filtering by sub_ids because I want to do habitat type queries

app/app.py

@@ -95,7 +95,7 @@ def main():
         sql_query = output.sql_query
         explanation =output.explanation
         if not sql_query: # if the chatbot can't generate a SQL query.
-            return pd.DataFrame({'id' : []}),'', explanation
+            return pd.DataFrame({'sub_id' : []}),'', explanation
         result = ca.sql(sql_query).execute()
         if result.empty:
             explanation = "This query did not return any results. Please try again with a different query."
@@ -139,7 +139,7 @@ def main():
                                 st.stop()
                                 
                             # output without mapping columns (id, geom)
-                            elif "id" not in llm_output.columns and "geom" not in llm_output.columns:
+                            elif "sub_id" not in llm_output.columns and "geom" not in llm_output.columns:
                                 st.write(llm_output)
                                 not_mapping = True
         
@@ -151,8 +151,8 @@ def main():
                                     st.code(sql_query,language = "sql") 
                                     
                         # extract ids, columns, bounds if present
-                        if "id" in llm_output.columns and not llm_output.empty:
-                            ids = list(set(llm_output['id'].tolist()))
+                        if "sub_id" in llm_output.columns and not llm_output.empty:
+                            ids = list(set(llm_output['sub_id'].tolist()))
                             llm_cols = extract_columns(sql_query)
                             bounds = llm_output.total_bounds.tolist()
                         else:
@@ -390,12 +390,12 @@ def main():
                     if ('geom' in llm_output.columns) and (not llm_output.empty):
                         llm_output = llm_output.drop('geom',axis = 1)
                     if not llm_output.empty:
-                        if 'name' in llm_output.columns and 'id' in llm_output.columns:
+                        if 'name' in llm_output.columns and 'sub_id' in llm_output.columns:
                             llm_grouped = (llm_output.groupby('name')
                                             .agg({col: ('sum' if col == 'acres' else 'first') 
                                               for col in llm_output.columns 
                                               if col != 'name'})).reset_index()
-                            llm_grouped.drop(['id'], axis=1, inplace = True)
+                            llm_grouped.drop(['sub_id'], axis=1, inplace = True)
                             st.dataframe(llm_grouped, use_container_width = True)
                         else:
                             st.dataframe(llm_output, use_container_width = True)

app/system_prompt.txt

@@ -32,7 +32,7 @@ Ensure the response contains only this JSON object, with no additional text, for
     - Unless the users asks about biodiversity overlap or coverage, use columns with the prefix "mean_" to quantify biodiversity. 
 
 # Column Descriptions
-- "id": The parent id for sub_id. "id" identifes a protected area, while "sub_id" identifies the particular feature in a protected area. "id" is necessary for displaying queried results on a map.  
+- "id": The parent id for sub_id. "id" identifes a protected area, while "sub_id" identifies the particular feature in a protected area. "sub_id" is necessary for displaying queried results on a map.  
 - "sub_id": unique string identifier for each feature. 
 - "established": The time range which the land was acquired, either "2024" or "pre-2024". 
 - "gap_code": The GAP code corresponds to the level of biodiversity protection for an area; GAP 1 has the highest protections whereas GAP 4 has the weakest. There are 4 gap codes and are defined as the following. In the data, non-conservation areas do not have a GAP code. 
@@ -73,7 +73,7 @@ Only use the following table:
 ## Example: 
 example_user: "Show me the best areas to go birdwatching in San Diego County."
 example_assistant: {{"sql_query":    
-    SELECT "id", "geom", "name", "acres", "county","mean_bird_richness"
+    SELECT "sub_id", "geom", "name", "acres", "county","mean_bird_richness"
     FROM mydata
     WHERE "county" = 'San Diego'
     ORDER BY "mean_bird_richness" DESC
@@ -90,7 +90,7 @@ limit the uses of private land to protect conservation values.
 ## Example:
 example_user: "Show me all non-profit land."
 example_assistant: {{"sql_query": 
-    SELECT "id", "geom", "name", "acres", "manager_type"
+    SELECT "sub_id", "geom", "name", "acres", "manager_type"
     FROM mydata 
     WHERE "manager_type" = 'Non Profit';
 "explanation":"I selected all data where `manager_type` is 'Non Profit'."
@@ -99,7 +99,7 @@ example_assistant: {{"sql_query":
 ## Example: 
 example_user: "Where are areas with high plant biodiversity"
 example_assistant: {{"sql_query":    
-    SELECT "id", "geom", "name", "acres", "mean_plant_richness"
+    SELECT "sub_id", "geom", "name", "acres", "mean_plant_richness"
     FROM mydata
     ORDER BY "mean_plant_richness" DESC
     LIMIT 50;
@@ -107,7 +107,7 @@ example_assistant: {{"sql_query":
 
 example_user: "Show me areas open to the public in disadvantaged communities."
 example_assistant: {{"sql_query":   
-    SELECT "id", "geom", "name", "acres", "access_type", "pct_disadvantaged_community" FROM mydata 
+    SELECT "sub_id", "geom", "name", "acres", "access_type", "pct_disadvantaged_community" FROM mydata 
     WHERE "access_type" = 'Open Access'
     WHERE "pct_disadvantaged_community" > 0;
 "explanation": "I used `access_type` to filter for areas that are open to the public (`Open Access`) and `pct_disadvantaged_community` > 0 to include only those located in disadvantaged communities. `pct_disadvantaged_community` identifies communities burdened by multiple sources of pollution with population characteristics that make them more sensitive to pollution."
@@ -141,13 +141,13 @@ sql_query:
 ## Example: 
 example_user: "Show me all land managed by the United States Forest Service."
 sql_query:  
-    SELECT "id", "geom", "name", "acres", "manager" FROM mydata 
+    SELECT "sub_id", "geom", "name", "acres", "manager" FROM mydata 
     WHERE LOWER("manager") LIKE '%united states forest service%';
 
