1. Updating parser spec.

Author: asladeofgreen

jobs/set_parser_config/__main__.py

@@ -18,7 +18,7 @@
 _GENERATORS = {
     constants.IDENTIFIER_TYPE_DATASET: set_dataset_id,
     constants.IDENTIFIER_TYPE_DIRECTORY: set_directory_id,
-    constants.IDENTIFIER_TYPE_FILENAME: set_filename_id
+    constants.IDENTIFIER_TYPE_FILENAME: set_filename_id,
 }
 
 
@@ -27,8 +27,8 @@ def _main():
 
     """
     for a in pyessv.get_cached():
-        for s in [i for i in a if i.data]:
-            for identifier_type, template in s.data.items():
+        for scope in [i for i in a if i.data]:
+            for identifier_type, template in scope.data.items():
                 try:
                     identifier_type = _IDENTIFIER_TYPES[identifier_type]
                 except KeyError:
@@ -38,10 +38,10 @@ def _main():
                     generator = _GENERATORS[identifier_type]
                 except KeyError:
                     continue
-                else:
-                    cfg = generator.get_config(a, s, template)
-                    if cfg is not None:
-                        io_manager.write_scope_parser_config(s, identifier_type, cfg)
+
+                cfg = generator.get_config(scope, template)
+                if cfg is not None:
+                    io_manager.write_scope_parser_config(scope, identifier_type, cfg)
 
 
 _main()

jobs/set_parser_config/set_dataset_id.py

@@ -1,37 +1,56 @@
-import typing
+import re
 
-from pyessv import Authority
 from pyessv import Scope
 
 
-def get_config(a: Authority, s: Scope, template_raw: str):
+def get_config(s: Scope, template_raw: str) -> dict:
     """Returns dataset identifier parser configuration information derived
        from a previously declared parsing template.
 
-    :param a: A vocabulary authority.
     :param s: A vocabulary scope.
     :param template: A raw dataset id parsing template.
     :returns: Dataset id parser configuration information.
 
     """
-    # Set specs, i.e. set of parser specification embedded in the template.
-    specs: typing.List[str] = template_raw.split(".")[1:]
-    specs = [i[2:-2] for i in specs]
-    specs = [i.replace("_", "-") for i in specs]
+    parts = [i.replace("_", "-") for i in re.findall("%\((\w*)\)s", template_raw)]
+    if s.namespace == "wcrp:input4mips":
+        parts = parts[1:]
+
+    return {
+        "seperator": ".",
+        "template": template_raw,
+        "specs": [_get_prefix_spec(s)] + [_get_part_spec(s, i) for i in parts],
+        "suffix": "#"
+    }
 
-    # Set spec overrides.
-    if s.namespace == "wcrp:cmip6":
-        specs[0] = "activity-id"
 
-    # Set template prefix.
+def _get_prefix_spec(s: Scope) -> dict:
+    """Maps a scope to a constant specifiction acting as identifier prefix.
+    
+    """
     if s.namespace in ("ecmwf:cc4e", "wcrp:cmip6"):
         prefix = s.canonical_name.upper()
+    elif s.namespace in ("wcrp:input4mips"):
+        prefix = s["activity-id"]["input4mips"].raw_name
     else:
         prefix = s.canonical_name
 
     return {
-        "template": template_raw,
-        "seperator": ".",
-        "specs": [f"const:{prefix}"] + [f"{s}:{i}" for i in specs],
-        "suffix": "#"
+        "type": "const",
+        "value": prefix,
+        "is_required": True
+    }
+
+
+def _get_part_spec(s: Scope, part: str) -> dict:
+    """Maps a template part to a collection specifiction.
+    
+    """
+    if s.namespace == "wcrp:cmip6" and part == "activity-drs":
+        part = "activity-id"
+
+    return {
+        "type": "collection",
+        "namespace": f"{s.namespace}:{part}",
+        "is_required": True
     }

jobs/set_parser_config/set_directory_id.py

@@ -1,6 +1,5 @@
-import typing
+import re
 
-from pyessv import Authority
 from pyessv import Scope
 
 
@@ -15,11 +14,10 @@
 }
 
 
-def get_config(a: Authority, s: Scope, template_raw: str):
+def get_config(s: Scope, template_raw: str) -> dict:
     """Returns directory identifier parser configuration information derived
        from a previously declared parsing template.
 
