ds-caselaw-marklogic-api-client 28.1.0__tar.gz → 29.0.0__tar.gz

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ds-caselaw-marklogic-api-client
-Version: 28.1.0
+Version: 29.0.0
 Summary: An API client for interacting with the underlying data in Find Caselaw.
 Home-page: https://github.com/nationalarchives/ds-caselaw-custom-api-client
 Keywords: national archives,caselaw
@@ -11,7 +11,7 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Requires-Dist: boto3 (>=1.26.112,<2.0.0)
-Requires-Dist: certifi (>=2024.8.30,<2024.9.0)
+Requires-Dist: certifi (>=2024.12.14,<2024.13.0)
 Requires-Dist: charset-normalizer (>=3.0.0,<4.0.0)
 Requires-Dist: django-environ (>=0.11.0,<0.12.0)
 Requires-Dist: ds-caselaw-utils (>=2.0.0,<3.0.0)
@@ -25,6 +25,7 @@ Requires-Dist: pytz (>=2024.1,<2025.0)
 Requires-Dist: requests (>=2.28.2,<3.0.0)
 Requires-Dist: requests-toolbelt (>=0.10.1,<1.1.0)
 Requires-Dist: saxonche (>=12.5.0,<13.0.0)
+Requires-Dist: sqids (>=0.5.0,<0.6.0)
 Requires-Dist: typing-extensions (>=4.7.1,<5.0.0)
 Description-Content-Type: text/markdown
 

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "ds-caselaw-marklogic-api-client"
-version = "28.1.0"
+version = "29.0.0"
 description = "An API client for interacting with the underlying data in Find Caselaw."
 authors = ["The National Archives"]
 homepage = "https://github.com/nationalarchives/ds-caselaw-custom-api-client"
@@ -12,7 +12,7 @@ packages = [
 
 [tool.poetry.dependencies]
 python = "^3.9"
-certifi = ">=2024.8.30,<2024.9.0"
+certifi = ">=2024.12.14,<2024.13.0"
 charset-normalizer = "^3.0.0"
 django-environ = "^0.11.0"
 idna = "^3.4"
@@ -28,6 +28,7 @@ mypy-boto3-sns = "^1.26.69"
 pytz = "^2024.1"
 python-dateutil = "^2.9.0-post.0"
 saxonche = "^12.5.0"
+sqids = "^0.5.0"
 
 [tool.poetry.group.dev.dependencies]
 coverage = "^7.2.3"

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/Client.py

@@ -1217,3 +1217,7 @@ class MarklogicApiClient:
             ),
         )
         return IdentifierResolutions.from_marklogic_output(raw_results)
+
+    def get_next_document_sequence_number(self) -> int:
+        """Increment the MarkLogic sequence number by one and return the value."""
+        return int(self._eval_and_decode({}, "get_next_document_sequence_number.xqy"))

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/documents/__init__.py

@@ -15,6 +15,8 @@ from caselawclient.errors import (
     NotSupportedOnVersion,
     OnlySupportedOnVersion,
 )
+from caselawclient.models.identifiers import Identifier
+from caselawclient.models.identifiers.fclid import FindCaseLawIdentifier, FindCaseLawIdentifierSchema
 from caselawclient.models.identifiers.unpacker import unpack_all_identifiers_from_etree
 from caselawclient.models.utilities import VersionsDict, extract_version, render_versions
 from caselawclient.models.utilities.aws import (
@@ -170,13 +172,11 @@ class Document:
         self.identifiers = unpack_all_identifiers_from_etree(identifiers_element_as_etree)
 
     @property
-    def best_human_identifier(self) -> Optional[str]:
-        """
-        Some identifier that is understood by legal professionals to refer to this legal event
-        that is not the name of the document.
-        Typically, this will be the neutral citation number, should it exist.
-        Should typically be overridden in subclasses.
-        """
+    def best_human_identifier(self) -> Optional[Identifier]:
+        """Return the preferred identifier for the document, providing that it is considered human readable."""
+        preferred_identifier = self.identifiers.preferred()
+        if preferred_identifier and preferred_identifier.schema.human_readable:
+            return preferred_identifier
         return None
 
