []
+                items_by_category[category][item_name].append((fork, item_data))
+
+    # Generate YAML files for each category
+    for category, (filename, spec_attr) in category_map.items():
+        if not items_by_category[category]:
+            continue
+
+        output_path = os.path.join(output_dir, filename)
+        entries = []
+
+        # Sort items alphabetically
+        for item_name in sorted(items_by_category[category].keys()):
+            forks_data = items_by_category[category][item_name]
+
+            # Find all unique versions of this item (where content differs)
+            versions = []  # List of (fork, item_data, spec_content)
+            prev_content = None
+
+            for fork, item_data in forks_data:
+                # Build the spec content based on category
+                if category == 'functions':
+                    spec_content = item_data
+                elif category in ['constant_vars', 'config_vars', 'preset_vars']:
+                    # item_data is a list: [type, value, ...]
+                    if isinstance(item_data, (list, tuple)) and len(item_data) >= 2:
+                        type_info = item_data[0]
+                        value = item_data[1]
+                        if type_info:
+                            spec_content = f"{item_name}: {type_info} = {value}"
+                        else:
+                            spec_content = f"{item_name} = {value}"
+                    else:
+                        spec_content = str(item_data)
+                elif category == 'ssz_objects':
+                    spec_content = item_data
+                elif category == 'dataclasses':
+                    spec_content = item_data.replace("@dataclass\n", "")
+                elif category == 'custom_types':
+                    # custom_types are simple type aliases: TypeName = SomeType
+                    spec_content = f"{item_name} = {item_data}"
+                else:
+                    spec_content = str(item_data)
+
+                # Only add this version if it's different from the previous one
+                if prev_content is None or spec_content != prev_content:
+                    versions.append((fork, item_data, spec_content))
+                    prev_content = spec_content
+
+            # Create entries based on number of unique versions
+            use_fork_suffix = len(versions) > 1
+
+            for idx, (fork, item_data, spec_content) in enumerate(versions):
+                # Calculate hash of current version
+                hash_value = hashlib.sha256(spec_content.encode('utf-8')).hexdigest()[:8]
+
+                # For multiple versions after the first, use diff style
+                if use_fork_suffix and idx > 0:
+                    # Get previous version for diff
+                    prev_fork, _, prev_spec_content = versions[idx - 1]
+
+                    # Generate diff
+                    diff_content = diff(prev_fork, strip_comments(prev_spec_content), fork, strip_comments(spec_content))
+
+                    # Build spec tag with style="diff"
+                    spec_tag = f'<spec {spec_attr}="{item_name}" fork="{fork}" style="diff" hash="{hash_value}">'
+                    content = diff_content
+                else:
+                    # Build spec tag without style="diff"
+                    spec_tag = f'<spec {spec_attr}="{item_name}" fork="{fork}" hash="{hash_value}">'
+                    content = spec_content
+
+                # Create entry
+                entry_name = f'{item_name}#{fork}' if use_fork_suffix else item_name
+                entry = {
+                    'name': entry_name,
+                    'sources': [],
+                    'spec': f'{spec_tag}\n{content}\n</spec>'
+                }
+                entries.append(entry)
+
+        # Write YAML file
+        if entries:
+            with open(output_path, 'w') as f:
+                for i, entry in enumerate(entries):
+                    if i > 0:
+                        f.write('\n')
+                    f.write(f'- name: {entry["name"]}\n')
+                    f.write('  sources: []\n')
+                    f.write('  spec: |\n')
+                    for line in entry['spec'].split('\n'):
+                        f.write(f'    {line}\n')
+
+    # Create .ethspecify.yml config file
+    config_path = os.path.join(output_dir, '.ethspecify.yml')
+    with open(config_path, 'w') as f:
+        f.write(f'version: {version}\n')
+        f.write('style: full\n')
+        f.write('\n')
+        f.write('specrefs:\n')
+        f.write('  files:\n')
+        for category, (filename, _) in category_map.items():
+            if items_by_category[category]:
+                f.write(f'    - {filename}\n')
+        f.write('\n')
+        f.write('  exceptions:\n')
+        f.write('    # Add any exceptions here\n')
+
+    # Strip trailing whitespace from all generated files
+    all_files = [config_path]
+    for category, (filename, _) in category_map.items():
+        if items_by_category[category]:
+            all_files.append(os.path.join(output_dir, filename))
+
+    for file_path in all_files:
+        with open(file_path, 'r') as f:
+            lines = f.readlines()
+        with open(file_path, 'w') as f:
+            for line in lines:
+                f.write(line.rstrip() + '\n')
+
+    return list(category_map.values())