Skip to content

Source objects

Wordcab API Source object.

AssemblyAISource dataclass

Bases: BaseSource

AssemblyAI source object using a local or remote AssemblyAI JSON file.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not valid.

Examples:

>>> from wordcab.core_objects import AssemblyAISource

>>> assemblyai_source = AssemblyAISource(filepath="path/to/assemblyai/file.json")
>>> assemblyai_source
AssemblyAISource(...)
>>> assemblyai_source.source
'assembly_ai'

Returns:

Type Description
AssemblyAISource

The AssemblyAI source object.
Source code in src/wordcab/core_objects/source.py 
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
@dataclass
class AssemblyAISource(BaseSource):
    """
    AssemblyAI source object using a local or remote AssemblyAI JSON file.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not valid.

    Examples
    --------
    >>> from wordcab.core_objects import AssemblyAISource

    >>> assemblyai_source = AssemblyAISource(filepath="path/to/assemblyai/file.json")  # doctest: +SKIP
    >>> assemblyai_source  # doctest: +SKIP
    AssemblyAISource(...)
    >>> assemblyai_source.source  # doctest: +SKIP
    'assembly_ai'

    Returns
    -------
    AssemblyAISource
        The AssemblyAI source object.
    """

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "assembly_ai"

        if self._suffix != ".json":
            raise ValueError(
                f"Please provide a valid AssemblyAI file format. {self._suffix} is not"
                " valid, it should be .json."
            )

        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote":
            self.file_object = self._load_file_from_url()

    def prepare_payload(self) -> str:
        """Prepare payload for API request."""
        self.payload = json.dumps(
            _get_assembly_utterances(json.loads(self.file_object))
        )
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Accept": "application/json",
            "Content-Type": "application/json",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "assembly_ai"

    if self._suffix != ".json":
        raise ValueError(
            f"Please provide a valid AssemblyAI file format. {self._suffix} is not"
            " valid, it should be .json."
        )

    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote":
        self.file_object = self._load_file_from_url()

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
548
549
550
551
552
553
554
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Accept": "application/json",
        "Content-Type": "application/json",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
541
542
543
544
545
546
def prepare_payload(self) -> str:
    """Prepare payload for API request."""
    self.payload = json.dumps(
        _get_assembly_utterances(json.loads(self.file_object))
    )
    return self.payload

AudioSource dataclass

Bases: BaseSource

The AudioSource object is required to create a job that uses an audio file as input.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not supported.
ValueError

If both filepath and url are provided.
TypeError

If the path is not a string or a Path object.
FileNotFoundError

If the file does not exist or is not accessible.

Examples:

>>> from wordcab.core_objects import AudioSource

>>> audio_source = AudioSource(filepath="path/to/audio/file.mp3")
>>> audio_source
AudioSource(...)

Returns:

Type Description
AudioSource

The audio source object.
Source code in src/wordcab/core_objects/source.py 
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
@dataclass
class AudioSource(BaseSource):
    """
    The AudioSource object is required to create a job that uses an audio file as input.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not supported.
    ValueError
        If both `filepath` and `url` are provided.
    TypeError
        If the path is not a string or a Path object.
    FileNotFoundError
        If the file does not exist or is not accessible.

    Examples
    --------
    >>> from wordcab.core_objects import AudioSource

    >>> audio_source = AudioSource(filepath="path/to/audio/file.mp3")  # doctest: +SKIP
    >>> audio_source  # doctest: +SKIP
    AudioSource(...)

    Returns
    -------
    AudioSource
        The audio source object.
    """

    file_object: bytes = field(init=False, repr=False)
    download: bool = field(default=False, repr=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "audio"
        if self._suffix not in AVAILABLE_AUDIO_FORMATS:
            raise ValueError(
                f"Please provide a valid file format. {self._suffix} is not valid."
            )
        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote" and self.download is True:
            self.file_object = self._load_file_from_url()
        else:
            self.file_object = None

    def prepare_payload(self) -> Dict[str, bytes]:
        """Prepare payload for API request."""
        self.payload = {"audio_file": self.file_object}
        return self.payload

    @no_type_check
    def prepare_headers(self) -> dict:
        """Prepare headers for API request."""
        self.headers = {}
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
365
366
367
368
369
370
371
372
373
374
375
376
377
378
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "audio"
    if self._suffix not in AVAILABLE_AUDIO_FORMATS:
        raise ValueError(
            f"Please provide a valid file format. {self._suffix} is not valid."
        )
    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote" and self.download is True:
        self.file_object = self._load_file_from_url()
    else:
        self.file_object = None

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
385
386
387
388
389
@no_type_check
def prepare_headers(self) -> dict:
    """Prepare headers for API request."""
    self.headers = {}
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
380
381
382
383
def prepare_payload(self) -> Dict[str, bytes]:
    """Prepare payload for API request."""
    self.payload = {"audio_file": self.file_object}
    return self.payload

BaseSource dataclass

Base class for AudioSource and GenericSource objects. It is not meant to be used directly.

