followthemoney 4.2.0__py3-none-any.whl → 4.2.2__py3-none-any.whl

followthemoney/__init__.py

@@ -9,7 +9,7 @@ from followthemoney.statement import Statement, StatementEntity, SE
 from followthemoney.dataset import Dataset, DefaultDataset, DS
 from followthemoney.util import set_model_locale
 
-__version__ = "4.2.0"
+__version__ = "4.2.2"
 
 # Data model singleton
 model = Model.instance()

followthemoney/property.py

@@ -1,3 +1,4 @@
+import re
 from banal import is_mapping, as_bool
 from typing import TYPE_CHECKING, Any, List, Optional, TypedDict
 
@@ -9,6 +10,18 @@ if TYPE_CHECKING:
     from followthemoney.schema import Schema
     from followthemoney.model import Model
 
+# Invalid property names.
+RESERVED = ["id", "caption", "schema", "schemata", "referents", "datasets"]
+PROP_NAME_RE = re.compile("^[a-z][a-zA-Z0-9]*$")
+
+
+def check_property_name(name: str) -> bool:
+    if name in RESERVED:
+        return False
+    if not PROP_NAME_RE.match(name):
+        return False
+    return True
+
 
 class ReverseSpec(TypedDict, total=False):
     name: str
@@ -66,9 +79,6 @@ class Property:
         "reverse",
     )
 
-    #: Invalid property names.
-    RESERVED = ["id", "caption", "schema", "schemata"]
-
     def __init__(self, schema: "Schema", name: str, data: PropertySpec) -> None:
         #: The schema which the property is defined for. This is always the
         #: most abstract schema that has this property, not the possible
@@ -77,11 +87,11 @@ class Property:
 
         #: Machine-readable name for this property.
         self.name = const(name)
+        if not check_property_name(self.name):
+            raise InvalidModel("Invalid name: %s" % self.name)
 
         #: Qualified property name, which also includes the schema name.
         self.qname = const("%s:%s" % (schema.name, self.name))
-        if self.name in self.RESERVED:
-            raise InvalidModel("Reserved name: %s" % self.name)
 
         self._hash = hash("<Property(%r)>" % self.qname)
 

followthemoney/schema/Company.yaml

@@ -58,20 +58,20 @@ Company:
       type: identifier
     caemCode:
       label: "COD CAEM"
-      description: "(RO) What kind of activity a legal entity is allowed to develop"
+      description: "Romanian classifier used to identify the types of economic activities that a business can provide in Romania"
       matchable: false
     kppCode:
       label: "KPP"
-      description: "(RU, КПП) in addition to INN for orgs; reason for registration at FNS"
+      description: "Russian code issued by the tax authority, identifies the reason for registration at the Federal Tax Service (Russian: КПП). A company may have multiple KPP codes (e.g. for different branches), and the codes are not unique across companies."
       type: identifier
       matchable: false
     okvedCode:
       label: "OKVED(2) Classifier"
-      description: "(RU, ОКВЭД) Economical activity classifier. OKVED2 is the same but newer"
+      description: "Russian classifier that that categorizes businesses by their (primary and secondary) economic activities (Russian: ОКВЭД)"
       matchable: false
     okopfCode:
       label: "OKOPF"
-      description: "(RU, ОКОПФ) What kind of business entity"
+      description: "Russian classifier that that categorizes different types of legal entities in Russia based on their organizational and legal structure (Russian: ОКОПФ)"
       matchable: false
     fnsCode:
       label: "Federal tax service code"

followthemoney/schema/LegalEntity.yaml

@@ -74,7 +74,7 @@ LegalEntity:
     idNumber:
       label: ID Number
       type: identifier
-      description: "ID number of any applicable ID"
+      description: "ID of any applicable personal identification document. Used mainly for people and their national ID cards."
     taxNumber:
       label: Tax Number
       type: identifier
@@ -118,13 +118,13 @@ LegalEntity:
       matchable: false
     innCode:
       label: "INN"
-      description: "Russian company ID"
+      description: "Russian tax identification number (Russian: ИНН). Issued to businesses and individuals in Russia"
       type: identifier
       format: inn
       maxLength: 32
     ogrnCode:
       label: "OGRN"
-      description: "Major State Registration Number"
+      description: "Identification number used in Russia's Unified State Register of Legal Entities (EGRUL) (Russian: ОГРН)"
       type: identifier
       format: ogrn
       maxLength: 32
@@ -148,7 +148,7 @@ LegalEntity:
       maxLength: 32
     npiCode:
       label: "NPI"
-      description: "National Provider Identifier"
+      description: "National Provider Identifier, issued to health care providers in the United States"
       type: identifier
       format: npi
       maxLength: 16

