fix(excel2json): find hidden worksheets in excel (DEV-1483) (#679)

src/dsp_tools/commands/excel2json/input_error.py

@@ -168,6 +168,22 @@ def execute_error_protocol(self) -> str:
         return msg
 
 
+@dataclass(frozen=True)
+class MoreThanOneSheetProblem:
+    """This class contains information if the excel containing the property values has more than one sheet."""
+
+    excelname: str
+    sheet_names: list[str]
+
+    def __str__(self) -> str:
+        msg = [
+            f"\nIn the '{self.excelname}' file only one sheet is allowed.",
+            f"The excel used contains the following sheets:{list_separator}{list_separator.join(self.sheet_names)}",
+            "Please delete all but one sheet.",
+        ]
+        return separator.join(msg)
+
+
 @dataclass(frozen=True)
 class JsonValidationPropertyProblem:
     """This class contains information about a JSON property section that fails its validation against the schema."""

src/dsp_tools/commands/excel2json/lists.py

@@ -10,7 +10,8 @@
 from openpyxl.cell import Cell
 from openpyxl.worksheet.worksheet import Worksheet
 
-from dsp_tools.models.exceptions import BaseError, UserError
+from dsp_tools.commands.excel2json.input_error import MoreThanOneSheetProblem
+from dsp_tools.models.exceptions import BaseError, InputError, UserError
 from dsp_tools.utils.shared import simplify_name
 
 list_of_lists_of_previous_cell_values: list[list[str]] = []
@@ -271,7 +272,7 @@ def _make_json_lists_from_excel(
     startcol = 1
 
     # make a dict with the language labels and the worksheets
-    lang_to_worksheet = {x.stem: load_workbook(x, read_only=True).worksheets[0] for x in excel_file_paths}
+    lang_to_worksheet: dict[str, Worksheet] = {x.stem: _read_and_check_workbook(x) for x in excel_file_paths}
 
     # take English as base file. If English is not available, take a random one.
     base_lang = "en" if "en" in lang_to_worksheet else list(lang_to_worksheet.keys())[0]
@@ -304,6 +305,14 @@ def _make_json_lists_from_excel(
     return finished_lists
 
 
+def _read_and_check_workbook(excelpath: Path) -> Worksheet:
+    all_worksheets = load_workbook(excelpath, read_only=True).worksheets
+    if len(all_worksheets) != 1:
+        msg = str(MoreThanOneSheetProblem(excelpath.name, [x.title for x in all_worksheets]))
+        raise InputError(msg)
+    return all_worksheets[0]
+
+
 def validate_lists_section_with_schema(
     path_to_json_project_file: Optional[str] = None,
     lists_section: Optional[list[dict[str, Any]]] = None,

src/dsp_tools/commands/excel2json/properties.py

@@ -13,6 +13,7 @@
     InvalidExcelContentProblem,
     JsonValidationPropertyProblem,
     MissingValuesInRowProblem,
+    MoreThanOneSheetProblem,
     PositionInExcel,
     Problem,
 )
@@ -26,7 +27,7 @@
     get_comments,
     get_labels,
     get_wrong_row_numbers,
-    read_and_clean_excel_file,
+    read_and_clean_all_sheets,
 )
 from dsp_tools.models.exceptions import InputError
 
@@ -449,7 +450,8 @@ def excel2properties(
         a tuple consisting of the "properties" section as a Python list,
             and the success status (True if everything went well)
     """
-    property_df = read_and_clean_excel_file(excelfile=excelfile)
+
+    property_df = _read_check_property_df(excelfile)
 
     property_df = _rename_deprecated_columnnames(df=property_df, excelfile=excelfile)
 
@@ -490,3 +492,11 @@ def excel2properties(
             print(f"properties section was created successfully and written to file '{path_to_output_file}'")
 
     return props, True
+
+
+def _read_check_property_df(excelfile: str) -> pd.DataFrame | None:
+    sheets_df_dict = read_and_clean_all_sheets(excelfile=excelfile)
+    if len(sheets_df_dict) != 1:
+        msg = str(MoreThanOneSheetProblem("properties.xlsx", list(sheets_df_dict.keys())))
+        raise InputError(msg)
+    return list(sheets_df_dict.values())[0]

src/dsp_tools/commands/excel2json/utils.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from pathlib import Path
 from typing import Any
 from unittest import mock
 
@@ -44,7 +45,7 @@ def read_and_clean_excel_file(excelfile: str, sheetname: str | int = 0) -> pd.Da
     return read_df
 
 
-def read_and_clean_all_sheets(excelfile: str) -> dict[str, pd.DataFrame]:
+def read_and_clean_all_sheets(excelfile: str | Path) -> dict[str, pd.DataFrame]:
     """
     This function reads an Excel file with all its sheets.
     If there is a ValueError, it patches the openpyxl part that causes the error
@@ -73,7 +74,7 @@ def read_and_clean_all_sheets(excelfile: str) -> dict[str, pd.DataFrame]:
     try:
         return {name.strip(""): clean_data_frame(df) for name, df in df_dict.items()}
     except AttributeError:
-        msg = InvalidSheetNameProblem(excelfile, list(df_dict.keys())).execute_error_protocol()
+        msg = InvalidSheetNameProblem(str(excelfile), list(df_dict.keys())).execute_error_protocol()
         raise InputError(msg) from None