Parameters:

Name Type Description Default
filepath Optional[Union[str, Path]]

Path to the local file, by default None.

 field(default=None, repr=False)
url Optional[str]

URL to the remote file, by default None.

 field(default=None, repr=False)
url_headers Optional[Dict[str, str]]

Headers to retrieve the file from the URL, by default None. Useful if the file requires authentication to be retrieved.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If neither filepath nor url are provided.
ValueError

If both filepath and url are provided.
TypeError

If filepath is not a string or a Path object.
FileNotFoundError

If filepath does not exist or is not accessible.

Attributes:

Name Type Description
source str

The source type.
source_type str

The source type.
_stem str

The stem of the file.
_suffix str

The suffix of the file.

Returns:

Type Description
BaseSource

The source object.
Source code in src/wordcab/core_objects/source.py 
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
@dataclass
class BaseSource:
    """Base class for AudioSource and GenericSource objects. It is not meant to be used directly.

    Parameters
    ----------
    filepath : Optional[Union[str, Path]], optional
        Path to the local file, by default None.
    url : Optional[str], optional
        URL to the remote file, by default None.
    url_headers : Optional[Dict[str, str]], optional
        Headers to retrieve the file from the URL, by default None.
        Useful if the file requires authentication to be retrieved.

    Raises
    ------
    ValueError
        If neither `filepath` nor `url` are provided.
    ValueError
        If both `filepath` and `url` are provided.
    TypeError
        If `filepath` is not a string or a Path object.
    FileNotFoundError
        If `filepath` does not exist or is not accessible.

    Attributes
    ----------
    source : str
        The source type.
    source_type : str
        The source type.
    _stem : str
        The stem of the file.
    _suffix : str
        The suffix of the file.

    Returns
    -------
    BaseSource
        The source object.
    """

    filepath: Optional[Union[str, Path]] = field(default=None, repr=False)
    url: Optional[str] = field(default=None, repr=False)
    url_headers: Optional[Dict[str, str]] = field(default=None, repr=False)
    source: str = field(init=False)
    source_type: str = field(init=False)
    _stem: str = field(init=False, repr=False)
    _suffix: str = field(init=False, repr=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        self.source = self.__class__.__name__
        if not self.filepath and not self.url:
            raise ValueError(
                "Please provide either a local or a remote source, respectively"
                " `filepath` or `url`."
            )
        if self.filepath and self.url:
            raise ValueError(
                "Please provide either a local or a remote source, not both `filepath`"
                " and `url`."
            )

        if self.filepath:
            if not isinstance(self.filepath, Path) and not isinstance(
                self.filepath, str
            ):
                raise TypeError(
                    "The path must be a string or a Path object, not"
                    f" {type(self.filepath)}"
                )

            if isinstance(self.filepath, str):
                self.filepath = Path(self.filepath)

            if not self.filepath.exists():
                raise FileNotFoundError(
                    f"File {self.filepath} does not exist or is not accessible."
                )

            self._stem = self.filepath.stem
            self._suffix = self.filepath.suffix
            self.source_type = "local"

        if self.url:
            if self._check_if_url_is_valid():
                filename = urllib.parse.unquote(self.url.split("/")[-1])
                self._stem = ".".join(filename.split(".")[:-1])
                self._suffix = f".{filename.split('.')[-1].split('?')[0]}"
                self.source_type = "remote"

    @no_type_check
    def _load_file_from_path(self) -> bytes:
        """Load file from local path."""
        with open(self.filepath, "rb") as f:
            return f.read()

    def _load_file_from_url(self) -> requests.Response.content:
        """Load file from URL."""
        if self.url_headers:
            file = requests.get(
                self.url, headers=self.url_headers, timeout=REQUEST_TIMEOUT
            )
        else:
            file = requests.get(self.url, timeout=REQUEST_TIMEOUT)

        return file.content

    def _check_if_url_is_valid(self) -> bool:
        """Check if URL is valid."""
        if not validators.url(self.url):
            raise ValueError(f"Please provide a valid URL. {self.url} is not valid.")

        return True

    def prepare_payload(self) -> Union[str, bytes, Dict[str, bytes]]:
        """Prepare payload."""
        raise NotImplementedError("Payload preparation is not implemented yet.")

