airbyte-source-google-sheets 0.9.5.dev202505142036__py3-none-any.whl → 0.10.0.dev202505231635__py3-none-any.whl

{airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info → airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: airbyte-source-google-sheets
-Version: 0.9.5.dev202505142036
+Version: 0.10.0.dev202505231635
 Summary: Source implementation for Google Sheets.
 License: Elv2
 Author: Airbyte

{airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info → airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info}/RECORD

@@ -1,6 +1,6 @@
 source_google_sheets/__init__.py,sha256=dYaZS0KrTjksk_yeSuXmNDXDsNKWctFnlZJSMgLryXE,135
 source_google_sheets/components/__init__.py,sha256=v7odPaLdz2S_NRzExsxkk6694Vmjbplz4Z5mA4lxdkA,343
-source_google_sheets/components/extractors.py,sha256=S7j78aZsoxYOj2ugCMWuXmphu2Eu8YfmYD3Xu_2CPAc,8832
+source_google_sheets/components/extractors.py,sha256=-d5hPW40ZBjU3gTUZNudayvqf1eHlveW1amwKJywDkU,9668
 source_google_sheets/components/partition_routers.py,sha256=SWo1V0K10ZdWE2TQ0KuQUfue04RTyHJe1f6BOj6c96s,1265
 source_google_sheets/manifest.yaml,sha256=HG8vpxv4lqwe5TCneq_zxuERaGNdtno7yAs61Wsk30E,15821
 source_google_sheets/models/__init__.py,sha256=Z-4MTpxG5t2jGhXzs4PPoIOa83zw3jRnUDx0N9Puv3s,61
@@ -8,9 +8,9 @@ source_google_sheets/models/spreadsheet.py,sha256=DEef7bWQEpY1Uqyk7RN6qLF8oxLSNz
 source_google_sheets/models/spreadsheet_values.py,sha256=-XRMuuILn9JN8svHNTj6-oG8mLTZOZ5Hejy2pJ5bILk,440
 source_google_sheets/run.py,sha256=eaPRcarWqkB2b2DokvI83w7rz1blmWPQCFahvCyCdSY,1887
 source_google_sheets/source.py,sha256=qO1KoGdphieu7F5VgDYtrbqs56AUvMWFGNvFHP2b9Z4,778
-source_google_sheets/spec.yaml,sha256=RIUILMhfS0is2r_mCkmIVrQfvND1D3eobDK1YElmzhU,5009
-source_google_sheets/utils.py,sha256=JEQIVLSFEAff-7zF3gPzsvFc9xLfCj9hVuFFYrSWiOo,2290
-airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info/METADATA,sha256=-6yDL8Kuv_mT4ImBf1toakCIbNX3l-2qTd3XfI9WILY,5384
-airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info/entry_points.txt,sha256=Dtsfjohe5IPUFyqojk49SIoP7CifCTlNLG_pgivzppo,69
-airbyte_source_google_sheets-0.9.5.dev202505142036.dist-info/RECORD,,
+source_google_sheets/spec.yaml,sha256=mFOiMN1IsrjyLRcjTRw7xa0b74LwkUXdVvUqau7iTXc,5709
+source_google_sheets/utils.py,sha256=NUw8pJSCORGKx8mrhd11x4xS1dGUraDZdh6WxVn7aQw,5150
+airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info/METADATA,sha256=MW6vZRfbucb1JqCsXRg2O93pL5s1hRalyUfVV-xKR6I,5385
+airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info/entry_points.txt,sha256=Dtsfjohe5IPUFyqojk49SIoP7CifCTlNLG_pgivzppo,69
+airbyte_source_google_sheets-0.10.0.dev202505231635.dist-info/RECORD,,

source_google_sheets/components/extractors.py

@@ -12,7 +12,7 @@ from airbyte_cdk.sources.declarative.decoders.json_decoder import JsonDecoder
 from airbyte_cdk.sources.declarative.extractors.dpath_extractor import DpathExtractor
 from airbyte_cdk.sources.declarative.interpolation.interpolated_string import InterpolatedString
 from airbyte_cdk.sources.types import Config
-from source_google_sheets.utils import name_conversion, safe_name_conversion
+from source_google_sheets.utils import experimental_safe_name_conversion, name_conversion, safe_name_conversion
 
