wordlift-sdk 2.7.6__tar.gz → 2.9.0__tar.gz

{wordlift_sdk-2.7.6 → wordlift_sdk-2.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wordlift-sdk
-Version: 2.7.6
+Version: 2.9.0
 Summary: 
 Author: David Riccitelli
 Author-email: david@wordlift.io
@@ -123,3 +123,8 @@ poetry install --with dev
 poetry run pytest
 ```
 
+## Documentation
+
+- [Google Sheets Lookup](docs/google_sheets_lookup.md): Utility for O(1) lookups from Google Sheets.
+
+

{wordlift_sdk-2.7.6 → wordlift_sdk-2.9.0}/README.md

@@ -96,3 +96,8 @@ Add `.ttl.liquid` files under `data/templates`. Templates render with `account`
 poetry install --with dev
 poetry run pytest
 ```
+
+## Documentation
+
+- [Google Sheets Lookup](docs/google_sheets_lookup.md): Utility for O(1) lookups from Google Sheets.
+

{wordlift_sdk-2.7.6 → wordlift_sdk-2.9.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "wordlift-sdk"
-version = "2.7.6"
+version = "2.9.0"
 description = ""
 authors = ["David Riccitelli <david@wordlift.io>"]
 readme = "README.md"

wordlift_sdk-2.9.0/wordlift_sdk/google_sheets/__init__.py

@@ -0,0 +1,3 @@
+from .google_sheets_lookup import GoogleSheetsLookup
+
+__all__ = ["GoogleSheetsLookup"]

wordlift_sdk-2.9.0/wordlift_sdk/google_sheets/google_sheets_lookup.py

@@ -0,0 +1,121 @@
+import logging
+import os
+from typing import Optional, Any, Dict
+
+import gspread
+from wordlift_sdk.configuration import ConfigurationProvider
+
+logger = logging.getLogger(__name__)
+
+
+class GoogleSheetsLookup:
+    """
+    A generic class to lookup values from a Google Sheet.
+    Preloads data upon initialization for O(1) lookup performance.
+    """
+
+    def __init__(
+        self,
+        spreadsheet_url: str,
+        sheet_name: str,
+        key_column: str,
+        value_column: str,
+        configuration_provider: ConfigurationProvider,
+        service_account_file: Optional[str] = None,
+    ):
+        """
+        Initialize the GoogleSheetsLookup.
+
+        :param spreadsheet_url: The URL of the Google Sheet.
+        :param sheet_name: The name of the specific worksheet (tab).
+        :param key_column: The header name of the column to use as keys.
+        :param value_column: The header name of the column to use as values.
+        :param configuration_provider: The ConfigurationProvider instance.
+        :param service_account_file: Optional path to the service account JSON file.
+                                     If not provided, it will be looked up in the configuration.
+        """
+        self.spreadsheet_url = spreadsheet_url
+        self.sheet_name = sheet_name
+        self.key_column = key_column
+        self.value_column = value_column
+        self.configuration_provider = configuration_provider
+        self.service_account_file = service_account_file
+
+        self._data: Dict[str, Any] = {}
+        self._load_data()
+
+    def _resolve_service_account_file(self) -> Optional[str]:
+        """
+        Resolves the service account file path.
+        Priority:
+        1. Argument passed to __init__
+        2. 'SERVICE_ACCOUNT_FILE' from ConfigurationProvider
+        """
+        if self.service_account_file:
+            return self.service_account_file
+
+        # Attempt to retrieve from ConfigurationProvider
+        return self.configuration_provider.get_value("SERVICE_ACCOUNT_FILE")
+
+        # Fallback: check environment variable directly if ConfigurationProvider
+        # behavior is different or it didn't return anything.
+        return os.getenv("SERVICE_ACCOUNT_FILE")
+
+    def _load_data(self):
+        """
+        Connects to Google Sheets and preloads the data into a dictionary.
+        """
+        credentials_file = self._resolve_service_account_file()
+
+        try:
+            if credentials_file:
+                logger.info(
+                    f"Connecting to Google Sheets using credentials: {credentials_file}"
+                )
+                gc = gspread.service_account(filename=credentials_file)
+            else:
+                logger.info(
+                    "Connecting to Google Sheets using default environment credentials"
+                )
+                gc = gspread.service_account()
+
+            # Open spreadsheet by URL
+            sh = gc.open_by_url(self.spreadsheet_url)
+
+            # Select worksheet
+            worksheet = sh.worksheet(self.sheet_name)
+
+            # Get all records
+            records = worksheet.get_all_records()
+            logger.info(
+                f"Fetched {len(records)} records from sheet '{self.sheet_name}'"
+            )
+
+            # Build lookup dictionary
+            for i, record in enumerate(records):
+                key = record.get(self.key_column)
+                value = record.get(self.value_column)
+
+                if key is None:
+                    logger.warning(
+                        f"Row {i + 2}: Key column '{self.key_column}' is missing or empty. Skipping."
+                    )
+                    continue
+
+                # We stringify the key to ensure consistent lookup, optional but recommended for mixed types
+                self._data[str(key)] = value
+
+            logger.info(f"Successfully loaded {len(self._data)} items into cache.")
+
+        except Exception as e:
+            logger.error(f"Failed to load data from Google Sheets: {e}")
+            raise
+
+    def get_value(self, key: Any) -> Optional[Any]:
+        """
+        Look up a value by its key.
+
+        :param key: The key to look up.
+        :return: The corresponding value, or None if not found.
+        """
+        return self._data.get(str(key))