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers."""
        raise NotImplementedError("Headers preparation is not implemented yet.")

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
def __post_init__(self) -> None:
    """Post-init method."""
    self.source = self.__class__.__name__
    if not self.filepath and not self.url:
        raise ValueError(
            "Please provide either a local or a remote source, respectively"
            " `filepath` or `url`."
        )
    if self.filepath and self.url:
        raise ValueError(
            "Please provide either a local or a remote source, not both `filepath`"
            " and `url`."
        )

    if self.filepath:
        if not isinstance(self.filepath, Path) and not isinstance(
            self.filepath, str
        ):
            raise TypeError(
                "The path must be a string or a Path object, not"
                f" {type(self.filepath)}"
            )

        if isinstance(self.filepath, str):
            self.filepath = Path(self.filepath)

        if not self.filepath.exists():
            raise FileNotFoundError(
                f"File {self.filepath} does not exist or is not accessible."
            )

        self._stem = self.filepath.stem
        self._suffix = self.filepath.suffix
        self.source_type = "local"

    if self.url:
        if self._check_if_url_is_valid():
            filename = urllib.parse.unquote(self.url.split("/")[-1])
            self._stem = ".".join(filename.split(".")[:-1])
            self._suffix = f".{filename.split('.')[-1].split('?')[0]}"
            self.source_type = "remote"

prepare_headers()

Prepare headers.

Source code in src/wordcab/core_objects/source.py 
158
159
160
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers."""
    raise NotImplementedError("Headers preparation is not implemented yet.")

prepare_payload()

Prepare payload.

Source code in src/wordcab/core_objects/source.py 
154
155
156
def prepare_payload(self) -> Union[str, bytes, Dict[str, bytes]]:
    """Prepare payload."""
    raise NotImplementedError("Payload preparation is not implemented yet.")

DeepgramSource dataclass

Bases: BaseSource

Deepgram source object using a local or remote Deepgram JSON file.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not valid.

Examples:

>>> from wordcab.core_objects import DeepgramSource

>>> deepgram_source = DeepgramSource(filepath="path/to/deepgram/file.json")
>>> deepgram_source
DeepgramSource(...)
>>> deepgram_source.source
'deepgram'

Returns:

Type Description
DeepgramSource

The Deepgram source object.
Source code in src/wordcab/core_objects/source.py 
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
@dataclass
class DeepgramSource(BaseSource):
    """
    Deepgram source object using a local or remote Deepgram JSON file.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not valid.

    Examples
    --------
    >>> from wordcab.core_objects import DeepgramSource

    >>> deepgram_source = DeepgramSource(filepath="path/to/deepgram/file.json")  # doctest: +SKIP
    >>> deepgram_source  # doctest: +SKIP
    DeepgramSource(...)
    >>> deepgram_source.source  # doctest: +SKIP
    'deepgram'

    Returns
    -------
    DeepgramSource
        The Deepgram source object.
    """

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "deepgram"

        if self._suffix != ".json":
            raise ValueError(
                f"Please provide a valid Deepgram file format. {self._suffix} is not"
                " valid, it should be .json."
            )

        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote":
            self.file_object = self._load_file_from_url()

    def prepare_payload(self) -> str:
        """Prepare payload for API request."""
        self.payload = json.dumps(
            _get_deepgram_utterances(json.loads(self.file_object))
        )
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Accept": "application/json",
            "Content-Type": "application/json",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "deepgram"

    if self._suffix != ".json":
        raise ValueError(
            f"Please provide a valid Deepgram file format. {self._suffix} is not"
            " valid, it should be .json."
        )

    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote":
        self.file_object = self._load_file_from_url()

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
613
614
615
616
617
618
619
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Accept": "application/json",
        "Content-Type": "application/json",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
606
607
608
609
610
611
def prepare_payload(self) -> str:
    """Prepare payload for API request."""
    self.payload = json.dumps(
        _get_deepgram_utterances(json.loads(self.file_object))
    )
    return self.payload

GenericSource dataclass

Bases: BaseSource

Generic source object.

