Filling XLSX: Everything except multiple choice.

Summary

For caosdb/customers/dimr/management#219:

Filling XLSX templates from JSON works now.

Focus

• Correct handling of foreign keys
• Correct handling of other lists

Test Environment

Run the unit tests.

Check List for the Author

Please, prepare your MR for a review. Be sure to write a summary and a focus and create gitlab comments for the reviewer. They should guide the reviewer through the changes, explain your changes and also point out open questions. For further good practices have a look at our review guidelines

• All automated tests pass
• Reference related issues
• Up-to-date CHANGELOG.md (or not necessary)
• Up-to-date JSON schema (or not necessary)
• Appropriate user and developer documentation (or not necessary)
  • How do I use the software? Assume "stupid" users.
  • How do I develop or debug the software? Assume novice developers.
• Annotations in code (Gitlab comments)
  • Intent of new code
  • Problems with old code
  • Why this implementation?

Check List for the Reviewer

• I understand the intent of this MR
• All automated tests pass
• Up-to-date CHANGELOG.md (or not necessary)
• Appropriate user and developer documentation (or not necessary)
• The test environment setup works and the intended behavior is reproducible in the test environment
• In-code documentation and comments are up-to-date.
• Check: Are there specifications? Are they satisfied?

For further good practices have a look at our review guidelines.

src/caosadvancedtools/table_json_conversion/fill_xlsx.py

@@ -26,10 +26,13 @@ import json
 import pathlib
 from collections import OrderedDict
 from types import SimpleNamespace
-from typing import Any, Dict, List, Optional, Union, TextIO
+from typing import Any, Dict, List, Optional, TextIO, Union
+from warnings import warn
 
 from jsonschema import FormatChecker, validate
-from openpyxl import load_workbook, Workbook
+from jsonschema.exceptions import ValidationError
+from openpyxl import Workbook, load_workbook
+from openpyxl.cell.cell import ILLEGAL_CHARACTERS_RE
 from openpyxl.worksheet.worksheet import Worksheet
 
 from .table_generator import ColumnType, RowType
@@ -53,11 +56,10 @@ def _get_column_types(sheet: Worksheet) -> OrderedDict:
     type_row_index = _get_row_type_column_index(sheet)
     for idx, col in enumerate(sheet.columns):
         type_cell = col[type_row_index]
-        result[idx] = type_cell.value
-        assert (hasattr(ColumnType, type_cell.value)
-                or type_cell.value == RowType.COL_TYPE.name
-                or type_cell.value is None), (
-            f"Unexpected column type value: {type_cell.value}")
+        result[idx] = type_cell.value if type_cell.value is not None else ColumnType.IGNORE.name
+        assert (hasattr(ColumnType, result[idx])
+                or result[idx] == RowType.COL_TYPE.name), (
+            f"Unexpected column type value ({idx}{type_row_index}): {type_cell.value}")
     return result
 
 
@@ -134,8 +136,9 @@ def _read_or_dict(data: Union[dict, str, TextIO]) -> dict:
 class TemplateFiller:
     """Class to fill XLSX templates.  Has an index for all relevant columns."""
 
-    def __init__(self, workbook: Workbook):
+    def __init__(self, workbook: Workbook, graceful: bool = False):
         self._workbook = workbook
+        self._graceful = graceful
         self._create_index()
 
     @property
@@ -183,6 +186,7 @@ class TemplateFiller:
             self._props[fullpath] = value
 
         def fill_from_data(self, data: Dict[str, Any]):
+            # TODO recursive for dicts and list?
             """Fill current level with all scalar elements of ``data``."""
             for name, value in data.items():
                 if not isinstance(value, (dict, list)):
@@ -272,20 +276,24 @@ out: union[dict, None]
 