 ## Example:
 example_user: "Show me areas with more than 25% overlap in bird species richness."
 sql_query:
-    SELECT "id", "geom", "name", "acres", "pct_bird_richness"
+    SELECT "sub_id", "geom", "name", "acres", "pct_bird_richness"
     FROM mydata
       AND (
         "pct_bird_richness" > 0.25 
@@ -165,7 +165,7 @@ sql_query:
 
 example_user: "Show me GAP 3 lands where more than 50% of the area overlaps with regions of high biodiversity."
 example_assistant: {{"sql_query":    
-    SELECT "id", "geom", "name", "acres", "county",
+    SELECT "sub_id", "geom", "name", "acres", "county",
            "pct_top_amphibian_richness", "pct_top_reptile_richness",
            "pct_top_bird_richness", "pct_top_mammal_richness",
            "pct_top_freshwater_richness", "pct_top_plant_richness"
@@ -186,21 +186,21 @@ example_assistant: {{"sql_query":
         SELECT PERCENTILE_CONT(0.85) WITHIN GROUP (ORDER BY "mean_amphibian_richness") AS mean_amphibian_richness_85_percentile
         FROM mydata
     )
-    SELECT "id", "geom", "name", "acres","mean_amphibian_richness"
+    SELECT "sub_id", "geom", "name", "acres","mean_amphibian_richness"
     FROM mydata
     WHERE "land_tenure" = 'Easement'
     AND "mean_amphibian_richness" >= (SELECT mean_amphibian_richness_85_percentile FROM percentile);
 
 example_user: "Show nonconserved areas in climate zone 2"
 example_assistant: {{"sql_query":    
-    SELECT "id", "geom", "name", "acres", "status", "climate_zone"
+    SELECT "sub_id", "geom", "name", "acres", "status", "climate_zone"
     FROM mydata
     WHERE "climate_zone" = 'Zone 2'
     AND "status" = 'Non-Conservation Area';
 
 example_user: "Show me working lands that 30x30 conservation areas"
 example_assistant: {{"sql_query":    
-    SELECT "id", "geom", "name", "acres", "land_tenure", "pct_farmland"
+    SELECT "sub_id", "geom", "name", "acres", "land_tenure", "pct_farmland"
     FROM mydata
     WHERE "status" = '30x30 Conservation Area'
     AND "pct_farmland" >0;
@@ -211,7 +211,7 @@ example_assistant: {{"sql_query":
         SELECT PERCENTILE_CONT(0.95) WITHIN GROUP (ORDER BY "mean_amphibian_richness") AS mean_mammal_richness_95_percentile
         FROM mydata
     )
-    SELECT "id", "geom", "name", "acres","mean_mammal_richness"
+    SELECT "sub_id", "geom", "name", "acres","mean_mammal_richness"
     FROM mydata
     WHERE "status" != '30x30 Conservation Area'
     AND "mean_mammal_richness" >= (SELECT mean_amphibian_richness_95_percentile FROM percentile);

app/utils.py

@@ -184,7 +184,7 @@ def get_summary_table_sql(ca, column, colors, ids, feature_col = None):
     """
     Generates a summary table using specific IDs as filters.
     """
-    combined_filter = _.id.isin(ids)
+    combined_filter = _.sub_id.isin(ids)
     df_network = get_summary(ca, combined_filter, [column], column, feature_col, colors)
     df_feature = get_summary(ca, combined_filter, [column], column, feature_col, colors, feature = True)
     return df_network, df_feature
@@ -206,7 +206,7 @@ def get_pmtiles_style(paint, pmtiles_file, low_res, filter_cols=None, filter_val
     Generates a MapLibre GL style for PMTiles with either filters or a list of IDs.
     """
     if ids:
-        filter_expr = ["in", ["get", "id"], ["literal", ids]]
+        filter_expr = ["in", ["get", "sub_id"], ["literal", ids]]
     else:
         # we don't want to overwhelm streamlit so if they didn't filter anything, don't provide filter arg 
         filter_length = sum([len(x) for x in filter_vals])

app/variables.py

@@ -463,8 +463,8 @@ help_message = '''
 
 example_queries = """
 Mapping queries:
-- Show me bird biodiversity hotspots not covered by the 30x30 network.
-- Show me GAP 3 lands with mean amphibian richness in the top 10%.
+- Show me amphibian biodiversity hotspots not covered by the 30x30 network.
+- Show me GAP 3 lands with mean bird richness in the top 10%.
 - Show me easements with 60% or more overlap with high plant biodiversity regions.
 - Show me protected areas that are open to the public in disadvantaged communities. 
 - Show me all 30x30 conservation areas managed by The Nature Conservancy.