| """Build compact-format schema sequence matching paper's HF SFT dataset: |
| table lists , columns = [ |
| lists.list_id | primary key ; type: integer ; values: 1 |
| ... |
| ] |
| foreign keys: |
| ... |
| """ |
| import sys, os |
| sys.path.insert(0, os.path.join(os.path.dirname(__file__), '..')) |
| from utils.db_utils import detect_special_char, add_quotation_mark |
|
|
|
|
| def get_compact_schema(schema, max_values=2): |
| out = "database schema:\n" |
| for table in schema["schema_items"]: |
| tname = table["table_name"] |
| if detect_special_char(tname): tname = add_quotation_mark(tname) |
| col_lines = [] |
| for cname, ctype, ccomment, ccontent, pk, *_ in zip( |
| table["column_names"], table["column_types"], table["column_comments"], |
| table["column_contents"], table["pk_indicators"] |
| ): |
| disp = add_quotation_mark(cname) if detect_special_char(cname) else cname |
| qualified = f"{tname}.{disp}" |
| parts = [] |
| if pk: parts.append("primary key") |
| parts.append(f"type: {ctype}") |
| if ccomment: parts.append(f"meaning: {ccomment}") |
| if ccontent: |
| vals = " , ".join(str(v) for v in ccontent[:max_values]) |
| parts.append(f"values: {vals}") |
| col_lines.append(f" {qualified} | " + " ; ".join(parts)) |
| out += f"table {tname} , columns = [\n" + "\n".join(col_lines) + "\n]\n" |
| if schema.get("foreign_keys"): |
| out += "foreign keys:\n" |
| for fk in schema["foreign_keys"]: |
| fk = [add_quotation_mark(p) if detect_special_char(p) else p for p in fk] |
| out += f"{fk[0]}.{fk[1]} = {fk[2]}.{fk[3]}\n" |
| else: |
| out += "foreign keys: None\n" |
| return out.strip() |
|
|
|
|
| if __name__ == "__main__": |
| import json |
| dev = json.load(open('data/rebuttal_sft_bird_dev_text2sql.json')) |
| s = dev[0] |
| print(get_compact_schema(s['schema'])[:2000]) |
|
|