-    :param a: A vocabulary authority.
     :param s: A vocabulary scope.
     :param template_raw: A raw dataset id parsing template.
     :returns: Directory name parser configuration information.
@@ -29,23 +27,41 @@ def get_config(a: Authority, s: Scope, template_raw: str):
     if s.namespace == "wcrp:e3sm":
         return
 
-    # Set specs.
-    specs: typing.List[str] = template_raw.split("%(")[2:]
-    specs = [i.split(")")[0] for i in specs]
-    specs = [i.replace("_", "-") for i in specs]
+    parts = [i.replace("_", "-") for i in re.findall("%\((\w*)\)s", template_raw)[2:]]
 
-    # Set spec overrides.
-    if s.namespace == "wcrp:cmip6":
-        specs[1] = "activity-id"
+    return {
+        "seperator": "/",
+        "template": template_raw,
+        "specs": [_get_prefix_spec(s)] + [_get_part_spec(s, i) for i in parts],
+        "suffix": "#"
+    }
 
-    # Set directory prefix. 
+
+def _get_prefix_spec(s: Scope) -> dict:
+    """Maps a scope to a constant specifiction acting as identifier prefix.
+    
+    """
     try:
         prefix: str = _PROJECT_PREFIX[s.namespace]
     except KeyError:
         prefix = s.canonical_name.upper()
 
     return {
-        "template": template_raw,
-        "seperator": "/",
-        "specs": [f"const:{prefix}"] + [f"{s}:{i}" for i in specs[1:]]
+        "type": "const",
+        "value": prefix,
+        "is_required": True
+    }
+
+
+def _get_part_spec(s: Scope, part: str) -> dict:
+    """Maps a template part to a collection specifiction.
+    
+    """
+    if s.namespace == "wcrp:cmip6" and part == "activity-drs":
+        part = "activity-id"
+
+    return {
+        "type": "collection",
+        "namespace": f"{s.namespace}:{part}",
+        "is_required": True
     }

jobs/set_parser_config/set_filename_id.py

@@ -1,42 +1,66 @@
-from pyessv import Authority
+import re
+
 from pyessv import Scope
 
 
-def get_config(a: Authority, s: Scope, template_raw: str):
+def get_config(s: Scope, template_raw: str) -> dict:
     """Returns directory identifier parser configuration information derived
        from a previously declared parsing template.
 
-    :param a: A vocabulary authority.
     :param s: A vocabulary scope.
     :param template_raw: A raw dataset id parsing template.
     :returns: File name parser configuration information.
 
     """
-    # Skip ill-defined.
     if s.namespace == "wcrp:e3sm":
         return
 
-    # Discard period start/end + file type.
-    template = template_raw.split("[")[0]
+    parts = [i.replace("_", "-") for i in re.findall("%\((\w*)\)s", template_raw)]
+    if parts[-2] == "period-start" and parts[-1] == "period-end":
+        parts = parts[:-2] + ["time_range"]
+
+    return {
+        "seperator": "_",
+        "template": template_raw,
+        "specs": [_get_part_spec(s, i) for i in parts] + [_get_suffix_spec()]
+    }
+
+
+def _get_part_spec(s: Scope, part: str) -> dict:
+    """Maps a template part to a collection specifiction.
+    
+    """
+    if s.namespace == "wcrp:cmip6" and part == "activity-drs":
+        part = "activity-id"
 
-    # Set specs.
-    specs = [i.split(")")[0] for i in template.split("%(")[1:]]
-    specs = [i.replace("_", "-") for i in specs]
-    specs = [f"{s}:{i}" for i in specs]
+    if s.namespace == "ecmwf:cc4e" and part == "project":
+        return {
+            "type": "const",
+            "value": "cc4e",
+            "is_required": True
+        }
 
-    # Set spec overrides.
-    if s.namespace == "ecmwf:cc4e":
-        specs[0] = "const:cc4e"
+    elif part == "time_range":
+        return {
+            "type": "regex",
+            "expression": "^[0-9]{4}-[0-9]{4}$",
+            "is_required": False
+        }
 
-    # Append period start - end regex.
-    if s.namespace != "wcrp:input4mips":
-        specs.append("regex:^[0-9]{4}-[0-9]{4}$")
+    else:
+        return {
+            "type": "collection",
+            "namespace": f"{s.namespace}:{part}",
+            "is_required": True
+        }
 
-    # Append file type.
-    specs.append("regex:^nc$")
 
+def _get_suffix_spec() -> dict:
+    """Maps a scope to a constant specifiction acting as identifier prefix.
+    
+    """
     return {
-        "template": template_raw,
-        "seperator": "_",
-        "specs": specs
+        "type": "regex",
+        "expression": "^nc$",
+        "is_required": True
     }

pyessv/parsing/identifiers/parser_spec.py

@@ -61,19 +61,19 @@ class RegExParsingSpecification(ParsingSpecification):
     """Encapsulates specification information related to a regex element parser.
 
     """
-    def __init__(self, regex, is_required):
+    def __init__(self, expression, is_required):
         """Instance initializer.
 
-        :param regex: A regular expression against which an identifier element will be validated.
+        :param expression: A regular expression against which an identifier element will be validated.
         :param is_required: Flag indicating whether the identifier element must exist.
 
         """
         super(RegExParsingSpecification, self).__init__("regex", is_required)
-        self.regex = regex
+        self.expression = expression
 
 
     def __repr__(self):
         """Instance representation.
 
         """
-        return f"parser-spec|regex::{self.regex}::{self.is_required}"
+        return f"parser-spec|regex::{self.expression}::{self.is_required}"