test: add corpus dist tests

DHARPA-Project · Oct 17, 2024 · 056b1cd · 056b1cd
1 parent 6110b01
commit 056b1cd
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 0 deletions.
diff --git a/tests/resources/jobs/init.yaml b/tests/resources/jobs/init.yaml
@@ -3,6 +3,9 @@ inputs:
   load_text_files__path: "${this_dir}/../data/text_corpus/data"
   lccn_metadata__column_name: "file_name"
   lccn_metadata__map: [["sn84037024","sn84037025"],["La Ragione","La Rassegna"]]
+  corpus_distribution__periodicity: "month"
+  corpus_distribution__date_col: "date"
+  corpus_distribution__publication_ref_col: "publication_ref"
   create_array__column_name: "content"
   preprocess_tokens__lowercase: true
   preprocess_tokens__isalpha: true

diff --git a/tests/resources/pipelines/init.yaml b/tests/resources/pipelines/init.yaml
@@ -16,6 +16,10 @@ steps:
       step_id: lccn_metadata
       input_links:
         corpus_table: create_table.table
+    - module_type: topic_modelling.corpus_distribution
+      step_id: corpus_distribution
+      input_links:
+        corpus_table: lccn_metadata.corpus_table
     - module_type: table.pick.column
       step_id: create_array
       input_links: