REVAI-3855: Update python sdk to support Super API

src/rev_ai/apiclient.py

@@ -2,9 +2,13 @@
 """Speech recognition tools for using Rev AI"""
 
 import json
-from .models import Account, CaptionType, Job, Transcript
-from .baseclient import BaseClient
+
 from . import utils
+from .baseclient import BaseClient
+from .models import Account, CaptionType, Job, Transcript
+from .models.asynchronous.summarization_options import SummarizationOptions
+from .models.asynchronous.summary import Summary
+from .models.asynchronous.translation_options import TranslationOptions
 
 try:
     from urllib.parse import urljoin
@@ -66,7 +70,9 @@ def submit_job_url(
             notification_config=None,
             skip_postprocessing=False,
             remove_atmospherics=False,
-            speakers_count=None):
+            speakers_count=None,
+            summarization_config: SummarizationOptions = None,
+            translation_config: TranslationOptions = None):
         """Submit media given a URL for transcription.
         The audio data is downloaded from the URL
         :param media_url: web location of the media file
@@ -116,6 +122,8 @@ def submit_job_url(
         :param remove_atmospherics: Atmospherics such as <laugh>, <affirmative>, etc. will not
             appear in the transcript.
         :param speakers_count: Use to specify the total number of unique speakers in the audio.
+        :param summarization_config: Use to request transcript summary.
+        :param translation_config: Use to request transcript translation.
         :returns: raw response data
         :raises: HTTPError
         """
@@ -128,7 +136,9 @@ def submit_job_url(
                                                    verbatim, rush, test_mode,
                                                    segments_to_transcribe, speaker_names,
                                                    source_config, notification_config,
-                                                   skip_postprocessing)
+                                                   skip_postprocessing,
+                                                   summarization_config=summarization_config,
+                                                   translation_config=translation_config)
 
         response = self._make_http_request(
             "POST",
@@ -161,7 +171,9 @@ def submit_job_local_file(
             notification_config=None,
             skip_postprocessing=False,
             remove_atmospherics=False,
-            speakers_count=None):
+            speakers_count=None,
+            summarization_config: SummarizationOptions = None,
+            translation_config: TranslationOptions = None):
         """Submit a local file for transcription.
         Note that the content type is inferred if not provided.
 
@@ -208,6 +220,8 @@ def submit_job_local_file(
         :param remove_atmospherics: Atmospherics such as <laugh>, <affirmative>, etc. will not
             appear in the transcript.
         :param speakers_count: Use to specify the total number of unique speakers in the audio.
+        :param summarization_config: Use to request transcript summary.
+        :param translation_config: Use to request transcript translation.
         :returns: raw response data
         :raises: HTTPError, ValueError
         """
@@ -222,7 +236,9 @@ def submit_job_local_file(
                                                    language, custom_vocabulary_id, transcriber,
                                                    verbatim, rush, test_mode,
                                                    segments_to_transcribe, speaker_names, None,
-                                                   notification_config, skip_postprocessing)
+                                                   notification_config, skip_postprocessing,
+                                                   summarization_config=summarization_config,
+                                                   translation_config=translation_config)
 
         with open(filename, 'rb') as f:
             files = {
@@ -451,6 +467,160 @@ def get_account(self):
 
         return Account.from_json(response.json())
 
+    def get_transcript_summary_text(self, id_):
+        """Get the transcript summary of a specific job as plain text.
+
+        :param id_: id of job to be requested
+        :returns: transcript data as text
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/summary'.format(id_)),
+            headers={'Accept': 'text/plain'}
+        )
+        return response.text
+
+    def get_transcript_summary_json(self, id_):
+        """Get the transcript summary of a specific job as json.
+
+        :param id_: id of job to be requested
+        :returns: transcript data as json
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/summary'.format(id_)),
+            headers={'Accept': 'application/json'}
+        )
+
+        return Summary.from_json(response.json())
+
+    def get_transcript_summary_json_as_stream(self, id_):
+        """Get the transcript summary of a specific job as streamed json.
+
+        :param id_: id of job to be requested
+        :returns: requests.models.Response HTTP response which can be used to stream
+            the payload of the response
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/summary'.format(id_)),
+            headers={'Accept': 'application/json'},
+            stream=True
+        )
+
+        return response
+
+    def get_translated_transcript_text(self, id_, language):
+        """Get the translated transcript of a specific job as plain text.
+
+        :param id_: id of job to be requested
+        :param language: requested language
+        :returns: transcript data as text
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/translation/{}'.format(id_, language)),
+            headers={'Accept': 'text/plain'}
+        )
+
+        return response.text
+
+    def get_translated_transcript_text_as_stream(self, id_, language):
+        """Get the translated transcript of a specific job as a plain text stream.
+
+        :param id_: id of job to be requested
+        :param language: requested language
+        :returns: requests.models.Response HTTP response which can be used to stream
+            the payload of the response
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/translation/{}'.format(id_, language)),
+            headers={'Accept': 'text/plain'},
+            stream=True
+        )
+
+        return response
+
+    def get_translated_transcript_json(self, id_, language):
+        """Get the translated transcript of a specific job as json.
+
+        :param id_: id of job to be requested
+        :param language: requested language
+        :returns: transcript data as json
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/translation/{}'.format(id_, language)),
+            headers={'Accept': self.rev_json_content_type}
+        )
+
+        return response.json()
+
+    def get_translated_transcript_json_as_stream(self, id_, language):
+        """Get the translated transcript of a specific job as streamed json.
+
+        :param id_: id of job to be requested
+        :param language: requested language
+        :returns: requests.models.Response HTTP response which can be used to stream
+            the payload of the response
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/translation/{}'.format(id_, language)),
+            headers={'Accept': self.rev_json_content_type},
+            stream=True
+        )
+
+        return response
+
+    def get_translated_transcript_object(self, id_, language):
+        """Get the translated transcript of a specific job as a python object`.
+
+        :param id_: id of job to be requested
+        :param language: requested language
+        :returns: transcript data as a python object
+        :raises: HTTPError
+        """
+        if not id_:
+            raise ValueError('id_ must be provided')
+
+        response = self._make_http_request(
+            "GET",
+            urljoin(self.base_url, 'jobs/{}/transcript/translation/{}'.format(id_, language)),
+            headers={'Accept': self.rev_json_content_type}
+        )
+
+        return Transcript.from_json(response.json())
+
     def _create_job_options_payload(
             self,
             media_url=None,
@@ -475,7 +645,9 @@ def _create_job_options_payload(
             notification_config=None,
             skip_postprocessing=False,
             remove_atmospherics=None,
-            speakers_count=None):
+            speakers_count=None,
+            summarization_config: SummarizationOptions = None,
+            translation_config: TranslationOptions = None):
         payload = {}
         if media_url:
             payload['media_url'] = media_url
@@ -512,7 +684,7 @@ def _create_job_options_payload(
         if segments_to_transcribe:
             payload['segments_to_transcribe'] = segments_to_transcribe
         if speaker_names:
-            payload['speaker_names'] =\
+            payload['speaker_names'] = \
                 utils._process_speaker_names(speaker_names)
         if source_config:
             payload['source_config'] = source_config.to_dict()
@@ -524,6 +696,10 @@ def _create_job_options_payload(
             payload['remove_atmospherics'] = remove_atmospherics
         if speakers_count:
             payload['speakers_count'] = speakers_count
+        if summarization_config:
+            payload['summarization_config'] = summarization_config.to_dict()
+        if translation_config:
+            payload['translation_config'] = translation_config.to_dict()
         return payload
 
