chore: merge pull request #12 from swerik-project/dev

swerik-project · Sep 12, 2024 · f50f777 · f50f777
2 parents 6b7c006 + 65f3082
commit f50f777
Show file tree

Hide file tree

Showing 4 changed files with 93 additions and 45 deletions.
diff --git a/.github/workflows/validate-cff.yml b/.github/workflows/validate-cff.yml
@@ -0,0 +1,20 @@
+name: "Pyriksdagen: Validate CITATION cff"
+
+on:
+  push:
+    paths:
+      - CITATION.cff
+      - .github/workflows/validate-cff.yml
+
+jobs:
+  Validate-CITATION-cff:
+    runs-on: ubuntu-latest
+    env:
+      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Validate CITATION.cff
+        uses: dieghernan/cff-validator@v3
diff --git a/CITATION.cff b/CITATION.cff
@@ -0,0 +1,20 @@
+cff-version: 1.2.0
+message: To cite this reposository, please use these metadata.
+title: "The Swedish Parliament Corpus: pyriksdagen"
+version: v1.2.0
+authors:
+  - given-names: Väinö
+    family-names: Yrjänäinen
+    alias: ninpnin
+  - family-names: Borges
+    given-names: Robert
+    orcid: "https://orcid.org/0000-0002-7647-4048"
+    alias: BobBorges
+date-released: 2024-04-25
+identifiers:
+  - description: Repository basename
+    type: other
+    value: pyriksdagen
+repository-code: "https://github.com/swerik-project/pyriksdagen"
+url: "https://github.com/swerik-project/the-swedish-parliament-corpus"
+type: dataset
diff --git a/pyriksdagen/metadata.py b/pyriksdagen/metadata.py
@@ -348,52 +348,60 @@ def clean_names(self):
 
 
 
-def load_Corpus_metadata(metadata_folder=None):
+def load_Corpus_metadata(metadata_folder=None, read_db_from=None):
 	"""
 	Populates Corpus object
 	"""
-	if metadata_folder is None:
-		metadata_folder = get_data_location("metadata")
-
-	corpus = Corpus()
-
-	corpus = corpus.add_mps(metadata_folder=metadata_folder)
-	corpus = corpus.add_ministers(metadata_folder=metadata_folder)
-	corpus = corpus.add_speakers(metadata_folder=metadata_folder)
-
-	corpus = corpus.add_persons(metadata_folder=metadata_folder)
-	corpus = corpus.add_location_specifiers(metadata_folder=metadata_folder)
-	corpus = corpus.add_names(metadata_folder=metadata_folder)
-
-	corpus = corpus.impute_dates(metadata_folder=metadata_folder)
-	corpus = corpus.impute_parties(metadata_folder=metadata_folder)
-	corpus = corpus.abbreviate_parties(metadata_folder=metadata_folder)
-	corpus = corpus.add_twitter(metadata_folder=metadata_folder)
-	corpus = corpus.clean_names()
-
-	# Clean up speaker role formatting
-	corpus["role"] = corpus["role"].replace({
-		'Sveriges riksdags talman':'speaker',
-		'andra kammarens andre vice talman':'ak_2_vice_speaker',
-		'andra kammarens förste vice talman':'ak_1_vice_speaker',
-		'andra kammarens talman':'ak_speaker',
-		'andra kammarens vice talman':'ak_1_vice_speaker',
-		'andre vice talman i första kammaren':'fk_2_vice_speaker',
-		'första kammarens talman':'fk_speaker',
-		'första kammarens vice talman':'fk_1_vice_speaker',
-		'förste vice talman i första kammaren':'fk_1_vice_speaker'
-		})
-
-	# Temporary ids
-	corpus['person_id'] = corpus['person_id']
-
-	# Drop individuals with missing names
-	corpus = corpus[corpus['name'].notna()]
-
-	# Remove redundancy and split file
-	corpus = corpus.drop_duplicates()
-	#print( corpus.loc[(pd.isna(corpus['start'])) | (pd.isna(corpus['end']))] )
-	corpus = corpus.dropna(subset=['name', 'start', 'end'])
-	corpus = corpus.sort_values(['person_id', 'start', 'end', 'name'])
+	if read_db_from is not None:
+		print("Reading metadata db from a file.")
+		try:
+			corpus = pd.read_csv(read_db_from)
+		except:
+			corpus = pd.read_pickle(read_db_from)
+	else:
+		print("Compiling metadata db from source.")
+		if metadata_folder is None:
+			metadata_folder = get_data_location("metadata")
+
+		corpus = Corpus()
+
+		corpus = corpus.add_mps(metadata_folder=metadata_folder)
+		corpus = corpus.add_ministers(metadata_folder=metadata_folder)
+		corpus = corpus.add_speakers(metadata_folder=metadata_folder)
+
+		corpus = corpus.add_persons(metadata_folder=metadata_folder)
+		corpus = corpus.add_location_specifiers(metadata_folder=metadata_folder)
+		corpus = corpus.add_names(metadata_folder=metadata_folder)
+
+		corpus = corpus.impute_dates(metadata_folder=metadata_folder)
+		corpus = corpus.impute_parties(metadata_folder=metadata_folder)
+		corpus = corpus.abbreviate_parties(metadata_folder=metadata_folder)
+		corpus = corpus.add_twitter(metadata_folder=metadata_folder)
+		corpus = corpus.clean_names()
+
+		# Clean up speaker role formatting
+		corpus["role"] = corpus["role"].replace({
+			'Sveriges riksdags talman':'speaker',
+			'andra kammarens andre vice talman':'ak_2_vice_speaker',
+			'andra kammarens förste vice talman':'ak_1_vice_speaker',
+			'andra kammarens talman':'ak_speaker',
+			'andra kammarens vice talman':'ak_1_vice_speaker',
+			'andre vice talman i första kammaren':'fk_2_vice_speaker',
+			'första kammarens talman':'fk_speaker',
+			'första kammarens vice talman':'fk_1_vice_speaker',
+			'förste vice talman i första kammaren':'fk_1_vice_speaker'
+			})
+
+		# Temporary ids
+		corpus['person_id'] = corpus['person_id']
+
+		# Drop individuals with missing names
+		corpus = corpus[corpus['name'].notna()]
+
+		# Remove redundancy and split file
+		corpus = corpus.drop_duplicates()
+		#print( corpus.loc[(pd.isna(corpus['start'])) | (pd.isna(corpus['end']))] )
+		corpus = corpus.dropna(subset=['name', 'start', 'end'])
+		corpus = corpus.sort_values(['person_id', 'start', 'end', 'name'])
 
 	return corpus
diff --git a/pyriksdagen/swerik_catalog.py b/pyriksdagen/swerik_catalog.py
@@ -113,7 +113,7 @@ def _identifier_template(auth_code, identifier:str) -> dict:
                     "name": "Swedish Portrait Archive ID"
                 },
             "UpUnAlID": {                                                     # P6821
-                    "formatter-url": "https://www.alvin-portal.org/alvin/view.jsf?pid=alvin-person%{}",
+                    "formatter-url": "https://www.alvin-portal.org/alvin/view.jsf?pid={}",
                     "name": "Uppsala University Alvin ID"
                 },
             "WiDaID": {