civics-cdf-validator 1.43.dev4__tar.gz → 1.43.dev6__tar.gz

{civics_cdf_validator-1.43.dev4/civics_cdf_validator.egg-info → civics_cdf_validator-1.43.dev6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: civics_cdf_validator
-Version: 1.43.dev4
+Version: 1.43.dev6
 Summary: Checks if an election feed follows best practices
 Home-page: https://github.com/google/civics_cdf_validator
 Author: Google Civics
@@ -15,6 +15,7 @@ Requires-Dist: requests>=2.10
 Requires-Dist: pygithub>=1.28
 Requires-Dist: networkx>=2.6.3
 Requires-Dist: pycountry==22.1.10
+Requires-Dist: frozendict>=2.4.4
 Requires-Dist: six
 
 civics_cdf_validator is a script that checks if an election data feed follows best practices and outputs errors, warnings and info messages for common issues.

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/README.md

@@ -32,6 +32,7 @@ You can use `civics_cdf_validator` to check different types of feed:
 * Results
 * Committee
 * Election Dates
+* Metadata
 
 ## List rules
 
@@ -81,6 +82,12 @@ default). Examples:
   civics_cdf_validator validate results_file.xml --xsd civics_cdf_spec.xsd --rule_set election_results
   ```
 
+  * Validate a metadata feed:
+
+  ```
+  civics_cdf_validator validate metadata_file.xml --xsd metadata_spec.xsd --rule_set metadata
+  ```
+
 One can choose to only validate one or more comma separated rules by using the `-i` flag
 
 ```

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6/civics_cdf_validator.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: civics_cdf_validator
-Version: 1.43.dev4
+Version: 1.43.dev6
 Summary: Checks if an election feed follows best practices
 Home-page: https://github.com/google/civics_cdf_validator
 Author: Google Civics
@@ -15,6 +15,7 @@ Requires-Dist: requests>=2.10
 Requires-Dist: pygithub>=1.28
 Requires-Dist: networkx>=2.6.3
 Requires-Dist: pycountry==22.1.10
+Requires-Dist: frozendict>=2.4.4
 Requires-Dist: six
 
 civics_cdf_validator is a script that checks if an election data feed follows best practices and outputs errors, warnings and info messages for common issues.

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/civics_cdf_validator.egg-info/requires.txt

@@ -4,4 +4,5 @@ requests>=2.10
 pygithub>=1.28
 networkx>=2.6.3
 pycountry==22.1.10
+frozendict>=2.4.4
 six

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/rules.py

@@ -25,9 +25,11 @@ from civics_cdf_validator import base
 from civics_cdf_validator import gpunit_rules
 from civics_cdf_validator import loggers
 from civics_cdf_validator import office_utils
+from frozendict import frozendict
 import language_tags
 from lxml import etree
 import networkx
+import pycountry
 from six.moves.urllib.parse import urlparse
 
 _PARTY_LEADERSHIP_TYPES = ["party-leader-id", "party-chair-id"]
@@ -75,6 +77,14 @@ _EXECUTIVE_OFFICE_ROLES = frozenset([
     "deputy head of government",
 ])
 
+_VALID_FEED_LONGEVITY_BY_FEED_TYPE = frozendict({
+    "committee": ["evergreen"],
+    "election-dates": ["evergreen"],
+    "election-results": ["limited", "yearly"],
+    "officeholder": ["evergreen"],
+    "pre-election": ["limited", "yearly"],
+})
+
 
 def _is_executive_office(office_roles):
   return not _EXECUTIVE_OFFICE_ROLES.isdisjoint(office_roles)
@@ -179,6 +189,14 @@ def element_has_text(element):
           and not element.text.isspace())
 
 
+def country_code_is_valid(country_code):
+  # EU is part of ISO 3166/MA
+  return (
+      country_code == "eu"
+      or pycountry.countries.get(alpha_2=country_code) is not None
+  )
+
+
 class Schema(base.TreeRule):
   """Checks if election file validates against the provided schema."""
 