     @property
@@ -432,6 +432,12 @@ class Document:
         if not self.is_publishable:
             raise CannotPublishUnpublishableDocument
 
+        ## If it doesn't already have one, get a new FCLID for this document and assign
+        if len(self.identifiers.of_type(FindCaseLawIdentifier)) < 1:
+            document_fclid = FindCaseLawIdentifierSchema.mint(self.api_client)
+            self.identifiers.add(document_fclid)
+            self.save_identifiers()
+
         publish_documents(uri_for_s3(self.uri))
         self.api_client.set_published(self.uri, True)
         announce_document_event(
@@ -500,13 +506,17 @@ class Document:
             "documentType": parser_type_noun,
             "metadata": {
                 "name": self.body.name or None,
-                "cite": self.best_human_identifier or None,
+                "cite": None,
                 "court": self.body.court or None,
                 "date": checked_date,
                 "uri": self.uri,
             },
         }
 
+        ## TODO: Remove this hack around the fact that NCNs are assumed to be present for all documents' metadata, but actually different document classes may have different metadata
+        if hasattr(self, "neutral_citation"):
+            parser_instructions["metadata"]["cite"] = self.neutral_citation
+
         request_parse(
             uri=self.uri,
             reference=self.consignment_reference,

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/identifiers/__init__.py

@@ -32,13 +32,20 @@ class IdentifierSchema(ABC):
     name: str
     namespace: str
 
+    human_readable: bool
+    """ Should this identifier type be considered for display as a 'human readable' identifier? """
+
+    base_score_multiplier: float = 1.0
+    """ A multiplier used to adjust the relative ranking of this identifier when calculating preferred identifiers. """
+
     def __init_subclass__(cls: type["IdentifierSchema"], **kwargs: Any) -> None:
         """Ensure that subclasses have the required attributes set."""
         for required in (
             "name",
             "namespace",
+            "human_readable",
         ):
-            if not getattr(cls, required, False):
+            if not hasattr(cls, required):
                 raise NotImplementedError(f"Can't instantiate IdentifierSchema without {required} attribute.")
         super().__init_subclass__(**kwargs)
 
@@ -101,6 +108,11 @@ class Identifier(ABC):
     def url_slug(self) -> str:
         return self.schema.compile_identifier_url_slug(self.value)
 
+    @property
+    def score(self) -> float:
+        """Return the score of this identifier, used to calculate the preferred identifier for a document."""
+        return 1 * self.schema.base_score_multiplier
+
     def same_as(self, other: "Identifier") -> bool:
         "Is this the same as another identifier (in value and schema)?"
         return self.value == other.value and self.schema == other.schema
@@ -127,6 +139,11 @@ class Identifiers(dict[str, Identifier]):
         else:
             super().__delitem__(key)
 
+    def of_type(self, identifier_type: type[Identifier]) -> list[Identifier]:
+        """Return a list of all identifiers of a given type."""
+        uuids = self.keys()
+        return [self[uuid] for uuid in list(uuids) if isinstance(self[uuid], identifier_type)]
+
     def delete_type(self, deleted_identifier_type: type[Identifier]) -> None:
         "For when we want an identifier to be the only valid identifier of that type, delete the others first"
         uuids = self.keys()
@@ -144,3 +161,22 @@ class Identifiers(dict[str, Identifier]):
             identifiers_root.append(identifier.as_xml_tree)
 
         return identifiers_root
+
+    def by_score(self, type: Optional[type[Identifier]] = None) -> list[Identifier]:
+        """
+        :param type: Optionally, an identifier type to constrain this list to.
+
+        :return: Return a list of identifiers, sorted by their score in descending order.
+        """
+        identifiers = self.of_type(type) if type else list(self.values())
+        return sorted(identifiers, key=lambda v: v.score, reverse=True)
+
+    def preferred(self, type: Optional[type[Identifier]] = None) -> Optional[Identifier]:
+        """
+        :param type: Optionally, an identifier type to constrain the results to.
+
+        :return: Return the highest scoring identifier of the given type (or of any type, if none is specified). Returns `None` if no identifier is available.
+        """
+        if len(self) == 0:
+            return None
+        return self.by_score(type)[0]

ds_caselaw_marklogic_api_client-29.0.0/src/caselawclient/models/identifiers/fclid.py