followthemoney/statement/entity.py

@@ -2,6 +2,8 @@ from hashlib import sha1
 from collections.abc import Mapping
 from typing import Any, Dict, List, Optional, Set, Type
 from typing import Generator, Iterable, Tuple, TypeVar
+from rigour.langs import LangStr
+from rigour.names.pick import pick_lang_name
 
 from followthemoney.model import Model
 from followthemoney.exc import InvalidData
@@ -163,6 +165,7 @@ class StatementEntity(EntityProxy):
                 else:
                     self.last_change = max(self.last_change, stmt.first_seen)
         else:
+            self._caption = None
             if stmt.prop not in self._statements:
                 self._statements[stmt.prop] = set()
             self._statements[stmt.prop].add(stmt)
@@ -296,13 +299,17 @@ class StatementEntity(EntityProxy):
         prop_name = self._prop_name(prop, quiet=quiet)
         if prop_name is None or prop_name not in self._statements:
             return []
+        if prop_name in self.schema.caption:
+            self._caption = None
         return list({s.value for s in self._statements.pop(prop_name, [])})
 
     def remove(self, prop: P, value: str, quiet: bool = True) -> None:
         prop_name = self._prop_name(prop, quiet=quiet)
-        if prop_name is not None and prop_name in self._properties:
+        if prop_name is not None and prop_name in self._statements:
             stmts = {s for s in self._statements[prop_name] if s.value != value}
             self._statements[prop_name] = stmts
+            if prop_name in self.schema.caption:
+                self._caption = None
 
     def itervalues(self) -> Generator[Tuple[Property, str], None, None]:
         for name, statements in self._statements.items():
@@ -335,6 +342,34 @@ class StatementEntity(EntityProxy):
     def properties(self) -> Dict[str, List[str]]:
         return {p: list({s.value for s in vs}) for p, vs in self._statements.items()}
 
+    @property
+    def caption(self) -> str:
+        """The user-facing label to be used for this entity. This checks a list
+        of properties defined by the schema (caption) and returns the first
+        available value. If no caption is available, return the schema label.
+
+        This implementation prefers statements where the language property is that
+        of the preferred system language."""
+        if self._caption is None:
+            for prop_ in self.schema.caption:
+                stmts = self._statements.get(prop_)
+                if stmts is None:
+                    continue
+                prop = self.schema.properties[prop_]
+                if prop.type == registry.name and len(stmts) > 1:
+                    values = [LangStr(s.value, lang=s.lang) for s in stmts]
+                    name = pick_lang_name(values)
+                    if name is not None:
+                        self._caption = name
+                        return self._caption
+
+                for stmt in sorted(stmts):
+                    self._caption = stmt.value
+                    return self._caption
+            if self._caption is None:
+                self._caption = self.schema.label
+        return self._caption
+
     def iterprops(self) -> List[Property]:
         return [self.schema.properties[p] for p in self._statements.keys()]
 
@@ -353,8 +388,8 @@ class StatementEntity(EntityProxy):
             raise InvalidData(msg % (self.id, e))
 
         if not isinstance(other, StatementEntity):
-            for prop, values in other._properties.items():
-                self.add(prop, values, cleaned=True, quiet=True)
+            for prop, value in other.itervalues():
+                self.unsafe_add(prop, value, cleaned=True, quiet=True)
             return self
         for stmt in other._iter_stmt():
             if self.id is not None:

{followthemoney-4.2.0.dist-info → followthemoney-4.2.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: followthemoney
-Version: 4.2.0
+Version: 4.2.2
 Summary: A data model for anti corruption data modeling and analysis.
 Project-URL: Documentation, https://followthemoney.tech/
 Project-URL: Repository, https://github.com/opensanctions/followthemoney.git
@@ -42,15 +42,15 @@ Requires-Dist: click<9.0.0,>=8.0
 Requires-Dist: networkx<3.5,>=2.5
 Requires-Dist: normality<4.0.0,>=3.0.1
 Requires-Dist: openpyxl<4.0.0,>=3.0.5
-Requires-Dist: orjson<4.0,>=3.10.18
+Requires-Dist: orjson<4.0,>=3.10.0
 Requires-Dist: phonenumbers<10.0.0,>=8.12.22
-Requires-Dist: prefixdate<1.0.0,>=0.4.0
-Requires-Dist: pydantic<3.0.0,>=2.11.7
+Requires-Dist: prefixdate<1.0.0,>=0.5.0
+Requires-Dist: pydantic<3.0.0,>=2.11.0
 Requires-Dist: pytz>=2021.1
 Requires-Dist: pyyaml<7.0.0,>=5.0.0
 Requires-Dist: rdflib<7.2.0,>=6.2.0
 Requires-Dist: requests<3.0.0,>=2.21.0
-Requires-Dist: rigour<2.0.0,>=1.2.0
+Requires-Dist: rigour<2.0.0,>=1.3.1
 Requires-Dist: sqlalchemy[mypy]<3.0.0,>=2.0.0
 Provides-Extra: dev
 Requires-Dist: build; extra == 'dev'