@@ -3573,6 +3591,199 @@ class AffiliationHasEitherPartyOrPerson(base.BaseRule):
       )
 
 
+class FeedTypeHasValidFeedLongevity(base.BaseRule):
+  """Feeds types should have valid corresponding FeedLongevity."""
+
+  def elements(self):
+    return ["Feed"]
+
+  def check(self, element):
+    feed_type_element = element.find("FeedType")
+    feed_longevity_element = element.find("FeedLongevity")
+    if element_has_text(feed_type_element) and element_has_text(
+        feed_longevity_element
+    ):
+      feed_type = feed_type_element.text.lower().replace("_", "-")
+      feed_longevity = feed_longevity_element.text.lower().replace("_", "-")
+      if (
+          feed_type in _VALID_FEED_LONGEVITY_BY_FEED_TYPE
+          and feed_longevity
+          not in _VALID_FEED_LONGEVITY_BY_FEED_TYPE[feed_type]
+      ):
+        raise loggers.ElectionError.from_message(
+            "Feed type {} has invalid feed longevity {}. Valid feed"
+            " longevities for this type are {}".format(
+                feed_type,
+                feed_longevity,
+                _VALID_FEED_LONGEVITY_BY_FEED_TYPE[feed_type],
+            ),
+            [element],
+        )
+
+
+class FeedIdsAreUnique(base.BaseRule):
+  """FeedId should be unique."""
+
+  def elements(self):
+    return ["FeedCollection"]
+
+  def check(self, element):
+    feed_ids = set()
+    error_log = []
+    for feed_element in element.findall("Feed"):
+      if element_has_text(feed_element.find("FeedId")):
+        feed_id = feed_element.find("FeedId").text
+        if feed_id in feed_ids:
+          msg = (
+              "FeedId {} appears multiple times in the metadata feed. Feed ids"
+              " must be unique.".format(feed_id)
+          )
+          error_log.append(
+              loggers.LogEntry(
+                  msg,
+                  [feed_element],
+              )
+          )
+        feed_ids.add(feed_id)
+
+    if error_log:
+      raise loggers.ElectionError(error_log)
+
+
+class SourceDirPathsAreUnique(base.BaseRule):
+  """All SourceDirPaths should be unique."""
+
+  def elements(self):
+    return ["FeedCollection"]
+
+  def check(self, element):
+    source_dir_paths = set()
+    error_log = []
+    for feed_element in element.findall("Feed"):
+      if element_has_text(feed_element.find("SourceDirPath")):
+        source_dir_path = feed_element.find("SourceDirPath").text
+        if source_dir_path in source_dir_paths:
+          msg = (
+              "SourceDirPath {} appears multiple times in the metadata feed."
+              " SourceDirPaths must be unique.".format(source_dir_path)
+          )
+          error_log.append(
+              loggers.LogEntry(
+                  msg,
+                  [feed_element],
+              )
+          )
+        source_dir_paths.add(source_dir_path)
+
+    if error_log:
+      raise loggers.ElectionError(error_log)
+
+
+class ElectionEventDatesAreSequential(base.DateRule):
+  """Dates in an ElectionEvent element should be sequential."""
+
+  def elements(self):
+    return ["ElectionEvent"]
+
+  def check(self, element):
+    self.reset_instance_vars()
+    self.gather_dates(element)
+    self.check_end_after_start()
+    if element_has_text(element.find("FullDeliveryDate")) and self.start_date:
+      full_delivery_date = base.PartialDate.init_partial_date(
+          element.find("FullDeliveryDate").text
+      )
+      date_delta = self.start_date.is_older_than(full_delivery_date)
+      if date_delta > 0:
+        self.error_log.append(
+            loggers.LogEntry(
+                "StartDate is older than FullDeliveryDate",
+                [element],
+            )
+        )
+    if element_has_text(
+        element.find("InitialDeliveryDate")
+    ) and element_has_text(element.find("FullDeliveryDate")):
+      initial_delivery_date = base.PartialDate.init_partial_date(
+          element.find("InitialDeliveryDate").text
+      )
+      full_delivery_date = base.PartialDate.init_partial_date(
+          element.find("FullDeliveryDate").text
+      )
+      date_delta = full_delivery_date.is_older_than(initial_delivery_date)
+      if date_delta > 0:
+        self.error_log.append(
+            loggers.LogEntry(
+                "FullDeliveryDate is older than InitialDeliveryDate",
+                [element],
+            )
+        )
+
+    if self.error_log:
+      raise loggers.ElectionError(self.error_log)
+
+
+class OfficeHolderSubFeedDatesAreSequential(base.DateRule):
+  """Dates in an OfficeHolderSubFeed element should be sequential."""
+
+  def elements(self):
+    return ["OfficeHolderSubFeed"]
+
+  def check(self, element):
+    if element_has_text(
+        element.find("InitialDeliveryDate")
+    ) and element_has_text(element.find("FullDeliveryDate")):
+      initial_delivery_date = base.PartialDate.init_partial_date(
+          element.find("InitialDeliveryDate").text
+      )
+      full_delivery_date = base.PartialDate.init_partial_date(
+          element.find("FullDeliveryDate").text
+      )
+      date_delta = full_delivery_date.is_older_than(initial_delivery_date)
+      if date_delta > 0:
+        raise loggers.ElectionError.from_message(
+            "FullDeliveryDate is older than InitialDeliveryDate",
+            [element],
+        )
+
+
+class FeedHasValidCountryCode(base.BaseRule):
+  """Feeds should have valid country code."""
+
+  def elements(self):
+    return ["Feed"]
+
+  def check(self, element):
+    country_code_element = element.find("CountryCode")
+    if element_has_text(country_code_element):
+      country_code = country_code_element.text.upper()
+      if not country_code_is_valid(country_code):
+        raise loggers.ElectionError.from_message(
+            "Invalid country code {}.".format(country_code),
+            [element],
+        )
+
+
+class FeedInactiveDateSetForNonEvergreenFeed(base.BaseRule):
+  """All non-evergreen feeds should have a FeedInactiveDate set."""
+
+  def elements(self):
+    return ["Feed"]
+
+  def check(self, element):
+    feed_longevity = element.find("FeedLongevity")
+    if (
+        element_has_text(feed_longevity)
+        and feed_longevity.text.lower() != "evergreen"
+        and not element_has_text(element.find("FeedInactiveDate"))
+    ):
+      raise loggers.ElectionError.from_message(
+          "FeedInactiveDate is not set for non-evergreen feed with FeedId {}."
+          .format(element.find("FeedId").text),
+          [element],
+      )
+
+
 class UnreferencedEntitiesBase(base.TreeRule):
   """All non-top-level entities in a feed should be referenced by at least one other entity.
 
