Revert "Fixed issues related formatting and linting"

This reverts commit 2e42182.

Author: bharath03-a

.github/workflows/ci.yaml

@@ -308,22 +308,22 @@ jobs:
           set -euo pipefail
           EXPECTED_SHA="${{ steps.sha-pin.outputs.SHA_TO_TEST }}"
           echo "Fetching exact commit: $EXPECTED_SHA"
-
+          
           # Fetch the specific commit SHA
           git fetch --no-tags --prune --no-recurse-submodules origin "$EXPECTED_SHA" || {
             echo "::error::Failed to fetch PR commit $EXPECTED_SHA. The commit may have been deleted."
             exit 1
           }
-
+          
           git checkout -b pr-to-test "$EXPECTED_SHA"
-
+          
           # Verify checkout
           ACTUAL_SHA="$(git rev-parse HEAD)"
           if [ "$ACTUAL_SHA" != "$EXPECTED_SHA" ]; then
             echo "::error::SHA verification failed! Expected $EXPECTED_SHA but got $ACTUAL_SHA"
             exit 1
           fi
-
+          
           echo "::notice title=Security::Successfully verified commit SHA: $ACTUAL_SHA"
 
       - name: Set up Python 3.11
@@ -365,13 +365,13 @@ jobs:
 
           SHA_TO_MERGE="${{ steps.sha-pin.outputs.SHA_TO_TEST }}"
           echo "Merging verified commit: $SHA_TO_MERGE"
-
+          
           git fetch --no-tags --prune --no-recurse-submodules origin "$SHA_TO_MERGE"
           git merge --no-ff --no-edit "$SHA_TO_MERGE" || {
             echo "::error::Failed to merge commit $SHA_TO_MERGE"
             exit 1
           }
-
+          
           echo "::notice title=Security::Successfully merged verified commit"
 
       - name: Add status comment

.github/workflows/validate_pr_template.yaml

@@ -38,10 +38,10 @@ jobs:
               owner, repo,
               pull_number: pr.number
             });
-
-            const isCommunityProviderPR = files.length === 1 &&
+            
+            const isCommunityProviderPR = files.length === 1 && 
                                           files[0].filename === 'COMMUNITY_PROVIDERS.md';
-
+            
             if (isCommunityProviderPR) {
               core.setOutput('is_community_provider', 'true');
               console.log('Community provider PR detected - relaxed validation will apply');

langextract/annotation.py

@@ -216,43 +216,41 @@ def _process_batch_with_retry(
       **kwargs,
   ) -> Iterator[list[core_types.ScoredOutput]]:
     """Process a batch of prompts with individual chunk retry capability.
-
+    
     This method processes each chunk individually and retries failed chunks
     due to transient errors (like 503 "model overloaded") while preserving
     successful chunks from the same batch.
-
+    
     Args:
       batch_prompts: List of prompts for the batch
       batch: List of TextChunk objects corresponding to the prompts
       **kwargs: Additional arguments passed to the language model
-
+      
     Yields:
       Lists of ScoredOutputs, with retries for failed chunks
     """
     try:
-      batch_results = list(
-          self._language_model.infer(
-              batch_prompts=batch_prompts,
-              **kwargs,
-          )
-      )
-
+      batch_results = list(self._language_model.infer(
+          batch_prompts=batch_prompts,
+          **kwargs,
+      ))
+      
       for result in batch_results:
         yield result
       return
-
+      
     except Exception as e:
       if not retry_utils.is_transient_error(e):
         raise
-
+      
       logging.warning(
           "Batch processing failed with transient error: %s. "
           "Falling back to individual chunk processing with retry.",
-          str(e),
+          str(e)
       )
-
+    
     individual_results = []
-
+    
     for i, (prompt, chunk) in enumerate(zip(batch_prompts, batch)):
       try:
         chunk_result = self._process_single_chunk_with_retry(
@@ -266,19 +264,16 @@ def _process_batch_with_retry(
             **kwargs,
         )
         individual_results.append(chunk_result)
-
+        
       except Exception as e:
         logging.error(
             "Failed to process chunk %d after retries: %s. "
             "Chunk info: document_id=%s, text_length=%d. "
             "Stopping document processing.",
-            i,
-            str(e),
-            chunk.document_id,
-            len(chunk.chunk_text),
+            i, str(e), chunk.document_id, len(chunk.chunk_text)
         )
         raise
-
+    
     for result in individual_results:
       yield result
 
