blp Python Library

By default, the Bloomberg API assumes a /ticker topic prefix (e.g., IBM US Equity). To use other identifiers like CUSIP or SEDOL, you must explicitly include the topic prefix in your security string.

Examples of explicit topic prefixes:

• CUSIP: /cusip/097023105
• SEDOL: /sedol1/2108601

If you do not specify a prefix, the API uses the defaultTopicPrefix from the SessionOptions object, which is /ticker by default. Note that the required format for a topic may vary depending on the specific subscription service being used.

# Examples of explicit topic prefixes:
/cusip/097023105
/sedol1/2108601

You can install the blp package using either pip or conda. blp provides a pythonic interface to the blpapi package and is designed for explicit separation of session management, event parsing, and event aggregation.

pip install blp
# or
conda install -c conda-forge blp

For technical details regarding the underlying Bloomberg API, use a Bloomberg Terminal to access the following resources:

• Developer's Guide: For general Bloomberg API documentation.
• Services and Schemas Reference Guide: For documentation on specific Bloomberg fields used to access data.

Access Path: From a Bloomberg Terminal, type WAPI <GO> and select API Developer's Guide.

For detailed information on using the underlying blpapi provided by Bloomberg, refer to the official Bloomberg documentation:

1. Via Bloomberg Terminal: Type WAPI <GO> and navigate to API Developer's Guide.
2. Online: Access the Bloomberg API Library at https://www.bloomberg.com/professional/support/api-library/.

The BlpParser converts various Bloomberg response types into standardized dictionary formats. Common schemas include:

1. ReferenceDataResponse: {'security': <str>, 'fields': <list of str>, 'data': <dict of field:value>}
2. HistoricalDataResponse: {'security': <str>, 'fields': <list of str>, 'data': <list of dict of field:value>}
3. IntradayBarResponse: {'security': <str>, 'events': [<str>], 'data': <list of {'time': <Timestamp>, 'open': <float>, ...}>}
4. IntradayTickResponse: {'security': <str>, 'events': <list of str>, 'data': <list of {'time': <Timestamp>, 'type': <str>, 'value': <float>, 'size': <int>}>}
5. fieldResponse: {'id': <list of str>, 'data': {<str>: {field: value}}}
6. InstrumentListResponse: {'security': <str>, 'description': <str>}
7. GetFillsResponse: {'Fills': <list of dicts>}

You can modify the behavior of a request by passing an overrides argument. Overrides are provided as a list of tuples in the format [(override_field, override_value), ...]. This is useful for changing currency, adjusting dates, or other field-specific behaviors.

# Example: Overriding currency for a BDP request
bquery.bdp(
    ["IBM US Equity"],
    ["CRNCY_ADJ_PX_LAST"],
    overrides=[("EQY_FUND_CRNCY", "EUR")]
)

# Example: Overriding an end date for a BDS request
bquery.bds(
    "DAX Index",
    "INDX_MWEIGHT_HIST",
    overrides=[("END_DATE_OVERRIDE", "20230630")]
)

To start making Bloomberg queries, instantiate blp.BlpQuery() and call .start(). You can also use blp.BlpQuery as a context manager to ensure proper cleanup.

from blp import blp

# Standard instantiation
bquery = blp.BlpQuery().start()

# Using as a context manager
with blp.BlpQuery() as bq:
    df = bq.bdh(
        ["GME US Equity"],
        ["PX_LAST", "VOLUME"],
        start_date="20210101",
        end_date="20210130",
    )

By default, invalid tickers or securities will raise errors (e.g., securityError). To prevent these errors from stopping your execution, instantiate BlpQuery with a BlpParser where raise_security_errors is set to False.

# Instantiate with a parser that ignores security errors
bquery = blp.BlpQuery(parser=blp.BlpParser(raise_security_errors=False)).start()

# This will no longer raise an exception for invalid tickers
bquery.bdh(
    ["NOT_A_TICKER US Equity", "SPY US Equity"],
    ["PX_LAST"],
    start_date="20190101",
    end_date="20190110",
)

Bloomberg allows identifying securities using SEDOLs. You can use the /sedol1/ prefix in your security list for both bdp() and bdh() calls.

# Using SEDOLs in BDP
bquery.bdp(["SPY US Equity", "TLT US Equity"], ["ID_SEDOL1"])

# Using SEDOLs in BDH
bquery.bdh(
    ["/sedol1/2840215", "/sedol1/2971546"],
    ["PX_LAST", "VOLUME"],
    start_date="20190101",
    end_date="20190110",
)

Use bquery.bds() to retrieve a set of data (e.g., index constituents or historical weights) for a security.

bquery.bds("BCOM Index", "INDX_MWEIGHT")

Use bquery.beqs() to run equity screens. It requires a screen name and a screen_type (e.g., "GLOBAL"). Additional configuration can be passed via options.

bquery.beqs(
    "Core Capital Ratios",
    screen_type="GLOBAL",
    options={"Group": "General"}
)