Read gviz column labels as headers

sync.py

@@ -39,7 +39,7 @@ def sheet_names(xlsx: bytes) -> list[str]:
 
 
 def read_gviz(sheet: str) -> list[list[str]]:
-    params = urlencode({"sheet": sheet, "headers": "0"})
+    params = urlencode({"sheet": sheet})
     url = f"https://docs.google.com/spreadsheets/d/{WORKBOOK_ID}/gviz/tq?{params}"
     response = requests.get(url, timeout=30)
     response.raise_for_status()
@@ -50,32 +50,56 @@ def read_gviz(sheet: str) -> list[list[str]]:
         text = text[:-1].rstrip()
     if not text.startswith(GVIZ_PREFIX) or not text.endswith(")"):
         raise ValueError(f"invalid gviz response for worksheet {sheet!r}")
-    payload = json.loads(text[len(GVIZ_PREFIX) : -1])
-    if payload.get("status") != "ok":
+    try:
+        payload = json.loads(text[len(GVIZ_PREFIX) : -1])
+    except json.JSONDecodeError as error:
+        raise ValueError(f"invalid gviz response for worksheet {sheet!r}") from error
+    if not isinstance(payload, dict) or payload.get("status") != "ok":
         raise ValueError(f"gviz query failed for worksheet {sheet!r}")
-    table = payload.get("table", {})
-    columns = table.get("cols", [])
-    rows = table.get("rows", [])
-    width = len(columns)
+    table = payload.get("table")
+    if not isinstance(table, dict):
+        raise ValueError(f"invalid gviz table for worksheet {sheet!r}")
+    columns = table.get("cols")
+    rows = table.get("rows")
+    if not isinstance(columns, list) or not isinstance(rows, list):
+        raise ValueError(f"invalid gviz table for worksheet {sheet!r}")
+
+    headers: list[str] = []
+    for column in columns:
+        if not isinstance(column, dict):
+            raise ValueError(f"invalid gviz column for worksheet {sheet!r}")
+        label = column.get("label")
+        if label is not None and not isinstance(label, str):
+            raise ValueError(f"invalid gviz column label for worksheet {sheet!r}")
+        headers.append(label or "")
+
+    while headers and headers[-1] == "":
+        headers.pop()
+    if not headers:
+        return []
 
     def cell_value(cell: dict | None) -> str:
-        if not cell:
+        if cell is None:
             return ""
-        value = cell.get("f") if cell.get("f") is not None else cell.get("v", "")
+        if not isinstance(cell, dict):
+            raise ValueError(f"invalid gviz cell for worksheet {sheet!r}")
+        value = cell.get("f") if cell.get("f") is not None else cell.get("v")
         if value is None:
             return ""
         if isinstance(value, bool):
             return "TRUE" if value else "FALSE"
         return str(value)
 
-    result: list[list[str]] = []
+    data_rows: list[list[str]] = []
     for row in rows:
-        cells = row.get("c", [])
-        if len(cells) > width:
+        if not isinstance(row, dict) or not isinstance(row.get("c", []), list):
+            raise ValueError(f"invalid gviz row for worksheet {sheet!r}")
+        cells = row["c"]
+        if len(cells) > len(columns):
             raise ValueError(f"gviz row is wider than columns in worksheet {sheet!r}")
         values = [cell_value(cell) for cell in cells]
-        result.append(values + [""] * (width - len(values)))
-    return result
+        data_rows.append((values + [""] * (len(columns) - len(values)))[: len(headers)])
+    return [headers] + data_rows
 
 
 def normalize_sheet(name: str, rows: list[list[str]]) -> tuple[list[str], list[list[str]]]: