exact search iteration

app.py

@@ -103,11 +103,17 @@ def process_results(results, highlight_terms):
 
 
 def process_exact_match_payload(payload, query):
+    datasets = set()
     results = payload["results"]
-    results_html = ""
+    results_html = (
+        "<p style='font-family: Arial;'>Total nubmer of results: {}</p>".format(
+            payload["num_results"]
+        )
+    )
     for result in results:
+        _, dataset, _ = result["docid"].split("/")
+        datasets.add(dataset)
         text = result["text"]
-        print(result, text, type(text))
         meta_html = format_meta(result)
 
         query_start = text.find(query)
@@ -125,7 +131,7 @@ def process_exact_match_payload(payload, query):
             )
         )
         results_html += result_html
-    return results_html + "<hr>"
+    return results_html + "<hr>", list(datasets)
 
 
 def process_bm25_match_payload(payload, language):
@@ -144,15 +150,20 @@ def process_bm25_match_payload(payload, language):
     if language == "detect_language":
         return (
             (
-                f"""<p style='font-family: Arial; color:MediumAquaMarine; text-align: center; line-height: 3em'>
+                (
+                    f"""<p style='font-family: Arial; color:MediumAquaMarine; text-align: center; line-height: 3em'>
             Detected language: <b>{results[0]["lang"]}</b></p><br><hr><br>"""
-                if len(results) > 0 and language == "detect_language"
-                else ""
-            )
-            + process_results(results, highlight_terms)
+                    if len(results) > 0 and language == "detect_language"
+                    else ""
+                )
+                + process_results(results, highlight_terms)
+            ),
+            [],
         )
 
     if language == "all":
+        datasets = set()
+        get_docid_html(result["docid"])
         results_html = ""
         for lang, results_for_lang in results.items():
             if len(results_for_lang) == 0:
@@ -168,14 +179,28 @@ def process_bm25_match_payload(payload, language):
                     {process_results(results_for_lang, highlight_terms)}
                 </details>"""
             results_html += collapsible_results
-        return results_html
+            for r in results_for_lang:
+                _, dataset, _ = r["docid"].split("/")
+                datasets.add(dataset)
+        return results_html, list(datasets)
 
-    return process_results(results, highlight_terms)
+    datasets = set()
+    for r in results:
+        _, dataset, _ = r["docid"].split("/")
+        datasets.add(dataset)
+    return process_results(results, highlight_terms), list(datasets)
 
 
-def scisearch(query, language, num_results=10, exact_search=False):
+def scisearch(query, language, num_results=10):
+    datasets = []
     try:
-        query = " ".join(query.split())
+        query = query.strip()
+        exact_search = False
+        if query.startswith('"') and query.endswith('"') and len(query) >= 2:
+            exact_search = True
+            query = query[1:-1]
+        else:
+            query = " ".join(query.split())
         if query == "" or query is None:
             return ""
         post_data = {"query": query, "k": num_results}
@@ -206,7 +231,7 @@ def scisearch(query, language, num_results=10, exact_search=False):
             """
         print(e)
         print(traceback.format_exc())
-    return results_html
+    return results_html, datasets
 
 
 def flag(query, language, num_results, issue_description):
@@ -254,7 +279,7 @@ if __name__ == "__main__":
             query = gr.Textbox(
                 lines=1,
                 max_lines=1,
-                placeholder="Type your query here...",
+                placeholder="Put your query in double quotes for exact search.",
                 label="Query",
             )
         with gr.Row():
@@ -282,18 +307,23 @@ if __name__ == "__main__":
         with gr.Row():
             k = gr.Slider(1, 100, value=10, step=1, label="Max Results")
         with gr.Row():
+            """
             with gr.Column(scale=1):
                 exact_search = gr.Checkbox(
                     value=False, label="Exact Search", variant="compact"
                 )
+            """
             with gr.Column(scale=4):
                 submit_btn = gr.Button("Submit")
+        with gr.Row(visible=False) as datasets_filter:
+            available_datasets = gr.Dropdown(
+                type="value",
+                choices=["ran", "swam", "ate", "slept"],
+                label="Datasets",
+                multiselect=True,
+            )
         with gr.Row():
             results = gr.HTML(label="Results")
-        flag_description = """
-            <p class='flagging'>
-            If you choose to flag your search, we will save the query, language and the number of results you requested.
-            Please consider adding any additional context in the box on the right.</p>"""
         with gr.Column(visible=False) as flagging_form:
             flag_txt = gr.Textbox(
                 lines=1,
@@ -304,25 +334,33 @@ if __name__ == "__main__":
             flag_btn = gr.Button("Flag Results")
             flag_btn.click(flag, inputs=[query, lang, k, flag_txt], outputs=[flag_txt])
 
-        def submit(query, lang, k, exact_search):
-            print("submitting", query, lang, k, exact_search)
+        def submit(query, lang, k, dropdown_input):
+            print("submitting", query, lang, k)
             query = query.strip()
             if query is None or query == "":
                 return "", ""
+            results_html, datasets = scisearch(query, lang, k)
+            print(datasets)
             return {
-                results: scisearch(query, lang, k, exact_search),
+                results: results_html,
                 flagging_form: gr.update(visible=True),
+                datasets_filter: gr.update(visible=True),
+                available_datasets: gr.Dropdown.update(choices=datasets),
             }
 
+        def filter_datasets():
+            pass
+
         query.submit(
             fn=submit,
-            inputs=[query, lang, k, exact_search],
-            outputs=[results, flagging_form],
+            inputs=[query, lang, k, available_datasets],
+            outputs=[results, flagging_form, datasets_filter, available_datasets],
         )
         submit_btn.click(
             submit,
-            inputs=[query, lang, k, exact_search],
-            outputs=[results, flagging_form],
+            inputs=[query, lang, k, available_datasets],
+            outputs=[results, flagging_form, datasets_filter, available_datasets],
         )
 
+        available_datasets.change(filter_datasets, inputs=[], outputs=[])
     demo.launch(enable_queue=True, debug=True)

spaces.code-workspace

@@ -0,0 +1,11 @@
+{
+	"folders": [
+		{
+			"path": ".."
+		},
+		{
+			"path": "../../roots-search-tool"
+		}
+	],
+	"settings": {}
+}