{wordlift_sdk-2.7.6 → wordlift_sdk-2.9.0}/wordlift_sdk/utils/__init__.py

@@ -1,10 +1,13 @@
 from .create_dataframe_from_google_sheets import create_dataframe_from_google_sheets
 from .create_dataframe_of_entities_by_types import create_dataframe_of_entities_by_types
-from .create_dataframe_of_entities_with_embedding_vectors import create_dataframe_of_entities_with_embedding_vectors
+from .create_dataframe_of_entities_with_embedding_vectors import (
+    create_dataframe_of_entities_with_embedding_vectors,
+)
 from .create_dataframe_of_url_iri import create_dataframe_of_url_iri
 from .create_entity_patch_request import create_entity_patch_request
 from .delayed import create_delayed
 from .get_me import get_me
+from .html_converter import HtmlConverter
 
 __all__ = [
     "create_dataframe_from_google_sheets",
@@ -13,5 +16,6 @@ __all__ = [
     "create_dataframe_of_url_iri",
     "create_entity_patch_request",
     "create_delayed",
-    "get_me"
+    "get_me",
+    "HtmlConverter",
 ]

wordlift_sdk-2.9.0/wordlift_sdk/utils/html_converter.py

@@ -0,0 +1,56 @@
+"""HTML to XHTML conversion utility."""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+_INVALID_XML_CHARS_RE = re.compile(r"[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]")
+_XML_NAME_RE = re.compile(r"^[A-Za-z_][A-Za-z0-9_.:-]*$")
+
+
+class HtmlConverter:
+    """Converts HTML to XHTML."""
+
+    def convert(self, html: str) -> str:
+        """
+        Convert an HTML string to a valid XHTML string.
+
+        Args:
+            html: The raw HTML string.
+
+        Returns:
+            A sanitized XHTML string.
+        """
+        html = re.sub(r"<!DOCTYPE[^>]*>", "", html, flags=re.IGNORECASE)
+        html = self._strip_invalid_xml_chars(html)
+        try:
+            from lxml import html as lxml_html
+        except ImportError as exc:
+            raise ImportError(
+                "lxml is required for XHTML output. Install with: pip install lxml"
+            ) from exc
+
+        try:
+            parser = lxml_html.HTMLParser(encoding="utf-8", recover=True)
+            doc = lxml_html.document_fromstring(html, parser=parser)
+            self._sanitize_xhtml_tree(doc)
+            xhtml = lxml_html.tostring(doc, encoding="unicode", method="xml")
+            return self._strip_invalid_xml_chars(xhtml)
+        except Exception as exc:
+            raise RuntimeError("Failed to convert HTML to XHTML.") from exc
+
+    def _strip_invalid_xml_chars(self, value: str) -> str:
+        return _INVALID_XML_CHARS_RE.sub("", value)
+
+    def _sanitize_xhtml_tree(self, doc: Any) -> None:
+        for element in doc.iter():
+            if not hasattr(element, "attrib"):
+                continue
+            for attr in list(element.attrib):
+                if not _XML_NAME_RE.match(attr):
+                    del element.attrib[attr]
+                    continue
+                value = element.attrib.get(attr)
+                if isinstance(value, str):
+                    element.attrib[attr] = self._strip_invalid_xml_chars(value)