airbyte-source-google-sheets 0.9.5.dev202505142036__tar.gz → 0.10.0.dev202505211933__tar.gz

{airbyte_source_google_sheets-0.9.5.dev202505142036 → airbyte_source_google_sheets-0.10.0.dev202505211933}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: airbyte-source-google-sheets
-Version: 0.9.5.dev202505142036
+Version: 0.10.0.dev202505211933
 Summary: Source implementation for Google Sheets.
 License: Elv2
 Author: Airbyte

{airbyte_source_google_sheets-0.9.5.dev202505142036 → airbyte_source_google_sheets-0.10.0.dev202505211933}/pyproject.toml

@@ -5,7 +5,7 @@ requires = [
 build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
-version = "0.9.5.dev202505142036"
+version = "0.10.0.dev202505211933"
 name = "airbyte-source-google-sheets"
 description = "Source implementation for Google Sheets."
 authors = [

{airbyte_source_google_sheets-0.9.5.dev202505142036 → airbyte_source_google_sheets-0.10.0.dev202505211933}/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

{airbyte_source_google_sheets-0.9.5.dev202505142036 → airbyte_source_google_sheets-0.10.0.dev202505211933}/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

{airbyte_source_google_sheets-0.9.5.dev202505142036 → airbyte_source_google_sheets-0.10.0.dev202505211933}/source_google_sheets/utils.py

@@ -37,6 +37,42 @@ 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 and combines number-word pairs (e.g., '50th' -> '50th').
+    """
+    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:
+            tokens.append("")
+
+    # Combine number followed by word (e.g., "50" and "th" -> "50th")
+    combined_tokens = []
+    i = 0
+    while i < len(tokens):
+        if i + 1 < len(tokens) and tokens[i].isdigit() and tokens[i + 1].isalpha():
+            combined_tokens.append(tokens[i] + tokens[i + 1])
+            i += 2
+        else:
+            combined_tokens.append(tokens[i])
+            i += 1
+
+    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 +82,15 @@ def safe_name_conversion(text: str) -> str:
     return new
 
 
+def experimental_safe_name_conversion(text: str) -> str:
+    if not text:
+        return text
+    new = experimental_name_conversion(text)
+    if not new:
+        raise Exception(f"initial string '{text}' converted to empty")
+    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 (