The GenericSource object is required to create a job that uses a generic file as input, such as .txt or .json file.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not supported.
ValueError

If both filepath and url are provided.
TypeError

If the path is not a string or a Path object.
FileNotFoundError

If the file does not exist or is not accessible.

Examples:

>>> from wordcab.core_objects import GenericSource

>>> generic_source = GenericSource(filepath="path/to/generic/file.txt")
>>> generic_source
GenericSource(...)
>>> generic_source.file_object
b'Hello, world!'
>>> generic_source.source_type
'local'
>>> generic_source._suffix
'.txt'
>>> generic_source._stem
'file'

Returns:

Type Description
GenericSource

The generic source object.
Source code in src/wordcab/core_objects/source.py 
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
@dataclass
class GenericSource(BaseSource):
    """Generic source object.

    The GenericSource object is required to create a job that uses a generic file as input,
    such as `.txt` or `.json` file.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not supported.
    ValueError
        If both `filepath` and `url` are provided.
    TypeError
        If the path is not a string or a Path object.
    FileNotFoundError
        If the file does not exist or is not accessible.

    Examples
    --------
    >>> from wordcab.core_objects import GenericSource

    >>> generic_source = GenericSource(filepath="path/to/generic/file.txt")  # doctest: +SKIP
    >>> generic_source  # doctest: +SKIP
    GenericSource(...)
    >>> generic_source.file_object  # doctest: +SKIP
    b'Hello, world!'
    >>> generic_source.source_type  # doctest: +SKIP
    'local'
    >>> generic_source._suffix  # doctest: +SKIP
    '.txt'
    >>> generic_source._stem  # doctest: +SKIP
    'file'

    Returns
    -------
    GenericSource
        The generic source object.
    """

    file_object: bytes = field(init=False, repr=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "generic"

        if self._suffix not in AVAILABLE_GENERIC_FORMATS:
            raise ValueError(
                f"Please provide a valid file format. {self._suffix} is not valid."
            )
        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote":
            self.file_object = self._load_file_from_url()

    def prepare_payload(self) -> str:
        """Prepare payload for API request."""
        if self._suffix == ".json":
            self.payload = json.dumps({"transcript": json.loads(self.file_object)})
        elif self._suffix == ".txt":
            self.payload = json.dumps(
                {"transcript": self.file_object.decode("utf-8").splitlines()}
            )
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Accept": "application/json",
            "Content-Type": "application/json",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
292
293
294
295
296
297
298
299
300
301
302
303
304
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "generic"

    if self._suffix not in AVAILABLE_GENERIC_FORMATS:
        raise ValueError(
            f"Please provide a valid file format. {self._suffix} is not valid."
        )
    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote":
        self.file_object = self._load_file_from_url()

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
316
317
318
319
320
321
322
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Accept": "application/json",
        "Content-Type": "application/json",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
306
307
308
309
310
311
312
313
314
def prepare_payload(self) -> str:
    """Prepare payload for API request."""
    if self._suffix == ".json":
        self.payload = json.dumps({"transcript": json.loads(self.file_object)})
    elif self._suffix == ".txt":
        self.payload = json.dumps(
            {"transcript": self.file_object.decode("utf-8").splitlines()}
        )
    return self.payload

InMemorySource dataclass

In-memory source object.

The in-memory source object is a special case of the generic source object. It is used to pass a pre-processed transcript to the API.

Parameters:

Name Type Description Default
obj Union[Dict[str, List[str]], List[str]]

The in-memory object. It can be a list of strings or a dict with a transcript key and a list of strings as value.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the in-memory object does not have a transcript key.
TypeError

If the in-memory object does not have a list as value for the transcript key.
TypeError

If the in-memory object is not a list or a dict.

Examples:

>>> from wordcab.core_objects import InMemorySource

>>> transcript = {"transcript": ["SPEAKER A: Hello.", "SPEAKER B: Hi."]}
>>> in_memory_source = InMemorySource(obj=transcript)
>>> in_memory_source
InMemorySource(...)
>>> in_memory_source.obj

Returns:

Type Description
InMemorySource

The in-memory source object.
Source code in src/wordcab/core_objects/source.py 
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
@dataclass
class InMemorySource:
    """In-memory source object.

