satif-ai 0.2.8__py3-none-any.whl → 0.2.10__py3-none-any.whl

satif_ai/utils/zip.py

@@ -0,0 +1,120 @@
+import asyncio
+import logging
+import zipfile
+from pathlib import Path
+from typing import List, Tuple
+
+logger = logging.getLogger(__name__)
+
+# Constants for ZIP file processing, kept local to this utility or passed as args if needed
+_IGNORED_ZIP_MEMBER_PREFIXES = ("__MACOSX/",)
+_IGNORED_ZIP_FILENAME_PREFIXES = ("._",)
+_IGNORED_ZIP_FILENAMES = (".DS_Store",)
+
+
+async def extract_zip_archive_async(
+    zip_path: Path,
+    extract_to: Path,
+    ignored_member_prefixes: Tuple[str, ...] = _IGNORED_ZIP_MEMBER_PREFIXES,
+    ignored_filename_prefixes: Tuple[str, ...] = _IGNORED_ZIP_FILENAME_PREFIXES,
+    ignored_filenames: Tuple[str, ...] = _IGNORED_ZIP_FILENAMES,
+) -> List[Path]:
+    """
+    Asynchronously extracts a ZIP archive to a specified directory, filtering out ignored files.
+
+    Args:
+        zip_path: Path to the ZIP archive.
+        extract_to: Directory where the contents will be extracted.
+        ignored_member_prefixes: Tuple of member path prefixes to ignore.
+        ignored_filename_prefixes: Tuple of filename prefixes to ignore.
+        ignored_filenames: Tuple of exact filenames to ignore.
+
+    Returns:
+        A list of paths to the successfully extracted files.
+
+    Raises:
+        ValueError: If the zip_path is invalid or corrupted.
+        RuntimeError: If any other error occurs during extraction.
+    """
+
+    def blocking_extract() -> List[Path]:
+        extracted_file_paths = []
+        logger.info(f"Extracting ZIP archive '{zip_path.name}' to '{extract_to}'...")
+        try:
+            extract_to.mkdir(
+                parents=True, exist_ok=True
+            )  # Ensure extract_to directory exists
+
+            with zipfile.ZipFile(zip_path, "r") as zip_ref:
+                # Security: Preliminary check for unsafe paths before extraction
+                for member_name in zip_ref.namelist():
+                    if member_name.startswith(("/", "..")):
+                        logger.error(
+                            f"Skipping potentially unsafe path in ZIP: {member_name}"
+                        )
+                        # Depending on security policy, might raise an error here
+                        continue
+
+                # Extract all members
+                zip_ref.extractall(extract_to)
+
+            # After extractall, collect all *file* paths, applying filters
+            # This second pass of filtering ensures that even if extractall creates them,
+            # we don't return paths to ignored files.
+            for root, _, files in extract_to.walk():
+                for filename in files:
+                    full_path = root / filename
+                    # Create a path relative to 'extract_to' to check against member prefixes
+                    # This ensures that '__MACOSX/file.txt' is correctly ignored,
+                    # not just a top-level '__MACOSX' directory.
+                    try:
+                        relative_path_to_check = full_path.relative_to(extract_to)
+                    except ValueError:
+                        # This can happen if full_path is not under extract_to,
+                        # which ideally shouldn't occur if zip_ref.extractall worked as expected
+                        # and target_path checks were effective.
+                        logger.warning(
+                            f"File {full_path} seems to be outside extraction root {extract_to}. Skipping."
+                        )
+                        continue
+
+                    path_str_to_check_prefixes = str(relative_path_to_check)
+
+                    if not (
+                        any(
+                            path_str_to_check_prefixes.startswith(p)
+                            for p in ignored_member_prefixes
+                        )
+                        or any(
+                            full_path.name.startswith(p)
+                            for p in ignored_filename_prefixes
+                        )
+                        or full_path.name in ignored_filenames
+                    ):
+                        extracted_file_paths.append(full_path)
+                    else:
+                        logger.debug(f"Ignoring file post-extraction: {full_path}")
+
+            if not extracted_file_paths:
+                logger.warning(
+                    f"ZIP archive '{zip_path.name}' is empty or contains no processable files after filtering."
+                )
+            else:
+                logger.info(
+                    f"Successfully extracted {len(extracted_file_paths)} file(s) from '{zip_path.name}'."
+                )
+            return extracted_file_paths
+        except zipfile.BadZipFile as e:
+            logger.error(
+                f"Invalid or corrupted ZIP file: {zip_path.name}", exc_info=True
+            )
+            raise ValueError(f"Invalid or corrupted ZIP file: {zip_path.name}") from e
+        except Exception as e:
+            logger.error(
+                f"Failed to extract ZIP archive '{zip_path.name}': {e}", exc_info=True
+            )
+            raise RuntimeError(
+                f"Unexpected error during ZIP extraction for '{zip_path.name}'"
+            ) from e
+
+    return await asyncio.to_thread(blocking_extract)

