Merge pull request #742 from ftnext/action-flake8

.github/workflows/lint.yml

@@ -0,0 +1,26 @@
+name: Static analysis
+
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    branches:
+      - master
+
+jobs:
+  flake8:
+    runs-on: ubuntu-latest
+
+    steps:
+        - uses: actions/checkout@v4
+        - uses: actions/setup-python@v5
+          with:
+            python-version: '3.11'
+        - name: Install dependencies
+          run: |
+            python -m pip install flake8
+        - name: Run flake8
+          run: |
+            # ignore errors for long lines and multi-statement lines
+            python -m flake8 --ignore=E501,E701,W503 speech_recognition tests examples setup.py

README.rst

@@ -276,7 +276,7 @@ To hack on this library, first make sure you have all the requirements listed in
 -  Documentation can be found in the ``reference/`` `directory <https://github.com/Uberi/speech_recognition/tree/master/reference>`__.
 -  Third-party libraries, utilities, and reference material are in the ``third-party/`` `directory <https://github.com/Uberi/speech_recognition/tree/master/third-party>`__.
 
-To install/reinstall the library locally, run ``python setup.py install`` in the project `root directory <https://github.com/Uberi/speech_recognition>`__.
+To install/reinstall the library locally, run ``python -m pip install -e .[dev]`` in the project `root directory <https://github.com/Uberi/speech_recognition>`__.
 
 Before a release, the version number is bumped in ``README.rst`` and ``speech_recognition/__init__.py``. Version tags are then created using ``git config gpg.program gpg2 && git config user.signingkey DB45F6C431DE7C2DCD99FF7904882258A4063489 && git tag -s VERSION_GOES_HERE -m "Version VERSION_GOES_HERE"``.
 
@@ -291,17 +291,19 @@ To run all the tests:
 
     python -m unittest discover --verbose
 
-Testing is also done automatically by TravisCI, upon every push. To set up the environment for offline/local Travis-like testing on a Debian-like system:
+To run static analysis:
 
 .. code:: bash
 
-    sudo docker run --volume "$(pwd):/speech_recognition" --interactive --tty quay.io/travisci/travis-python:latest /bin/bash
-    su - travis && cd /speech_recognition
-    sudo apt-get update && sudo apt-get install swig libpulse-dev
-    pip install --user pocketsphinx && pip install --user flake8 rstcheck && pip install --user -e .
-    python -m unittest discover --verbose # run unit tests
-    python -m flake8 --ignore=E501,E701 speech_recognition tests examples setup.py # ignore errors for long lines and multi-statement lines
-    python -m rstcheck README.rst reference/*.rst # ensure RST is well-formed
+    python -m flake8 --ignore=E501,E701,W503 speech_recognition tests examples setup.py
+
+To ensure RST is well-formed:
+
+.. code:: bash
+
+    python -m rstcheck README.rst reference/*.rst
+
+Testing is also done automatically by GitHub Actions, upon every push.
 