         insertables: Dict[str, Any] = {}
         for name, content in data.items():
+            # TODO is this the best way to do it????
+            if name == "file":
+                continue
             path = current_path + [name]
             next_context = context.next_level(name)
             # preprocessing
             if isinstance(content, list):
-                if not content:
+                if not content:  # empty list
                     continue
-                # Must be all of the same type.
+                # List elements must be all of the same type.
                 assert len(set(type(entry) for entry in content)) == 1
-                if isinstance(content[0], dict):
+
+                if isinstance(content[0], dict):  # all elements are dicts
                     # An array of objects: must go into exploded sheet
                     for entry in content:
                         self._handle_data(data=entry, current_path=path, context=next_context)
                     continue
-            elif isinstance(content, dict):
+            elif isinstance(content, dict):  # we recurse and simply use the result
                 if not current_path:  # Special handling for top level
                     self._handle_data(content, current_path=path, context=next_context)
                     continue
@@ -296,26 +304,27 @@ out: union[dict, None]
                 insertables.update(insert)
                 continue
             else:  # scalars
-                content = [content]
+                content = [content]  # make list for unified treatment below
 
             # collecting the data
             assert isinstance(content, list)
-            if len(content) == 1:
-                value = content[0]
-            else:
-                value = ";".join(content)
+            content = [str(x) for x in content]
+            value = ";".join(content)  # TODO we need escaping of values
             path_str = p2s(path)
             assert path_str not in insertables
             insertables[path_str] = value
         if only_collect_insertables:
             return insertables
-        if not current_path:
+        if not current_path:  # Top level returns, because there are only sheets for the children.
             return None
 
         # actual data insertion
         insert_row = None
         sheet = None
         for path_str, value in insertables.items():
+            if self._graceful and path_str not in self._sheet_index:
+                warn(f"Ignoring path with missing sheet index: {path_str}")
+                continue
             sheet_meta = self._sheet_index[path_str]
             if sheet is None:
                 sheet = sheet_meta.sheet
@@ -324,14 +333,20 @@ out: union[dict, None]
             if insert_row is None:
                 insert_row = _next_row_index(sheet)
 
-            sheet.cell(row=insert_row+1, column=col_index+1, value=value)
+            sheet.cell(row=insert_row+1, column=col_index+1,
+                       value=ILLEGAL_CHARACTERS_RE.sub("", value))
 
         # Insert foreign keys
         if insert_row is not None and sheet is not None and _is_exploded_sheet(sheet):
-            foreigns = _get_foreign_key_columns(sheet)
+            try:
+                foreigns = _get_foreign_key_columns(sheet)
+            except ValueError:
+                print(f"Sheet: {sheet}")
+                raise
             for index, path in ((f.index, f.path) for f in foreigns.values()):
                 value = context[path]
-                sheet.cell(row=insert_row+1, column=index+1, value=value)
+                sheet.cell(row=insert_row+1, column=index+1,
+                           value=ILLEGAL_CHARACTERS_RE.sub("", value))
 
         return None
 
@@ -353,7 +368,8 @@ result: str
   Path for the result XLSX.
 validation_schema: dict, optional
   If given, validate the date against this schema first.  This raises an exception if the validation
-  fails.
+  fails.  If no validation schema is given, try to ignore more errors in the data when filling the
+  XLSX template.
 """
     data = _read_or_dict(data)
     assert isinstance(data, dict)
@@ -361,11 +377,17 @@ validation_schema: dict, optional
     # Validation
     if validation_schema is not None:
         validation_schema = _read_or_dict(validation_schema)
-        validate(data, validation_schema, format_checker=FormatChecker())
+        try:
+            validate(data, validation_schema, format_checker=FormatChecker())
+        except ValidationError as ve:
+            print(ve.message)
+            raise RuntimeError("Validation failed")
+    else:
+        print("No validation schema given, continue at your own risk.")
 
     # Filling the data
     result_wb = load_workbook(template)
-    template_filler = TemplateFiller(result_wb)
+    template_filler = TemplateFiller(result_wb, graceful=(validation_schema is None))
     template_filler.fill_data(data=data)
 
     parentpath = pathlib.Path(result).parent