@@ -3831,6 +4042,13 @@ METADATA_RULES = (
     Encoding,
     OptionalAndEmpty,
     UniqueLabel,
+    FeedTypeHasValidFeedLongevity,
+    FeedIdsAreUnique,
+    SourceDirPathsAreUnique,
+    ElectionEventDatesAreSequential,
+    OfficeHolderSubFeedDatesAreSequential,
+    FeedHasValidCountryCode,
+    FeedInactiveDateSetForNonEvergreenFeed,
 )
 
 ALL_RULES = frozenset(

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/setup.py

@@ -58,6 +58,7 @@ setup(
         'pygithub>=1.28',
         'networkx>=2.6.3',
         'pycountry==22.1.10',
+        'frozendict>=2.4.4',
         'six',
     ],
     setup_requires=['pytest-runner'],

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/validator.py

@@ -33,7 +33,6 @@ from civics_cdf_validator import gpunit_rules
 from civics_cdf_validator import loggers
 from civics_cdf_validator import rules
 from civics_cdf_validator import version
-import pycountry
 
 _REGISTRY_KEY = "registry"
 _METADATA_KEY = "metadata"
@@ -82,14 +81,9 @@ def _validate_country_codes(parser, arg):
   """
   country_code = arg.strip().lower()
 
-  # EU is part of ISO 3166/MA
-  if country_code == "eu":
+  if rules.country_code_is_valid(country_code):
     return country_code
 
-  for country in pycountry.countries:
-    if country_code == country.alpha_2.lower():
-      return country_code
-
   parser.error(
       "Invalid country code. Please make sure it is listed under the officially"
       " assigned ISO 3166-1 alpha-2 codes."

{civics_cdf_validator-1.43.dev4 → civics_cdf_validator-1.43.dev6}/version.py

@@ -5,4 +5,4 @@ No dependencies should be added to this module.
 See https://packaging.python.org/guides/single-sourcing-package-version/
 """
 
-__version__ = '1.43.dev4'
+__version__ = '1.43.dev6'