 FLAC Executables
 ~~~~~~~~~~~~~~~~

examples/microphone_recognition.py

@@ -91,11 +91,11 @@
 except sr.UnknownValueError:
     print("Whisper could not understand audio")
 except sr.RequestError as e:
-    print("Could not request results from Whisper")
+    print(f"Could not request results from Whisper; {e}")
 
 # recognize speech using Whisper API
 OPENAI_API_KEY = "INSERT OPENAI API KEY HERE"
 try:
     print(f"Whisper API thinks you said {r.recognize_whisper_api(audio, api_key=OPENAI_API_KEY)}")
 except sr.RequestError as e:
-    print("Could not request results from Whisper API")
+    print(f"Could not request results from Whisper API; {e}")

setup.cfg

@@ -4,6 +4,9 @@
 universal=1
 
 [options.extras_require]
+dev =
+    flake8
+    rstcheck
 whisper-local =
     openai-whisper
     soundfile

setup.py

@@ -21,12 +21,15 @@ def run(self):
             if os.path.basename(output_path) in FILES_TO_MARK_EXECUTABLE:
                 log.info("setting executable permissions on {}".format(output_path))
                 stat_info = os.stat(output_path)
+                OWNER_CAN_READ_EXECUTE = stat.S_IRUSR | stat.S_IXUSR
+                GROUP_CAN_READ_EXECUTE = stat.S_IRGRP | stat.S_IXGRP
+                OTHERS_CAN_READ_EXECUTE = stat.S_IROTH | stat.S_IXOTH
                 os.chmod(
                     output_path,
-                    stat_info.st_mode |
-                    stat.S_IRUSR | stat.S_IXUSR |  # owner can read/execute
-                    stat.S_IRGRP | stat.S_IXGRP |  # group can read/execute
-                    stat.S_IROTH | stat.S_IXOTH  # everyone else can read/execute
+                    stat_info.st_mode
+                    | OWNER_CAN_READ_EXECUTE
+                    | GROUP_CAN_READ_EXECUTE
+                    | OTHERS_CAN_READ_EXECUTE,
                 )
 
 

speech_recognition/__init__.py

@@ -1087,7 +1087,8 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
             aws_secret_access_key=secret_access_key,
             region_name=region)
 
-        s3 = boto3.client('s3', 
+        s3 = boto3.client(
+            's3',
             aws_access_key_id=access_key_id,
             aws_secret_access_key=secret_access_key,
             region_name=region)
@@ -1107,7 +1108,6 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
         except ClientError as exc:
             print('Error creating bucket %s: %s' % (bucket_name, exc))
         s3res = session.resource('s3')
-        bucket = s3res.Bucket(bucket_name)
         if audio_data is not None:
             print('Uploading audio data...')
             wav_data = audio_data.get_wav_data()
@@ -1124,7 +1124,7 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
             try:
                 status = transcribe.get_transcription_job(TranscriptionJobName=job_name)
             except ClientError as exc:
-                print('!'*80)
+                print('!' * 80)
                 print('Error getting job:', exc.response)
                 if exc.response['Error']['Code'] == 'BadRequestException' and "The requested job couldn't be found" in str(exc):
                     # Some error caused the job we recorded to not exist on AWS.
@@ -1137,7 +1137,7 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
                 else:
                     # Some other error happened, so re-raise.
                     raise
-            
+
             job = status['TranscriptionJob']
             if job['TranscriptionJobStatus'] in ['COMPLETED'] and 'TranscriptFileUri' in job['Transcript']:
 
@@ -1152,12 +1152,12 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
                         confidences.append(float(item['alternatives'][0]['confidence']))
                     confidence = 0.5
                     if confidences:
-                        confidence = sum(confidences)/float(len(confidences))
+                        confidence = sum(confidences) / float(len(confidences))
                     transcript = d['results']['transcripts'][0]['transcript']
 
                     # Delete job.
                     try:
-                        transcribe.delete_transcription_job(TranscriptionJobName=job_name) # cleanup
+                        transcribe.delete_transcription_job(TranscriptionJobName=job_name)  # cleanup
                     except Exception as exc:
                         print('Warning, could not clean up transcription: %s' % exc)
                         traceback.print_exc()
@@ -1167,17 +1167,17 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
 
                     return transcript, confidence
             elif job['TranscriptionJobStatus'] in ['FAILED']:
-            
+
                 # Delete job.
                 try:
-                    transcribe.delete_transcription_job(TranscriptionJobName=job_name) # cleanup
+                    transcribe.delete_transcription_job(TranscriptionJobName=job_name)  # cleanup
                 except Exception as exc:
                     print('Warning, could not clean up transcription: %s' % exc)
                     traceback.print_exc()
 
                 # Delete S3 file.
