tidata

> ## ⚠️ Deprecated — the API behind this package is closed

>

> The TradeInsight beta data API that `tidata` calls stopped answering at

> 23:59 UTC on 2 September 2026. Every request now returns

> `410 Gone`, and no key will work again — the beta credential pool is

> switched off, and beta signup is closed, so a new key cannot be issued either.

>

> Key holders were emailed on 26 August 2026. There is nothing to migrate to

> that keeps this package working.

>

> Nothing in this package can be made to work. It has no unauthenticated

> mode: without a key it never sent an `Authorization` header, and the server

> refused the call.

>

> Where the data went: TradeInsight Desk — the

> same congressional, insider and price data, reached through an agent rather

> than a raw API.

>

> If you are reading this from a failing script, the error you want is

> `tidata.tifinance.exceptions.APIClosedError` (code `API_CLOSED`), raised in

> place of the auth error the 410 would otherwise look like.

The rest of this README is kept for anyone reading old code. **It no longer

describes anything that runs.**

Provides a `Ticker` class with a `history()` method that returns a pandas DataFrame

with yfinance-compatible column names.

Installation

pip install tidata

Or install from source:

git clone https://github.com/TradeInsight-Info/tidata.git
cd tidata
pip install -e .

Quick Start

~~Set your API key in the environment~~ — keys are no longer issued and existing

ones are rejected. The snippet below is what the client used to do:

from tidata.tifinance import Ticker

# API key is read from TIDATA_API_KEY env var automatically
t = Ticker("AAPL")

# Adjusted prices (yfinance-compatible)
df = t.history(start="2024-01-01", end="2024-12-31")
print(df.head())
#                  Open        High         Low       Close      Volume  Dividends  Stock Splits
# Date
# 2024-01-02  184.210...  185.880...  183.430...  185.200...  79047200.0        0.0           0.0

# Raw (unadjusted) prices — adds an "Adj Close" column, like yfinance
df_raw = t.history(start="2024-01-01", end="2024-12-31", auto_adjust=False)

yfinance drop-in

`Ticker.history()` mirrors yfinance for the daily-OHLCV case:

from tidata.tifinance import Ticker, download

# period shorthands: 1d 5d 1mo 3mo 6mo 1y 2y 5y 10y ytd max
df = Ticker("AAPL").history(period="1y")

# non-zero dividends / splits over full available history (period="max")
divs = Ticker("AAPL").dividends
spl = Ticker("AAPL").splits

# bulk download — flat frame for one ticker, column MultiIndex for many
one = download("AAPL", period="6mo")
many = download("AAPL MSFT", period="6mo")   # columns: (field, ticker)

Behaviour notes

• On API errors, `history()` logs a warning and returns an empty DataFrame

(yfinance behaviour). Pass `raise_errors=True` to raise the typed exception

instead.

• `period="max"` requests full available history (from a `1900-01-01` floor),

not a fixed window.

• Only `interval="1d"` is currently supported.
• The index is a tz-naive `DatetimeIndex` named `Date`.

Configuration

Exceptions

All exceptions inherit from `APIError` which exposes `.code` and `.message`.

License

MIT