[app][fix] dont crash info for total quotes

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "digiflow"
-version = "5.5.8"
+version = "5.5.9"
 description = "Father's Little Digitization Workflow Helper"
 readme = "README.md"
 requires-python = ">=3.8"

src/digiflow/record/record_handler.py

@@ -11,6 +11,7 @@
 
 RECORD_STATE_MASK_FRAME = 'other_load'
 SETSPEC_SPLITTER = '##'
+STRING_QUOTES = "\"'"
 
 
 class RecordHandlerException(Exception):
@@ -375,13 +376,23 @@ def _merge(self_record, other_record):
     self_record[df_r.FIELD_STATE] = other_record[df_r.FIELD_STATE]
     self_record[df_r.FIELD_STATETIME] = other_record[df_r.FIELD_STATETIME]
     try:
-        self_info = ast.literal_eval(self_record[df_r.FIELD_INFO])
-        other_info = ast.literal_eval(other_record[df_r.FIELD_INFO])
+        self_info = ast.literal_eval(_clear_trailing_quotes(self_record[df_r.FIELD_INFO]))
+        other_info = ast.literal_eval(_clear_trailing_quotes(other_record[df_r.FIELD_INFO]))
         self_info.update(other_info)
         self_record[df_r.FIELD_INFO] = str(self_info)
     except (SyntaxError, ValueError):
         self_record[df_r.FIELD_INFO] = other_record[df_r.FIELD_INFO]
 
+def _clear_trailing_quotes(raw_string:str):
+    """Remove evil trailing chars like double/single 
+    quotation marks"""
+
+    if raw_string[0] in STRING_QUOTES:
+        raw_string = raw_string[1:]
+    if raw_string[-1] in STRING_QUOTES:
+        raw_string = raw_string[:-1]
+    return raw_string
+
 
 def _is_unset(self_record):
     if self_record[df_r.FIELD_STATE] == df_r.UNSET_LABEL:

tests/test_digiflow_record_handler.py

@@ -806,7 +806,7 @@ def test_record_handler_merge_info_dicts(tmp_path):
 
     Please note:
     This merge will only work if both INFO fields
-    can the evaluated to dictionaries!
+    can evaluate to dictionaries!
     """
 
     # arrange
@@ -840,4 +840,39 @@ def test_record_handler_merge_info_dicts(tmp_path):
     assert results['ignores'] == 0
     assert results['appendeds'] == 0
     assert dst_hndlr.total_len == 2
-    assert merged_record.info == {'n_ocr': 20, 'pages': 23, 'ods_created':'1984-10-03'}
+    assert merged_record.info == {'n_ocr': 20, 'pages': 23, 'ods_created': '1984-10-03'}
+
+
+def test_record_handler_merge_write_read(tmp_path):
+    """Two lists info field merged too but managed
+    to handle quotations around the info-string.
+    """
+
+    # arrange
+    path_oai_list_a = tmp_path / 'oai_list_a'
+    data_fresh = [
+        "123\tn.a.\t2015-08-25T20:00:35Z\t{'pages':23, 'ods_created':'1984-10-03'}\tu.a.\tn.a.\n"
+    ]
+    write_datalist(path_oai_list_a, data_fresh, LEGACY_HEADER_STR)
+    dst_hndlr = df_r.RecordHandler(
+        path_oai_list_a,
+        data_fields=df_r.LEGACY_HEADER,
+        transform_func=df_r.row_to_record)
+
+    list_merge = tmp_path / 'oai_list_b'
+    data2 = [
+        "123\tn.a.\t2015-08-25T20:00:35Z\t\"{'xml_invalid': \"Element 'mods:subtitle': This element is not expected.\"}\"\tocr_done\t2024-10-18_11:12:00\n",
+    ]
+    write_datalist(list_merge, data2, LEGACY_HEADER_STR)
+
+    # act
+    dst_hndlr.merges(list_merge, dry_run=False)
+    new_hndlr = df_r.RecordHandler(path_oai_list_a,
+                                   data_fields=df_r.LEGACY_HEADER,
+                                   transform_func=df_r.row_to_record)
+
+    # assert
+    tha_record: df_r.Record = new_hndlr.next_record(state='ocr_done')
+    assert tha_record.info == {'pages': 23,
+                               'ods_created': '1984-10-03',
+                               'xml_invalid': "Element 'mods:subtitle': This element is not expected."}