{followthemoney-4.2.0.dist-info → followthemoney-4.2.2.dist-info}/RECORD

@@ -1,4 +1,4 @@
-followthemoney/__init__.py,sha256=NqtJnba71wycGIhiZrBIfOYbafL_Tm_Qd0ouNm2EYPA,856
+followthemoney/__init__.py,sha256=pTkibO7jJUHnObuIGY1rsjFL9siMpBusU7E9OylQDBI,856
 followthemoney/compare.py,sha256=bZlnj2VMoe67q4Lyq_VwS1a-EJnEK1kC8prbs8jyL9E,5774
 followthemoney/entity.py,sha256=bBiX7hNquXemS3vYCUHKtWI_IqX43Z6i8RQDbZ7gXsg,3449
 followthemoney/exc.py,sha256=GyMgwY4QVm87hLevDfV7gM1MJsDqfNCi_UQw7F_A8X8,858
@@ -9,7 +9,7 @@ followthemoney/model.py,sha256=bWFVNa-DhYzc8BdSXBZdG2ev6Nh9uHx6i4tin8DvEEU,7374
 followthemoney/names.py,sha256=LODQqExKEHdH4z6Mmbhlm0KeKRzGcptaSWzYXZ7lONI,1120
 followthemoney/namespace.py,sha256=utggu9IGA8bhgEYom3OUB1KxkAJR_TrMNbY5MUF_db8,4536
 followthemoney/ontology.py,sha256=WWY_PYQGl5Ket4zZBuZglzQxD2Bh9UqHok6GJNNX7GA,3001
-followthemoney/property.py,sha256=RDTzTXJeeLFLptQL1_gr1S1T-vdDe-8MGMwsRaGQh0I,7665
+followthemoney/property.py,sha256=9qZ_o2iA-1llLMJ3O2hsW7c2XhkFU1YbvVqretGYUSA,7913
 followthemoney/proxy.py,sha256=LD4K1oPABXMX212UZxwLu7XOHRDyVBwTlqudTUsUZRQ,19619
 followthemoney/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 followthemoney/schema.py,sha256=WYnPE4Lego0pJHlojECEv0aO9Miw_YIvEb35HoDo4Zk,18087
@@ -54,7 +54,7 @@ followthemoney/schema/Audio.yaml,sha256=Eb1rZGUEOX7XDAj_1YIN28NCBzMvkopQBNwgHt_k
 followthemoney/schema/BankAccount.yaml,sha256=60v-VD296lW1Qq7fx--CzxfPNwfCcyMV6xIl8OrSy5g,1431
 followthemoney/schema/Call.yaml,sha256=kbVCnVxucBrEplxehXHThLSJAJjy_GhWan-IeZZjr0M,980
 followthemoney/schema/CallForTenders.yaml,sha256=2IWonTzfSbrkynMoEWqv5fekUeFM_xDKpKIbRe1XDbo,3227
-followthemoney/schema/Company.yaml,sha256=v16tU4mQhVitC3BOc9OPbubCxPgf1c-Iudnyq6IF0ys,3057
+followthemoney/schema/Company.yaml,sha256=v6OFHZjEIMQPSP9re6nAhI2kLk1MHfP9LfcCc7M-Ifc,3432
 followthemoney/schema/Contract.yaml,sha256=aSPB64T1h-0nuLDv6krasUvvoPZgo6sWUbv60c3vmzI,1541
 followthemoney/schema/ContractAward.yaml,sha256=b2spaZHYCaP1yR1RCsrI7mUjk-fAF7BUE3dc8Vl3cUQ,1689
 followthemoney/schema/CourtCase.yaml,sha256=lcovnY0Ne_xcggvkqfCW_RHvsRKo8kFTCPCyovAXRtI,599
@@ -75,7 +75,7 @@ followthemoney/schema/Identification.yaml,sha256=6txjZs6-3Kn94c3G4tDeDt9Jb4FW55-
 followthemoney/schema/Image.yaml,sha256=wuznboWECGiV96_GQiXq1-oKNoxO8zKisR4xyusnEn8,394
 followthemoney/schema/Interest.yaml,sha256=VUrehmsN1WgtS1oAa5jn_JGtSkZGGYLGNahp-R5JhOQ,282
 followthemoney/schema/Interval.yaml,sha256=8YJQ51GI-GxvbjYs3uC593kQtCepWW_7ZiNnlbPm2aM,2084