    The in-memory source object is a special case of the generic source object.
    It is used to pass a pre-processed transcript to the API.

    Parameters
    ----------
    obj : Union[Dict[str, List[str]], List[str]]
        The in-memory object. It can be a list of strings or a dict with a `transcript` key
        and a list of strings as value.

    Raises
    ------
    ValueError
        If the in-memory object does not have a `transcript` key.
    TypeError
        If the in-memory object does not have a list as value for the `transcript` key.
    TypeError
        If the in-memory object is not a list or a dict.

    Examples
    --------
    >>> from wordcab.core_objects import InMemorySource

    >>> transcript = {"transcript": ["SPEAKER A: Hello.", "SPEAKER B: Hi."]}
    >>> in_memory_source = InMemorySource(obj=transcript)
    >>> in_memory_source
    InMemorySource(...)
    >>> in_memory_source.obj

    Returns
    -------
    InMemorySource
        The in-memory source object.
    """

    obj: Optional[Union[Dict[str, List[str]], List[str]]] = field(
        default=None, repr=False
    )
    source: str = field(init=False)
    source_type: str = field(init=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        self.source = "generic"
        self.source_type = "in_memory"
        if isinstance(self.obj, dict):
            if "transcript" not in self.obj:
                raise ValueError(
                    "Please provide a valid in-memory object. It must have a"
                    " `transcript` key."
                )
            elif not isinstance(self.obj["transcript"], list):
                raise TypeError(
                    "Please provide a valid in-memory object. The `transcript` key must"
                    " be a list."
                )
        elif isinstance(self.obj, list):
            self.obj = {"transcript": self.obj}
        else:
            raise TypeError(
                "Please provide a valid in-memory object. It must be a list or a dict."
            )

    def prepare_payload(self) -> str:
        """Prepare payload for API request."""
        self.payload = json.dumps(self.obj)
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Accept": "application/json",
            "Content-Type": "application/json",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
def __post_init__(self) -> None:
    """Post-init method."""
    self.source = "generic"
    self.source_type = "in_memory"
    if isinstance(self.obj, dict):
        if "transcript" not in self.obj:
            raise ValueError(
                "Please provide a valid in-memory object. It must have a"
                " `transcript` key."
            )
        elif not isinstance(self.obj["transcript"], list):
            raise TypeError(
                "Please provide a valid in-memory object. The `transcript` key must"
                " be a list."
            )
    elif isinstance(self.obj, list):
        self.obj = {"transcript": self.obj}
    else:
        raise TypeError(
            "Please provide a valid in-memory object. It must be a list or a dict."
        )

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
234
235
236
237
238
239
240
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Accept": "application/json",
        "Content-Type": "application/json",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
229
230
231
232
def prepare_payload(self) -> str:
    """Prepare payload for API request."""
    self.payload = json.dumps(self.obj)
    return self.payload

RevSource dataclass

Bases: BaseSource

Rev.ai source object using a local or remote Rev.ai JSON file.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not valid.

Examples:

>>> from wordcab.core_objects import RevSource

>>> rev_source = RevSource(filepath="path/to/rev/file.json")
>>> rev_source
RevSource(...)
>>> rev_source.source
'rev_ai'

Returns:

Type Description
RevSource

The Rev.ai source object.
Source code in src/wordcab/core_objects/source.py 
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
@dataclass
class RevSource(BaseSource):
    """
    Rev.ai source object using a local or remote Rev.ai JSON file.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not valid.

    Examples
    --------
    >>> from wordcab.core_objects import RevSource

    >>> rev_source = RevSource(filepath="path/to/rev/file.json")  # doctest: +SKIP
    >>> rev_source  # doctest: +SKIP
    RevSource(...)
    >>> rev_source.source  # doctest: +SKIP
    'rev_ai'

    Returns
    -------
    RevSource
        The Rev.ai source object.
    """

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "rev_ai"

        if self._suffix != ".json":
            raise ValueError(
                f"Please provide a valid Rev.ai file format. {self._suffix} is not"
                " valid, it should be .json."
            )

        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote":
            self.file_object = self._load_file_from_url()

    def prepare_payload(self) -> str:
        """Prepare payload for API request."""
        self.payload = json.dumps(_get_rev_monologues(json.loads(self.file_object)))
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Accept": "application/json",
            "Content-Type": "application/json",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "rev_ai"

    if self._suffix != ".json":
        raise ValueError(
            f"Please provide a valid Rev.ai file format. {self._suffix} is not"
            " valid, it should be .json."
        )