 
 class RawSchemaParser:
@@ -54,9 +54,12 @@ class RawSchemaParser:
         schema_pointer: List[Union[InterpolatedString, str]],
         key_pointer: List[Union[InterpolatedString, str]],
         names_conversion: bool,
+        experimental_names_conversion: bool,
     ):
         """
-        1. Parses sheet headers from the provided raw schema, skipping any headers that are empty or contain only whitespace.
+        1. Parses sheet headers from the provided raw schema. This method assumes that data is contiguous
+            i.e: every cell contains a value and the first cell which does not contain a value denotes the end
+            of the headers.
         2. Makes name conversion if required.
         3. Removes duplicated fields from the schema.
         Return a list of tuples with correct property index (by found in array), value and raw_schema
@@ -68,8 +71,11 @@ class RawSchemaParser:
         for property_index, raw_schema_property in enumerate(raw_schema_properties):
             raw_schema_property_value = self._extract_data(raw_schema_property, key_pointer)
             if not raw_schema_property_value or raw_schema_property_value.isspace():
-                continue
-            if names_conversion:
+                break
+            # Apply experimental conversion if enabled; otherwise, apply standard conversion if enabled
+            if experimental_names_conversion:
+                raw_schema_property_value = experimental_safe_name_conversion(raw_schema_property_value)
+            elif names_conversion:
                 raw_schema_property_value = safe_name_conversion(raw_schema_property_value)
 
             if raw_schema_property_value in seen_values:
@@ -87,12 +93,13 @@ class RawSchemaParser:
     def parse(self, schema_type_identifier, records: Iterable[MutableMapping[Any, Any]]):
         """Removes duplicated fields and makes names conversion"""
         names_conversion = self.config.get("names_conversion", False)
+        experimental_names_conversion = self.config.get("experimental_names_conversion", False)
         schema_pointer = schema_type_identifier.get("schema_pointer")
         key_pointer = schema_type_identifier["key_pointer"]
         parsed_properties = []
         for raw_schema_data in records:
             for _, parsed_value, raw_schema_property in self.parse_raw_schema_values(
-                raw_schema_data, schema_pointer, key_pointer, names_conversion
+                raw_schema_data, schema_pointer, key_pointer, names_conversion, experimental_names_conversion
             ):
                 self._set_data(parsed_value, raw_schema_property, key_pointer)
                 parsed_properties.append(raw_schema_property)
@@ -138,16 +145,20 @@ class DpathSchemaMatchingExtractor(DpathExtractor, RawSchemaParser):
         self._values_to_match_key = parameters["values_to_match_key"]
         schema_type_identifier = parameters["schema_type_identifier"]
         names_conversion = self.config.get("names_conversion", False)
+        experimental_names_conversion = self.config.get("experimental_names_conversion", False)
         self._indexed_properties_to_match = self.extract_properties_to_match(
-            parameters["properties_to_match"], schema_type_identifier, names_conversion=names_conversion
+            parameters["properties_to_match"],
+            schema_type_identifier,
+            names_conversion=names_conversion,
+            experimental_names_conversion=experimental_names_conversion,
         )
 
-    def extract_properties_to_match(self, properties_to_match, schema_type_identifier, names_conversion):
+    def extract_properties_to_match(self, properties_to_match, schema_type_identifier, names_conversion, experimental_names_conversion):
         schema_pointer = schema_type_identifier.get("schema_pointer")
         key_pointer = schema_type_identifier["key_pointer"]
         indexed_properties = {}
         for property_index, property_parsed_value, _ in self.parse_raw_schema_values(
-            properties_to_match, schema_pointer, key_pointer, names_conversion
+            properties_to_match, schema_pointer, key_pointer, names_conversion, experimental_names_conversion
         ):
             indexed_properties[property_index] = property_parsed_value
         return indexed_properties