-followthemoney/schema/LegalEntity.yaml,sha256=u6GOemHj1vnL_6dcC3XyLaP72JQrNecXlbocXxJo5AI,4502
+followthemoney/schema/LegalEntity.yaml,sha256=S56ALyeCZVgjpNy9EqkVkPBIs6QaGuZw31s9bdtEeC4,4789
 followthemoney/schema/License.yaml,sha256=bXESXY-JpSmc5sthZe4sssXhx50UoLPAMED9FvEUyRU,534
 followthemoney/schema/Membership.yaml,sha256=IPmaOX4Ai2r4sGcA5ig2WmLvWHb38akdxp4smEdDWOE,710
 followthemoney/schema/Mention.yaml,sha256=nBeulR_Jm4x75aJ7yNF0TAVhHJqXQaEzOutLIn_YU-4,1086
@@ -114,7 +114,7 @@ followthemoney/schema/Vessel.yaml,sha256=zWHUfSK8g6Pz58ZyCaK0AFJ4u_UHjEIUGC4c_7o
 followthemoney/schema/Video.yaml,sha256=LY3DYMWTHXiAhL0hxBCNCz50cp2sPbUlEhhig5Fbjos,327
 followthemoney/schema/Workbook.yaml,sha256=iikWPElz4klA7SkWH7eae6xqhbkMCIP_3zdeXzFEMU0,354
 followthemoney/statement/__init__.py,sha256=7m2VUCAuqNZXIY0WFJRFkw5UG14QuxATL4f_xbqKwhw,633
-followthemoney/statement/entity.py,sha256=MKHGmFeDwcW2lTbAeKSdU53YwDuoLm5iKy4roeb8_lo,16172
+followthemoney/statement/entity.py,sha256=mUf-6uvrYvqq9azdHu7qFGkGK_0j9gT6nxscX_JXWR0,17673
 followthemoney/statement/serialize.py,sha256=9eXzQ1biR2mSxWRID5C7xDdku4b4ZImHeRJ53yLZ0yo,7225
 followthemoney/statement/statement.py,sha256=Ae-EYuzS8S12BkaRqrvMuI1C7YwlRKa5C_pTBELyNMM,8029
 followthemoney/statement/util.py,sha256=B-ozuRc1TWvpop52873Pqt5OPj8H6uk4KyRJLfAhr10,780
@@ -161,8 +161,8 @@ followthemoney/types/phone.py,sha256=r8uRqWinS0CYnYBTs405k5gO4jeatUDgjdzzijoMKJE
 followthemoney/types/string.py,sha256=fqyTauAm4mNnNaoH-yH087RBbNh-G5ZZUO3awTGQUUg,1230
 followthemoney/types/topic.py,sha256=Mi0Gx0m3bDeTmyuvM6jdRMqv81O03U4eI99R13KGu2Y,4503
 followthemoney/types/url.py,sha256=QFpS_JIV8unFHuh_uGv22SWUUkocBoOpzLsAJWom_gI,1455
-followthemoney-4.2.0.dist-info/METADATA,sha256=I1JXOtpniNIMi2jPoM1Vpx_IaYezqCP8l9e3qhldGtc,6748
-followthemoney-4.2.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-followthemoney-4.2.0.dist-info/entry_points.txt,sha256=caoFTlf213jhg5sz3TNSofutjUTzaKtWATuSIdd9Cps,653
-followthemoney-4.2.0.dist-info/licenses/LICENSE,sha256=H6_EVXisnJC0-18CjXIaqrBSFq_VH3OnS7u3dccOv6g,1148
-followthemoney-4.2.0.dist-info/RECORD,,
+followthemoney-4.2.2.dist-info/METADATA,sha256=Q7n2yZuDQJM27GTkVB0nMJgbMSVJLJ1RNCMmzABRFbY,6747
+followthemoney-4.2.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+followthemoney-4.2.2.dist-info/entry_points.txt,sha256=caoFTlf213jhg5sz3TNSofutjUTzaKtWATuSIdd9Cps,653
+followthemoney-4.2.2.dist-info/licenses/LICENSE,sha256=H6_EVXisnJC0-18CjXIaqrBSFq_VH3OnS7u3dccOv6g,1148
+followthemoney-4.2.2.dist-info/RECORD,,