@@ -294,7 +289,7 @@ def _process_single_chunk_with_retry(
       **kwargs,
   ) -> list[core_types.ScoredOutput]:
     """Process a single chunk with retry logic.
-
+    
     Args:
       prompt: The prompt for this chunk
       chunk: The TextChunk object
@@ -304,69 +299,59 @@ def _process_single_chunk_with_retry(
       retry_backoff_factor: Backoff multiplier for retries
       retry_max_delay: Maximum delay between retries
       **kwargs: Additional arguments for the language model
-
+      
     Returns:
       List containing a single ScoredOutput for this chunk
     """
     last_exception = None
     delay = retry_initial_delay
-
+    
     for attempt in range(max_retries + 1):
       try:
-        batch_results = list(
-            self._language_model.infer(
-                batch_prompts=[prompt],
-                **kwargs,
-            )
-        )
-
+        batch_results = list(self._language_model.infer(
+            batch_prompts=[prompt],
+            **kwargs,
+        ))
+        
         if not batch_results:
           raise exceptions.InferenceOutputError(
               f"No results returned for chunk in document {chunk.document_id}"
           )
-
+        
         return batch_results[0]
-
+        
       except Exception as e:
         last_exception = e
-
+        
         if not retry_transient_errors or not retry_utils.is_transient_error(e):
           logging.debug(
-              "Not retrying chunk processing: retry_disabled=%s,"
-              " is_transient=%s, error=%s",
-              not retry_transient_errors,
-              retry_utils.is_transient_error(e),
-              str(e),
+              "Not retrying chunk processing: retry_disabled=%s, is_transient=%s, error=%s",
+              not retry_transient_errors, retry_utils.is_transient_error(e), str(e)
           )
           raise
-
+        
         if attempt >= max_retries:
           logging.error(
               "Chunk processing failed after %d retries: %s",
-              max_retries,
-              str(e),
+              max_retries, str(e)
           )
           raise
-
+        
         current_delay = min(delay, retry_max_delay)
-
+        
         import random
-
         jitter_amount = current_delay * 0.1 * random.random()
         current_delay += jitter_amount
-
+        
         logging.warning(
-            "Chunk processing failed on attempt %d/%d due to transient error:"
-            " %s. Retrying in %.2f seconds...",
-            attempt + 1,
-            max_retries + 1,
-            str(e),
-            current_delay,
+            "Chunk processing failed on attempt %d/%d due to transient error: %s. "
+            "Retrying in %.2f seconds...",
+            attempt + 1, max_retries + 1, str(e), current_delay
         )
-
+        
         time.sleep(current_delay)
         delay = min(delay * retry_backoff_factor, retry_max_delay)
-
+    
     if last_exception:
       raise last_exception
     raise RuntimeError("Chunk retry logic failed unexpectedly")

langextract/extraction.py

@@ -333,11 +333,11 @@ def extract(
 
   # Add retry parameters to alignment kwargs
   retry_kwargs = {
-      "retry_transient_errors": retry_transient_errors,
-      "max_retries": max_retries,
-      "retry_initial_delay": retry_initial_delay,
-      "retry_backoff_factor": retry_backoff_factor,
-      "retry_max_delay": retry_max_delay,
+      'retry_transient_errors': retry_transient_errors,
+      'max_retries': max_retries,
+      'retry_initial_delay': retry_initial_delay,
+      'retry_backoff_factor': retry_backoff_factor,
+      'retry_max_delay': retry_max_delay,
   }
   alignment_kwargs.update(retry_kwargs)
 

langextract/plugins.py

@@ -19,9 +19,9 @@
 """
 from __future__ import annotations
 
+from functools import lru_cache
 from importlib import import_module
 from importlib.metadata import entry_points
-from functools import lru_cache
 from typing import Dict, List, Type
 
 from absl import logging

langextract/providers/__init__.py

@@ -19,8 +19,8 @@
 management in build systems.
 """
 
-from importlib import metadata
 import importlib
+from importlib import metadata
 import os
 
 from absl import logging

langextract/providers/gemini.py

@@ -23,7 +23,6 @@
 
 from absl import logging
 
-from langextract import retry_utils
 from langextract.core import base_model
 from langextract.core import data
 from langextract.core import exceptions
@@ -32,6 +31,7 @@
 from langextract.providers import patterns
 from langextract.providers import router
 from langextract.providers import schemas
+from langextract import retry_utils
 
 _API_CONFIG_KEYS: Final[set[str]] = {
     'response_mime_type',