                 s3.delete_object(Bucket=bucket_name, Key=filename)
-                
+
                 exc = TranscriptionFailed()
                 exc.job_name = None
                 exc.file_key = None
@@ -1193,11 +1193,6 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
         else:
 
             # Launch the transcription job.
-            # try:
-                # transcribe.delete_transcription_job(TranscriptionJobName=job_name) # pre-cleanup
-            # except:
-                # # It's ok if this fails because the job hopefully doesn't exist yet.
-                # pass
             try:
                 transcribe.start_transcription_job(
                     TranscriptionJobName=job_name,
@@ -1210,7 +1205,7 @@ def recognize_amazon(self, audio_data, bucket_name=None, access_key_id=None, sec
                 exc.file_key = None
                 raise exc
             except ClientError as exc:
-                print('!'*80)
+                print('!' * 80)
                 print('Error starting job:', exc.response)
                 if exc.response['Error']['Code'] == 'LimitExceededException':
                     # Could not start job. Cancel everything.
@@ -1277,9 +1272,7 @@ def read_file(filename, chunk_size=5242880):
 
             # Queue file for transcription.
             endpoint = "https://api.assemblyai.com/v2/transcript"
-            json = {
-              "audio_url": upload_url
-            }
+            json = {"audio_url": upload_url}
             headers = {
                 "authorization": api_token,
                 "content-type": "application/json"
@@ -1436,23 +1429,23 @@ def recognize_whisper(self, audio_data, model="base", show_dict=False, load_opti
             return result
         else:
             return result["text"]
-            
+
     def recognize_vosk(self, audio_data, language='en'):
         from vosk import KaldiRecognizer, Model
-        
+
         assert isinstance(audio_data, AudioData), "Data must be audio data"
-        
+
         if not hasattr(self, 'vosk_model'):
             if not os.path.exists("model"):
                 return "Please download the model from https://github.com/alphacep/vosk-api/blob/master/doc/models.md and unpack as 'model' in the current folder."
-                exit (1)
+                exit(1)
             self.vosk_model = Model("model")
 
-        rec = KaldiRecognizer(self.vosk_model, 16000);
-        
-        rec.AcceptWaveform(audio_data.get_raw_data(convert_rate=16000, convert_width=2));
+        rec = KaldiRecognizer(self.vosk_model, 16000)
+
+        rec.AcceptWaveform(audio_data.get_raw_data(convert_rate=16000, convert_width=2))
         finalRecognition = rec.FinalResult()
-        
+
         return finalRecognition
 
 

speech_recognition/audio.py

@@ -113,7 +113,7 @@ def get_raw_data(self, convert_rate=None, convert_width=None):
                     audioop.error
                 ):  # this version of audioop doesn't support 24-bit audio (probably Python 3.3 or less)
                     raw_data = b"".join(
-                        raw_data[i + 1 : i + 4]
+                        raw_data[i + 1: i + 4]
                         for i in range(0, len(raw_data), 4)
                     )  # since we're in little endian, we discard the first byte from each 32-bit sample to get a 24-bit sample
                 else:  # 24-bit audio fully supported, we don't need to shim anything
@@ -188,8 +188,8 @@ def get_aiff_data(self, convert_rate=None, convert_width=None):
         ):  # ``audioop.byteswap`` was only added in Python 3.4
             raw_data = audioop.byteswap(raw_data, sample_width)
         else:  # manually reverse the bytes of each sample, which is slower but works well enough as a fallback
-            raw_data = raw_data[sample_width - 1 :: -1] + b"".join(
-                raw_data[i + sample_width : i : -1]
+            raw_data = raw_data[sample_width - 1:: -1] + b"".join(
+                raw_data[i + sample_width: i: -1]
                 for i in range(sample_width - 1, len(raw_data), sample_width)
             )
 

tests/recognizers/test_google.py

@@ -79,7 +79,7 @@ def test_parse_without_confidence(
 
     @patch(f"{CLASS_UNDER_TEST}.find_best_hypothesis")
     @patch(f"{CLASS_UNDER_TEST}.convert_to_result")
-    def test_parse_without_confidence(
+    def test_parse_with_confidence(
         self, convert_to_result, find_best_hypothesis
     ):
         convert_to_result.return_value = {"alternative": "dummy3"}

tests/recognizers/test_whisper.py

@@ -32,7 +32,7 @@ def test_recognize_pass_arguments(self, OpenAI, BytesIO, environ):
         recognizer = MagicMock(spec=Recognizer)
         audio_data = MagicMock(spec=AudioData)
 
-        actual = whisper.recognize_whisper_api(
+        _ = whisper.recognize_whisper_api(
             recognizer, audio_data, model="x-whisper", api_key="OPENAI_API_KEY"
         )
 

tests/test_recognition.py

@@ -1,10 +1,8 @@
 #!/usr/bin/env python3
 # -*- coding: utf-8 -*-
 
-import http
 import os
 import unittest
-from unittest.mock import patch, MagicMock
 
 import speech_recognition as sr