# News (Python SDK)

Retrieve news articles for any supported stock symbol.

## Making Requests

Use the `news()` method on the `stocks` resource to fetch news articles. The method supports multiple output formats:

| Output Format | Return Type                                         | Description                                                                                                                                         |
|---------------|-----------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------|
| **DATAFRAME** | `pandas.DataFrame` or `polars.DataFrame`            | Returns a DataFrame with news articles indexed by symbol (default).                                                                                 |
| **INTERNAL**  | `list[StockNews]` or `list[StockNewsHumanReadable]` | Returns a list of StockNews objects. When `use_human_readable=True`, returns a list of StockNewsHumanReadable objects with capitalized field names. |
| **JSON**      | `dict`                                              | Returns the raw JSON response as a dictionary.                                                                                                      |
| **CSV**       | `str`                                               | Writes CSV data to file and returns the filename string.                                                                                            |

<a name="news"></a>
## news

```python
def news(
    symbol: str,
    *,
    date: str | datetime.datetime = None,
    from_date: str | datetime.datetime = None,
    to_date: str | datetime.datetime = None,
    countback: int = None,
    output_format: OutputFormat = OutputFormat.DATAFRAME,
    date_format: DateFormat = None,
    columns: list[str] = None,
    add_headers: bool = None,
    use_human_readable: bool = False,
    mode: Mode = None,
    filename: str | Path = None,
) -> list[StockNews] | list[StockNewsHumanReadable] | dict | str | MarketDataClientErrorResult
```

Fetches news articles for a stock symbol. The `symbol` parameter can be passed as the first positional argument or as a keyword argument. All other parameters must be keyword-only.

### Parameters

- `symbol` (str)

  The stock symbol for which to fetch news articles.

- `date` (str | datetime.datetime, optional)

  Specific date for the news articles.

- `from_date` (str | datetime.datetime, optional)

  Start date for the date range.

- `to_date` (str | datetime.datetime, optional)

  End date for the date range.

- `countback` (int, optional)

  Number of news articles to return, counting backwards from the `to_date`.

- `output_format` ([OutputFormat](https://www.marketdata.app/docs/sdk/py/settings#output-format), optional)

  The format of the returned data. Defaults to `OutputFormat.DATAFRAME`. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- `date_format` ([DateFormat](https://www.marketdata.app/docs/sdk/py/settings#date-format), optional)

  The date format to use in the response. Defaults to `DateFormat.UNIX`. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- [`columns`](https://www.marketdata.app/docs/sdk/py/settings#columns) (optional)

  Specify which columns to include in the response. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- [`add_headers`](https://www.marketdata.app/docs/sdk/py/settings#headers) (optional)

  Whether to include headers in the response. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- [`use_human_readable`](https://www.marketdata.app/docs/sdk/py/settings#human-readable) (optional)

  Whether to use human-readable format for values. Only applies when `output_format=OutputFormat.INTERNAL`. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- `mode` ([Mode](https://www.marketdata.app/docs/sdk/py/settings#data-mode), optional)

  The data mode to use. See [Settings](https://www.marketdata.app/docs/sdk/py/settings) for details.

- `filename` (str | Path, optional)

  File path for CSV output (only used with `output_format=OutputFormat.CSV`).

### Returns

- `list[StockNews]` | `list[StockNewsHumanReadable]` | `dict` | `str` | `MarketDataClientErrorResult`

  The news articles in the requested format, or a `MarketDataClientErrorResult` if an error occurred.

### Example (DataFrame)

```python
from marketdata import MarketDataClient

client = MarketDataClient()

# Get stock news as DataFrame (default)
# symbol can be passed positionally or as keyword
df = client.stocks.news("AAPL")
# or
df = client.stocks.news(symbol="AAPL")

print(df)
```

### Example (Internal)

```python
from marketdata import MarketDataClient, OutputFormat

client = MarketDataClient()

# Get stock news as internal objects
news_list = client.stocks.news("AAPL", output_format=OutputFormat.INTERNAL)

# Access individual news article properties
for news in news_list:
    print(f"Symbol: {news.symbol}")
    print(f"Headline: {news.headline}")
    print(f"Source: {news.source}")
    print(f"Publication Date: {news.publicationDate}")
    print(f"Updated: {news.updated}")
```

### Example (JSON)

```python
from marketdata import MarketDataClient, OutputFormat

client = MarketDataClient()

# Get stock news as JSON
news = client.stocks.news("AAPL", output_format=OutputFormat.JSON)

print(news)
```

### Example (CSV)

```python
from marketdata import MarketDataClient, OutputFormat
from pathlib import Path

client = MarketDataClient()

# Get stock news as CSV
csv_file = client.stocks.news(
    "AAPL",
    output_format=OutputFormat.CSV,
    filename=Path("news.csv")
)

print(f"CSV file saved to: {csv_file}")
```

### Example (Human Readable)

```python
from marketdata import MarketDataClient, OutputFormat

client = MarketDataClient()

# Get stock news in human-readable format
news_list = client.stocks.news(
    "AAPL",
    output_format=OutputFormat.INTERNAL,
    use_human_readable=True
)

# Access individual news article properties
for news in news_list:
    print(f"Symbol: {news.Symbol}")
    print(f"Headline: {news.headline}")
    print(f"Source: {news.source}")
    print(f"Publication Date: {news.publicationDate}")
    print(f"Date: {news.Date}")
```

<a name="StockNews"></a>
## StockNews

```python
@dataclass
class StockNews:
    symbol: str
    headline: str
    content: str
    source: str
    publicationDate: datetime.datetime
    updated: datetime.datetime
```

StockNews represents a single news article for a stock, encapsulating headline, content, source, and publication information.

### Properties

- `symbol` (str): The stock symbol.
- `headline` (str): The news headline.
- `content` (str): The news article content.
- `source` (str): The news source.
- `publicationDate` (datetime.datetime): The publication date (automatically converted from timestamp).
- `updated` (datetime.datetime): The time when the news was last updated (automatically converted from timestamp).

### Notes

- The `publicationDate` and `updated` fields are automatically converted to `datetime.datetime` objects from Unix timestamps.

<a name="StockNewsHumanReadable"></a>
## StockNewsHumanReadable

```python
@dataclass
class StockNewsHumanReadable:
    Symbol: str
    headline: str
    content: str
    source: str
    publicationDate: datetime.datetime
    Date: datetime.datetime
```

StockNewsHumanReadable represents a news article in human-readable format with capitalized field names and formatted values.

### Properties

- `Symbol` (str): The stock symbol.
- `headline` (str): The news headline.
- `content` (str): The news article content.
- `source` (str): The news source.
- `publicationDate` (datetime.datetime): The publication date (automatically converted from timestamp).
- `Date` (datetime.datetime): The time when the news was last updated (automatically converted from timestamp).

### Notes

- The `publicationDate` and `Date` fields are automatically converted to `datetime.datetime` objects from Unix timestamps.
- Field names use capitalized format with underscores where applicable (e.g., `Symbol` instead of `symbol`, `Date` instead of `updated`).