@@ -0,0 +1,50 @@
+import re
+from typing import TYPE_CHECKING
+
+from sqids import Sqids
+
+from . import Identifier, IdentifierSchema
+
+if TYPE_CHECKING:
+    from caselawclient.Client import MarklogicApiClient
+
+
+VALID_FCLID_PATTERN = re.compile(r"^[bcdfghjkmnpqrstvwxyz23456789]{4,}$")
+
+FCLID_MINIMUM_LENGTH = 8
+FCLID_ALPHABET = "bcdfghjkmnpqrstvwxyz23456789"
+
+sqids = Sqids(
+    min_length=FCLID_MINIMUM_LENGTH,
+    alphabet=FCLID_ALPHABET,
+)
+
+
+class FindCaseLawIdentifierSchema(IdentifierSchema):
+    """
+    Identifier schema describing a Find Case Law Identifier.
+    """
+
+    name = "Find Case Law Identifier"
+    namespace = "fclid"
+    human_readable = False
+    base_score_multiplier = 0.8
+
+    @classmethod
+    def validate_identifier(cls, value: str) -> bool:
+        return bool(VALID_FCLID_PATTERN.match(value))
+
+    @classmethod
+    def compile_identifier_url_slug(cls, value: str) -> str:
+        return "tna." + value
+
+    @classmethod
+    def mint(cls, api_client: "MarklogicApiClient") -> "FindCaseLawIdentifier":
+        """Generate a totally new Find Case Law identifier."""
+        next_sequence_number = api_client.get_next_document_sequence_number()
+        new_identifier = sqids.encode([next_sequence_number])
+        return FindCaseLawIdentifier(value=new_identifier)
+
+
+class FindCaseLawIdentifier(Identifier):
+    schema = FindCaseLawIdentifierSchema

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/identifiers/neutral_citation.py

@@ -30,6 +30,8 @@ class NeutralCitationNumberSchema(IdentifierSchema):
 
     name = "Neutral Citation Number"
     namespace = "ukncn"
+    human_readable = True
+    base_score_multiplier = 1.5
 
     @classmethod
     def validate_identifier(cls, value: str) -> bool:

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/identifiers/unpacker.py

@@ -3,9 +3,11 @@ from typing import Optional
 from lxml import etree
 
 from . import IDENTIFIER_UNPACKABLE_ATTRIBUTES, Identifier, Identifiers, InvalidIdentifierXMLRepresentationException
+from .fclid import FindCaseLawIdentifier
 from .neutral_citation import NeutralCitationNumber
 
 IDENTIFIER_NAMESPACE_MAP: dict[str, type[Identifier]] = {
+    "fclid": FindCaseLawIdentifier,
     "ukncn": NeutralCitationNumber,
 }
 

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/judgments.py

@@ -25,20 +25,17 @@ class Judgment(NeutralCitationMixin, Document):
         super().__init__(self.document_noun, uri, *args, **kwargs)
 
     @cached_property
-    def neutral_citation(self) -> NeutralCitationString:
-        return NeutralCitationString(
-            self.body.get_xpath_match_string(
-                "/akn:akomaNtoso/akn:*/akn:meta/akn:proprietary/uk:cite/text()",
-                {
-                    "uk": "https://caselaw.nationalarchives.gov.uk/akn",
-                    "akn": "http://docs.oasis-open.org/legaldocml/ns/akn/3.0",
-                },
-            )
+    def neutral_citation(self) -> Optional[NeutralCitationString]:
+        value_in_xml = self.body.get_xpath_match_string(
+            "/akn:akomaNtoso/akn:*/akn:meta/akn:proprietary/uk:cite/text()",
+            {
+                "uk": "https://caselaw.nationalarchives.gov.uk/akn",
+                "akn": "http://docs.oasis-open.org/legaldocml/ns/akn/3.0",
+            },
         )
-
-    @property
-    def best_human_identifier(self) -> str:
-        return self.neutral_citation
+        if value_in_xml:
+            return NeutralCitationString(value_in_xml)
+        return None
 