     def _create_captions_query(self, speaker_channel):

src/rev_ai/models/asynchronous/job.py

@@ -1,7 +1,8 @@
 # -*- coding: utf-8 -*-
 """Job model"""
-
+from .summarization_options import Summarization
 from .job_status import JobStatus
+from .translation_options import Translation
 
 
 class Job:
@@ -28,7 +29,9 @@ def __init__(
             rush=None,
             segments_to_transcribe=None,
             remove_atmospherics=None,
-            speakers_count=None):
+            speakers_count=None,
+            summarization: Summarization = None,
+            translation: Translation = None):
         """
         :param id_: unique id of job
         :param created_on: date and time at which this job was started
@@ -85,6 +88,8 @@ def __init__(
         self.segments_to_transcribe = segments_to_transcribe
         self.remove_atmospherics = remove_atmospherics
         self.speakers_count = speakers_count
+        self.summarization = summarization
+        self.translation = translation
 
     def __eq__(self, other):
         """Override default equality operator"""
@@ -120,5 +125,7 @@ def from_json(cls, json):
             rush=json.get('rush'),
             segments_to_transcribe=json.get('segments_to_transcribe'),
             remove_atmospherics=json.get('remove_atmospherics'),
-            speakers_count=json.get('speakers_count')
+            speakers_count=json.get('speakers_count'),
+            summarization=Summarization.from_json(json.get('summarization')),
+            translation=Translation.from_json(json.get('translation'))
         )

src/rev_ai/models/asynchronous/summarization_formatting_options.py

@@ -0,0 +1,7 @@
+from enum import Enum
+"""Summarization formatting options."""
+
+
+class SummarizationFormattingOptions(str, Enum):
+    PARAGRAPH = "paragraph"
+    BULLETS = "bullets"

src/rev_ai/models/asynchronous/summarization_job_status.py

@@ -0,0 +1,10 @@
+# -*- coding: utf-8 -*-
+"""Enum for Summarization Job statuses"""
+
+from enum import Enum
+
+
+class SummarizationJobStatus(str, Enum):
+    IN_PROGRESS = "in_progress"
+    FAILED = "failed"
+    COMPLETED = "completed"

src/rev_ai/models/asynchronous/summarization_options.py

@@ -0,0 +1,63 @@
+from .summarization_formatting_options import SummarizationFormattingOptions
+from .summarization_job_status import SummarizationJobStatus
+from ..nlp_model import NlpModel
+
+"""Summarization request options."""
+
+
+class SummarizationOptions:
+    def __init__(
+            self,
+            prompt: str = None,
+            model: NlpModel = None,
+            formattingType: SummarizationFormattingOptions = None
+    ):
+        self.prompt = prompt
+        self.model = model
+        self.type = formattingType
+
+    def to_dict(self):
+        """Returns the raw form of the url data object as the api
+        expects them"""
+        dict_result = {}
+        if self.prompt:
+            dict_result['prompt'] = self.prompt
+        if self.model:
+            dict_result['model'] = self.model
+        if self.type:
+            dict_result['type'] = self.type
+
+        return dict_result
+
+
+"""Summarization options."""
+
+
+class Summarization(SummarizationOptions):
+    def __init__(
+            self,
+            prompt: str = None,
+            model: NlpModel = None,
+            formattingType: SummarizationFormattingOptions = None,
+            status: SummarizationJobStatus = None,
+            completed_on: str = None,
+            failure: str = None
+    ):
+        super().__init__(prompt, model, formattingType)
+        self.status = status
+        self.completed_on = completed_on
+        self.failure = failure
+
+    @classmethod
+    def from_json(cls, json):
+        if json is None:
+            return None
+
+        return cls(
+            json.get('prompt'),
+            json.get('model'),
+            json.get('type'),
+            json.get('status'),
+            json.get('completed_on'),
+            json.get('failure')
+        )