Upload 6 files

app.py

@@ -1,22 +1,12 @@
-import gradio as gr
-
-def oracle_plsql_autocomplete(prompt):
-    suggestions = {
-        "select": "SELECT * FROM employees WHERE ROWNUM <= 10;",
-        "insert": "INSERT INTO employees (emp_id, name) VALUES (101, 'John');",
-        "create": "CREATE TABLE departments (dept_id NUMBER, dept_name VARCHAR2(50));",
-        "begin": "BEGIN\n  NULL;\nEND;",
-        "if": "IF salary > 10000 THEN\n  DBMS_OUTPUT.PUT_LINE('High salary');\nEND IF;"
-    }
-    key = prompt.lower().strip().split()[0]
-    return suggestions.get(key, "-- No matching Oracle PL/SQL snippet found.")
-
-interface = gr.Interface(
-    fn=oracle_plsql_autocomplete,
-    inputs=gr.Textbox(lines=2, placeholder="Start typing PL/SQL... (e.g. select, insert)"),
-    outputs=gr.Textbox(),
-    title="Oracle LLM Autocomplete (Mock)",
-    description="Start typing a PL/SQL keyword and get a suggested code snippet."
-)
-
-interface.launch()
+import gradio as gr
+from model import oracle_sql_suggester
+
+interface = gr.Interface(
+    fn=oracle_sql_suggester,
+    inputs=gr.Textbox(lines=3, placeholder="Describe the SQL query you want..."),
+    outputs=gr.Textbox(),
+    title="Oracle SQL Generator (Phase 1 - Rule Based)",
+    description="Describe your requirement in text, get a suggested SQL."
+)
+
+interface.launch()

data_loader.py

@@ -0,0 +1,10 @@
+# data_loader.py
+
+def load_rules(file_path="data/train_data.txt"):
+    data = {}
+    with open(file_path, "r", encoding="utf-8") as file:
+        for line in file:
+            if "=" in line:
+                key, value = line.strip().split("=", 1)
+                data[key.strip().lower()] = value.strip()
+    return data

model.py

@@ -0,0 +1,50 @@
+# model.py
+from sentence_transformers import SentenceTransformer, util
+from sql_templates import sql_templates  # new import
+import torch
+
+# Load training rules (string-to-SQL map)
+from data_loader import load_rules  # you can split this for cleanliness
+rules = load_rules()
+
+# Load embedding model (lightweight, fast)
+model = SentenceTransformer("sentence-transformers/paraphrase-MiniLM-L6-v2")
+
+# Pre-compute embeddings of training prompts
+train_prompts = list(rules.keys())
+train_embeddings = model.encode(train_prompts, convert_to_tensor=True)
+
+def oracle_sql_suggester(prompt):
+    prompt_clean = prompt.strip().lower()
+    
+    # Try direct rule match
+    if prompt_clean in rules:
+        return rules[prompt_clean]
+    
+    # Semantic matching
+    user_embedding = model.encode(prompt_clean, convert_to_tensor=True)
+    cosine_scores = util.cos_sim(user_embedding, train_embeddings)
+    
+    top_match_index = torch.argmax(cosine_scores).item()
+    top_score = cosine_scores[0][top_match_index].item()
+    
+    if top_score >= 0.7:
+        matched_prompt = train_prompts[top_match_index]
+        return rules[matched_prompt]
+    
+     # Check template keywords
+    for key in sql_templates:
+        if key.replace("_", " ") in prompt_clean or key in prompt_clean:
+            return sql_templates[key]
+
+    # Semantic match
+    user_embedding = model.encode(prompt_clean, convert_to_tensor=True)
+    cosine_scores = util.cos_sim(user_embedding, train_embeddings)
+    top_match_index = torch.argmax(cosine_scores).item()
+    top_score = cosine_scores[0][top_match_index].item()
+
+    if top_score >= 0.7:
+        matched_prompt = train_prompts[top_match_index]
+        return rules[matched_prompt]
+    
+    return "🤖 Sorry, I couldn’t understand that. Please try rephrasing your request."

my_gradio_demo.py

@@ -0,0 +1,22 @@
+import gradio as gr
+
+def oracle_plsql_autocomplete(prompt):
+    suggestions = {
+        "select": "SELECT * FROM employees WHERE ROWNUM <= 10;",
+        "insert": "INSERT INTO employees (emp_id, name) VALUES (101, 'John');",
+        "create": "CREATE TABLE departments (dept_id NUMBER, dept_name VARCHAR2(50));",
+        "begin": "BEGIN\n  NULL;\nEND;",
+        "if": "IF salary > 10000 THEN\n  DBMS_OUTPUT.PUT_LINE('High salary');\nEND IF;"
+    }
+    key = prompt.lower().strip().split()[0]
+    return suggestions.get(key, "-- No matching Oracle PL/SQL snippet found.")
+
+interface = gr.Interface(
+    fn=oracle_plsql_autocomplete,
+    inputs=gr.Textbox(lines=2, placeholder="Start typing PL/SQL... (e.g. select, insert)"),
+    outputs=gr.Textbox(),
+    title="Oracle LLM Autocomplete (Mock)",
+    description="Start typing a PL/SQL keyword and get a suggested code snippet."
+)
+
+interface.launch()

requirements.txt

@@ -0,0 +1,3 @@
+gradio
+torch
+sentence-transformers

sql_templates.py

@@ -0,0 +1,10 @@
+sql_templates = {
+    "basic_select": "SELECT column1, column2 FROM table_name;",
+    "select_where": "SELECT column1 FROM table_name WHERE condition;",
+    "join_example": "SELECT a.col1, b.col2 FROM table1 a JOIN table2 b ON a.id = b.a_id;",
+    "group_by": "SELECT dept, COUNT(*) FROM employees GROUP BY dept;",
+    "having": "SELECT dept, COUNT(*) FROM employees GROUP BY dept HAVING COUNT(*) > 5;",
+    "insert": "INSERT INTO table_name (col1, col2) VALUES (val1, val2);",
+    "update": "UPDATE table_name SET col1 = val1 WHERE condition;",
+    "delete": "DELETE FROM table_name WHERE condition;"
+}