     @cached_property
     def linked_document(self) -> Optional["PressSummary"]:

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/neutral_citation_mixin.py

@@ -1,9 +1,10 @@
 from abc import ABC, abstractmethod
 from functools import cached_property
-from typing import Any
+from typing import Any, Optional
 
 from ds_caselaw_utils import neutral_url
 from ds_caselaw_utils.types import NeutralCitationString
+from typing_extensions import deprecated
 
 
 class NeutralCitationMixin(ABC):
@@ -38,12 +39,15 @@ class NeutralCitationMixin(ABC):
 
     @cached_property
     @abstractmethod
-    def neutral_citation(self) -> NeutralCitationString: ...
+    @deprecated("Legacy usage of NCNs is deprecated; you should be moving to the Identifiers framework")
+    def neutral_citation(self) -> Optional[NeutralCitationString]: ...
 
     @cached_property
+    @deprecated("Legacy usage of NCNs is deprecated; you should be moving to the Identifiers framework")
     def has_ncn(self) -> bool:
-        return bool(self.neutral_citation)
+        return self.neutral_citation is not None and self.neutral_citation != ""
 
     @cached_property
+    @deprecated("Legacy usage of NCNs is deprecated; you should be moving to the Identifiers framework")
     def has_valid_ncn(self) -> bool:
-        return self.has_ncn and neutral_url(self.neutral_citation) is not None
+        return self.neutral_citation is not None and neutral_url(self.neutral_citation) is not None

{ds_caselaw_marklogic_api_client-28.1.0 → ds_caselaw_marklogic_api_client-29.0.0}/src/caselawclient/models/press_summaries.py

@@ -27,19 +27,16 @@ class PressSummary(NeutralCitationMixin, Document):
         super().__init__(self.document_noun, uri, *args, **kwargs)
 
     @cached_property
-    def neutral_citation(self) -> NeutralCitationString:
-        return NeutralCitationString(
-            self.body.get_xpath_match_string(
-                "/akn:akomaNtoso/akn:doc/akn:preface/akn:p/akn:neutralCitation/text()",
-                {
-                    "akn": "http://docs.oasis-open.org/legaldocml/ns/akn/3.0",
-                },
-            )
+    def neutral_citation(self) -> Optional[NeutralCitationString]:
+        value_in_xml = self.body.get_xpath_match_string(
+            "/akn:akomaNtoso/akn:doc/akn:preface/akn:p/akn:neutralCitation/text()",
+            {
+                "akn": "http://docs.oasis-open.org/legaldocml/ns/akn/3.0",
+            },
         )
-
-    @property
-    def best_human_identifier(self) -> str:
-        return self.neutral_citation
+        if value_in_xml:
+            return NeutralCitationString(value_in_xml)
+        return None
 
     @cached_property
     def linked_document(self) -> Optional[Judgment]:

ds_caselaw_marklogic_api_client-29.0.0/src/caselawclient/xquery/get_next_document_sequence_number.xqy

@@ -0,0 +1,14 @@
+xquery version "1.0-ml";
+declare option xdmp:transaction-mode "update";
+
+let $_ := xdmp:set-transaction-mode("update")
+let $state_doc := fn:doc("state.xml")
+let $counter_node := $state_doc/state/document_counter
+
+let $current_counter := $counter_node/text()
+let $new_counter := fn:sum(($current_counter, 1))
+
+let $_ := xdmp:node-replace($counter_node, <document_counter>{$new_counter}</document_counter>)
+let $_ := xdmp:commit()
+
+return $new_counter