source_google_sheets/spec.yaml

@@ -33,6 +33,14 @@ connectionSpecification:
       title: Convert Column Names to SQL-Compliant Format
       description: Enables the conversion of column names to a standardized, SQL-compliant format. For example, 'My Name' -> 'my_name'. Enable this option if your destination is SQL-based.
       default: false
+    experimental_names_conversion:
+      type: boolean
+      title: Experimental Convert Column Names to SQL-Compliant Format
+      description: >-
+        Adds additional sanitization to column names before converting to SQL-compliant format, such as removing leading and trailing spaces.
+        This option may change behavior in the future, which may cause column names to update in your destination on future updates.
+        Due to this, it is recommended that you also change the "Detect and propagate schema changes" to "Approve all changes myself" in the connection advances settings.
+        If enabled, this option will supersede the `Convert Column Names to SQL-Compliant Format` option.
     credentials:
       type: object
       title: Authentication

source_google_sheets/utils.py

@@ -37,6 +37,61 @@ def name_conversion(text: str) -> str:
     return text
 
 
+def experimental_name_conversion(text: str) -> str:
+    """
+    Convert name using a set of rules, for example: '1MyName' -> '_1_my_name'
+    Removes leading/trailing spaces, combines number-word pairs (e.g., '50th' -> '50th'),
+    letter-number pairs (e.g., 'Q3' -> 'Q3'), and removes special characters without adding underscores.
+    Spaces are converted to underscores for snake_case.
+    """
+    text = unidecode.unidecode(text.strip())  # Strip leading/trailing spaces
+
+    tokens = []
+    for m in TOKEN_PATTERN.finditer(text):
+        if m.group("NoToken") is None:
+            tokens.append(m.group(0))
+        else:
+            # Only add an empty token for spaces to preserve snake_case; skip other special characters
+            if m.group(0).isspace():
+                tokens.append("")
+            # Otherwise, skip the special character entirely
+
+    # Combine single uppercase letter followed by number (e.g., "Q" and "3" -> "Q3"), then number-word pairs
+    combined_tokens = []
+    i = 0
+    while i < len(tokens):
+        # Check for letter-number pair (e.g., "Q3")
+        if i + 1 < len(tokens) and len(tokens[i]) == 1 and tokens[i].isupper() and tokens[i + 1].isdigit():
+            combined_tokens.append(tokens[i] + tokens[i + 1])
+            i += 2
+        # Check for number-word pair (e.g., "50th")
+        elif i + 1 < len(tokens) and tokens[i].isdigit() and tokens[i + 1].isalpha():
+            combined_tokens.append(tokens[i] + tokens[i + 1])
+            i += 2
+        else:
+            # Only add the token if it's non-empty to avoid underscores from spaces near special characters
+            if tokens[i]:
+                combined_tokens.append(tokens[i])
+            i += 1
+
+    # Remove trailing empty tokens to avoid trailing underscores
+    while combined_tokens and combined_tokens[-1] == "":
+        combined_tokens.pop()
+    # Remove leading empty tokens to avoid leading underscores
+    while combined_tokens and combined_tokens[0] == "":
+        combined_tokens.pop(0)
+
+    if len(combined_tokens) >= 3:
+        combined_tokens = combined_tokens[:1] + [t for t in combined_tokens[1:-1] if t] + combined_tokens[-1:]
+
+    if combined_tokens and combined_tokens[0].isdigit():
+        combined_tokens.insert(0, "")
+
+    text = DEFAULT_SEPARATOR.join(combined_tokens)
+    text = text.lower()
+    return text
+
+
 def safe_name_conversion(text: str) -> str:
     if not text:
         return text
@@ -46,6 +101,24 @@ def safe_name_conversion(text: str) -> str:
     return new
 
 
+import csv
+
+
+def experimental_safe_name_conversion(text: str, output_file: str = "conversion_results2.csv") -> str:
+    if not text:
+        return text
+    new = experimental_name_conversion(text)
+    if not new:
+        raise Exception(f"initial string '{text}' converted to empty")
+
+    # Write to CSV
+    with open(output_file, mode="a", newline="", encoding="utf-8") as file:
+        writer = csv.writer(file)
+        writer.writerow([text, new])  # Write the original and converted text
+
+    return new
+
+
 def exception_description_by_status_code(code: int, spreadsheet_id) -> str:
     if code in [status_codes.INTERNAL_SERVER_ERROR, status_codes.BAD_GATEWAY, status_codes.SERVICE_UNAVAILABLE]:
         return (