add formatting

app.py

@@ -35,7 +35,7 @@ st.set_page_config(
 page_description = """
 # GEM Data Card Input Form
 
-This application was designed to support the GEM v2 data hackathon.
+This application was designed for GEM.
 
 It allows users to fill out all of the information going into the data documentation when submitting a new dataset.
 
@@ -43,6 +43,8 @@ Use the left sidebar to navigate:
 - "**Dataset at a Glance**" shows selected information and tracks progress
 - Each of the "**Section:**" pages opens a form for a specific section of the card
 - Go to "**Review and Save**" when you are done to save your data card
+
+Please use **markdown** formatting for each field.
 """
 
 _N_FIELDS = datacards.considerations.N_FIELDS + \

datacards/overview.py

@@ -34,7 +34,7 @@ def overview_page():
     st.session_state.card_dict["overview"] = st.session_state.card_dict.get(
         "overview", {}
     )
-    with st.expander("Where to find the data and its documentation", expanded=True):
+    with st.expander("What is this dataset?", expanded=True):
         key_pref = ["overview", "what"]
         st.session_state.card_dict["overview"]["what"] = st.session_state.card_dict[
             "overview"

formatting/key_to_question.json

@@ -1,6 +1,17 @@
 {
   "overview": {
     "section-title": "Dataset Overview",
+    "what": {
+      "section-title": "What is this dataset?",
+      "dataset": {
+        "title": "Summary",
+        "info": "Provide a summary of this dataset in 3-4 sentences.",
+        "flags": [
+          "quick"
+        ],
+        "scope": "telescope"
+      }
+    },
     "where": {
       "section-title": "Where to find the Data and its Documentation",
       "website": {

formatting/reformat_json.py

@@ -52,10 +52,16 @@ def construct_json(dataset_name: str, data_card_data: dict, text_by_key: dict):
     authors = data_card_data["overview"]["credit"]["creators"]
   except KeyError:
     authors = ""
+  try:
+    summary = data_card_data["overview"]["what"]["dataset"]
+  except KeyError:
+    summary = "Placeholder"
+
+
 
   new_json = {
       "name": dataset_name,
-      "summary": "Placeholder",
+      "summary": summary,
       "sections": [
       ],
   }
@@ -94,9 +100,10 @@ def construct_json(dataset_name: str, data_card_data: dict, text_by_key: dict):
           total_questions += 1
           answer = data_card_data[main_key][second_key].get(final_key, "N/A")
         except:
-          print(main_key, second_key, final_key)
-          print("==="*50)
-          print(data_card_data)
+          # print(main_key, second_key, final_key)
+          # print("==="*50)
+          # print(data_card_data)
+          continue
         # Skip empty answers.
         if isinstance(answer, str):
           if answer.lower() == "n/a":
@@ -140,6 +147,7 @@ if __name__ == "__main__":
       data_card_data = read_json_file(data_card_path)
       data_card_json, total_cur_words = construct_json(dataset, data_card_data, text_by_key)
       total_words_across_everything += total_cur_words
+
       save_file(new_path, data_card_json)
     else:
       print(f"{dataset} has no data card!")