    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote":
        self.file_object = self._load_file_from_url()

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
676
677
678
679
680
681
682
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Accept": "application/json",
        "Content-Type": "application/json",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
671
672
673
674
def prepare_payload(self) -> str:
    """Prepare payload for API request."""
    self.payload = json.dumps(_get_rev_monologues(json.loads(self.file_object)))
    return self.payload

VTTSource dataclass

Bases: BaseSource

VTT source object using a local or remote VTT file.

Parameters:

Name Type Description Default
filepath Union[str, Path]

The path to the local file.

 field(default=None, repr=False)
url str

The URL to the remote file.

 field(default=None, repr=False)

Raises:

Type Description
ValueError

If the file format is not valid.

Examples:

>>> from wordcab.core_objects import VTTSource

>>> vtt_source = VTTSource(filepath="path/to/vtt/file.vtt")
>>> vtt_source
VTTSource(...)
>>> vtt_source.source
'vtt'

Returns:

Type Description
VTTSource

The VTT source object.
Source code in src/wordcab/core_objects/source.py 
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
@dataclass
class VTTSource(BaseSource):
    """
    VTT source object using a local or remote VTT file.

    Parameters
    ----------
    filepath : Union[str, Path]
        The path to the local file.
    url : str
        The URL to the remote file.

    Raises
    ------
    ValueError
        If the file format is not valid.

    Examples
    --------
    >>> from wordcab.core_objects import VTTSource

    >>> vtt_source = VTTSource(filepath="path/to/vtt/file.vtt")  # doctest: +SKIP
    >>> vtt_source  # doctest: +SKIP
    VTTSource(...)
    >>> vtt_source.source  # doctest: +SKIP
    'vtt'

    Returns
    -------
    VTTSource
        The VTT source object.
    """

    filename: str = field(init=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        super().__post_init__()
        self.source = "vtt"

        if self._suffix != ".vtt":
            raise ValueError(
                f"Please provide a valid VTT file format. {self._suffix} is not valid,"
                " it should be .vtt."
            )

        if self.source_type == "local":
            self.file_object = self._load_file_from_path()
        elif self.source_type == "remote":
            self.file_object = self._load_file_from_url()

        self.filename = self.filepath.name  # type: ignore

    def prepare_payload(self) -> bytes:
        """Prepare payload for API request."""
        self.payload: bytes = self.file_object
        return self.payload

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {
            "Content-Disposition": f"attachment; filename={self.filename}",
        }
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
def __post_init__(self) -> None:
    """Post-init method."""
    super().__post_init__()
    self.source = "vtt"

    if self._suffix != ".vtt":
        raise ValueError(
            f"Please provide a valid VTT file format. {self._suffix} is not valid,"
            " it should be .vtt."
        )

    if self.source_type == "local":
        self.file_object = self._load_file_from_path()
    elif self.source_type == "remote":
        self.file_object = self._load_file_from_url()

    self.filename = self.filepath.name  # type: ignore

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
743
744
745
746
747
748
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {
        "Content-Disposition": f"attachment; filename={self.filename}",
    }
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
738
739
740
741
def prepare_payload(self) -> bytes:
    """Prepare payload for API request."""
    self.payload: bytes = self.file_object
    return self.payload

WordcabTranscriptSource dataclass

Wordcab transcript source object using a Wordcab transcript ID.

Parameters:

Name Type Description Default
transcript_id str

The Wordcab transcript ID to use as input.

 field(default=None)

Raises:

Type Description
ValueError

If the transcript_id is not provided.

Examples:

>>> from wordcab.core_objects import WordcabTranscriptSource

>>> wordcab_transcript_source = WordcabTranscriptSource(transcript_id="transcript_12345")
>>> wordcab_transcript_source
WordcabTranscriptSource(transcript_id=transcript_12345)

Returns:

Type Description
WordcabTranscriptSource

The Wordcab transcript source object.
Source code in src/wordcab/core_objects/source.py 
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
@dataclass
class WordcabTranscriptSource:
    """
    Wordcab transcript source object using a Wordcab transcript ID.

    Parameters
    ----------
    transcript_id : str
        The Wordcab transcript ID to use as input.