{satif_ai-0.2.8.dist-info → satif_ai-0.2.10.dist-info}/METADATA

@@ -1,17 +1,19 @@
 Metadata-Version: 2.3
 Name: satif-ai
-Version: 0.2.8
+Version: 0.2.10
 Summary: AI Agents for Satif
 License: MIT
-Author: Bryan Djafer
-Author-email: bryan.djafer@syncpulse.fr
-Requires-Python: >=3.10,<4.0
+Author: Syncpulse
+Maintainer: Bryan Djafer
+Maintainer-email: bryan.djafer@syncpulse.fr
+Requires-Python: >=3.10,<3.14
 Classifier: License :: OSI Approved :: MIT License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Provides-Extra: xlsx
 Requires-Dist: openai-agents (>=0.0.9,<0.0.10)
 Requires-Dist: satif-sdk (>=0.1.0,<1.0.0)
 Requires-Dist: sdif-mcp (>=0.1.0,<1.0.0)

satif_ai-0.2.10.dist-info/RECORD

@@ -0,0 +1,20 @@
+satif_ai/__init__.py,sha256=cqJ6Kd9IolVodPi9yOBPnfhYQXH5a1JgRB3HfLOtP_4,611
+satif_ai/adapters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+satif_ai/adapters/tidy.py,sha256=1g7Wcq8agAZhaAqQDhhD8yh3iO5gZ4mwdKHsiNN3hHY,18540
+satif_ai/standardize.py,sha256=TgAB_nhcHY8zqlfT1PpgfgSswqdE-ly-dheQz-7NC7Q,5674
+satif_ai/standardizers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+satif_ai/standardizers/ai.py,sha256=jtYM-ChjLtkpFaubz980CTCNAoC33iYxB3pq0_hn2lU,21045
+satif_ai/standardizers/ai_csv.py,sha256=LbCRaLleujQRgSRRyt9ujbED-PIGRq1J8zRnejGM5nc,25437
+satif_ai/standardizers/ai_xlsx.py,sha256=558Bzfy8WGuk5mdnjMvvtakQXcU3rmwK3ykPjpXKwmQ,15863
+satif_ai/transform.py,sha256=g5XNeVCIKUgDW3UIhf02MN9xkXnWF3EJXS0Eig_hfD8,7677
+satif_ai/transformation_builders/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+satif_ai/transformation_builders/syncpulse.py,sha256=c59BZicNnqs3NDKpflBAPqw42pGb6nYB2Zps0ChGyaM,11368
+satif_ai/utils/__init__.py,sha256=F-usaCt_vX872mXvtukuZdNMPnkVqDb8RaDgox2uow4,212
+satif_ai/utils/merge_sdif.py,sha256=y4C6pgkdyer0QugroFKUck4Eud4Ap-tJzM-eclMo3Rw,25629
+satif_ai/utils/openai_mcp.py,sha256=duCQZXG0mBs9DOOFIUvzraJhxD2IDzegWO9iOiLfFwY,3938
+satif_ai/utils/zip.py,sha256=G_GK8629Iw0TLFCQJfnqOscv7MoKF5zdzxvEAbL7Gss,5186
+satif_ai-0.2.10.dist-info/LICENSE,sha256=kS8EN6yAaGZd7V5z6GKSn_x3ozcZltrfRky4vMPRCw8,1072
+satif_ai-0.2.10.dist-info/METADATA,sha256=O5QWv8YJFtB5AIniv0LRgmSgpEaRLVdlz8WHZAru1X8,719
+satif_ai-0.2.10.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
+satif_ai-0.2.10.dist-info/entry_points.txt,sha256=Mz2SwYALjktap1bF-Q3EWBgiZVNT6QJCVsCs_fCV33Y,43
+satif_ai-0.2.10.dist-info/RECORD,,

satif_ai/code_builders/adaptation.py

@@ -1,9 +0,0 @@
-from satif_core.code_builders.base import AsyncCodeBuilder, CodeBuilder
-
-
-class AdaptationCodeBuilder(CodeBuilder):
-    pass
-
-
-class AdaptationAsyncCodeBuilder(AsyncCodeBuilder):
-    pass

satif_ai-0.2.8.dist-info/RECORD

@@ -1,13 +0,0 @@
-satif_ai/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-satif_ai/adapters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-satif_ai/adapters/tidy.py,sha256=2oYj7Gz3vOQtzcpoJI4JbftWlMKvOWL8rdwthjg-zUE,19884
-satif_ai/code_builders/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-satif_ai/code_builders/adaptation.py,sha256=E29YM0S6pMtAfB0uzSUexoeWKwXfF8iJVyYUCKWQz5k,188
-satif_ai/code_builders/transformation.py,sha256=5B7a6lDv-gqQo83F8fQeSw2gHpDgznoDfjXsASkLc60,11870
-satif_ai/standardizers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-satif_ai/standardizers/ai_csv.py,sha256=c0CKnIib610GgwGqaF8NaqT_P4pZ2BupO-BTSNuIhoc,25385
-satif_ai-0.2.8.dist-info/LICENSE,sha256=kS8EN6yAaGZd7V5z6GKSn_x3ozcZltrfRky4vMPRCw8,1072
-satif_ai-0.2.8.dist-info/METADATA,sha256=CZBbNd1A-KL8eoOzmmz7bW3ue4HGOC2Qic60wQ-v6z8,670
-satif_ai-0.2.8.dist-info/WHEEL,sha256=b4K_helf-jlQoXBBETfwnf4B04YC67LOev0jo4fX5m8,88
-satif_ai-0.2.8.dist-info/entry_points.txt,sha256=Mz2SwYALjktap1bF-Q3EWBgiZVNT6QJCVsCs_fCV33Y,43
-satif_ai-0.2.8.dist-info/RECORD,,