    Raises
    ------
    ValueError
        If the `transcript_id` is not provided.

    Examples
    --------
    >>> from wordcab.core_objects import WordcabTranscriptSource

    >>> wordcab_transcript_source = WordcabTranscriptSource(transcript_id="transcript_12345")
    >>> wordcab_transcript_source
    WordcabTranscriptSource(transcript_id=transcript_12345)

    Returns
    -------
    WordcabTranscriptSource
        The Wordcab transcript source object.
    """

    transcript_id: Optional[str] = field(default=None)
    source: str = field(init=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        if self.transcript_id is None:
            raise ValueError(
                "Please provide a `transcript_id` to initialize a"
                " WordcabTranscriptSource object."
            )
        self.source = "wordcab_transcript"

    def __repr__(self) -> str:
        """Representation method."""
        return f"WordcabTranscriptSource(transcript_id={self.transcript_id})"

    def prepare_payload(self) -> None:
        """Prepare payload for API request."""
        return None

    def prepare_headers(self) -> Dict[str, str]:
        """Prepare headers for API request."""
        self.headers = {"Accept": "application/json"}
        return self.headers

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
469
470
471
472
473
474
475
476
def __post_init__(self) -> None:
    """Post-init method."""
    if self.transcript_id is None:
        raise ValueError(
            "Please provide a `transcript_id` to initialize a"
            " WordcabTranscriptSource object."
        )
    self.source = "wordcab_transcript"

__repr__()

Representation method.

Source code in src/wordcab/core_objects/source.py 
478
479
480
def __repr__(self) -> str:
    """Representation method."""
    return f"WordcabTranscriptSource(transcript_id={self.transcript_id})"

prepare_headers()

Prepare headers for API request.

Source code in src/wordcab/core_objects/source.py 
486
487
488
489
def prepare_headers(self) -> Dict[str, str]:
    """Prepare headers for API request."""
    self.headers = {"Accept": "application/json"}
    return self.headers

prepare_payload()

Prepare payload for API request.

Source code in src/wordcab/core_objects/source.py 
482
483
484
def prepare_payload(self) -> None:
    """Prepare payload for API request."""
    return None

YoutubeSource dataclass

Youtube source object using a Youtube video url.

Parameters:

Name Type Description Default
url str

The Youtube video url to use as input.

 field(default=None)

Examples:

>>> from wordcab.core_objects import YoutubeSource

>>> youtube_source = YoutubeSource(url="https://www.youtube.com/watch?v=12345")
>>> youtube_source
YoutubeSource(url='https://www.youtube.com/watch?v=12345')

Returns:

Type Description
YoutubeSource

The Youtube source object.
Source code in src/wordcab/core_objects/source.py 
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
@dataclass
class YoutubeSource:
    """
    Youtube source object using a Youtube video url.

    Parameters
    ----------
    url : str
        The Youtube video url to use as input.

    Examples
    --------
    >>> from wordcab.core_objects import YoutubeSource

    >>> youtube_source = YoutubeSource(url="https://www.youtube.com/watch?v=12345")  # doctest: +SKIP
    >>> youtube_source
    YoutubeSource(url='https://www.youtube.com/watch?v=12345')

    Returns
    -------
    YoutubeSource
        The Youtube source object.
    """

    url: str = field(default=None)
    source: str = field(init=False)
    source_type: str = field(init=False)

    def __post_init__(self) -> None:
        """Post-init method."""
        if self.url is None:
            raise ValueError(
                "Please provide a `url` to initialize a YoutubeSource object."
            )

        if not _is_youtube_link(self.url):
            raise ValueError(
                f"Please provide a valid Youtube URL. {self.url} is not valid. "
                "Use AudioSource instead if you want to use a signed URL."
            )

        self.source = "youtube"
        self.source_type = "remote"

__post_init__()

Post-init method.

Source code in src/wordcab/core_objects/source.py 
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
def __post_init__(self) -> None:
    """Post-init method."""
    if self.url is None:
        raise ValueError(
            "Please provide a `url` to initialize a YoutubeSource object."
        )

    if not _is_youtube_link(self.url):
        raise ValueError(
            f"Please provide a valid Youtube URL. {self.url} is not valid. "
            "Use AudioSource instead if you want to use a signed URL."
        )

    self.source = "youtube"
    self.source_type = "remote"
Last update: 2023-09-25
Created: 2023-09-25