Bloomberg Server API and Ruby/Python

Bloomberg Server API and Ruby/Python

Did you check out some questions at SO on this. It might help you

• Bloomberg API request timing out
• Asynchronous data through Bloomberg's new data API (COM v3) with Python?

Resolver is an spreadsheet implementation in IronPython and has a very good integration for Bloomberg API

• http://www.resolversystems.com/documentation/apidocs/MarketData_Bloomberg.html

Asynchronous data through Bloomberg's new data API (COM v3) with Python?

I finally figured it out. I did a fair bit of combrowse.py detective work, and I compared with the JAVA, C, C++, and .NET examples in the BBG API download. Interestingly enough the Bloomberg Helpdesk people knew pretty much null when it came to these things, or perhaps I was just talking to the wrong person.

Here is my code.

asynchronousHandler.py:

import win32com.client
from pythoncom import PumpWaitingMessages
from time import time, strftime
import constants

class EventHandler:
    def OnProcessEvent(self, result):
        event = win32com.client.gencache.EnsureDispatch(result) 
        if event.EventType == constants.SUBSCRIPTION_DATA:
            self.getData(event)
        elif event.EventType == constants.SUBSCRIPTION_STATUS:
            self.getStatus(event)
        else:
            self.getMisc(event)
    def getData(self, event):
        iterator = event.CreateMessageIterator()
        while iterator.Next():
            message = iterator.Message  
            dataString = ''
            for fieldIndex, field in enumerate(constants.fields):           
                if message.AsElement.HasElement(field):
                    element = message.GetElement(field)
                    if element.IsNull:
                        theValue = ''
                    else:
                        theValue = ', Value: ' + str(element.Value) 
                    dataString = dataString + ', (Type: ' + element.Name + theValue + ')'
            print strftime('%m/%d/%y %H:%M:%S') + ', MessageType: ' + message.MessageTypeAsString + ', CorrelationId: ' + str(message.CorrelationId) + dataString
    def getMisc(self, event):
        iterator = event.CreateMessageIterator()
        while iterator.Next():
            message = iterator.Message
            print strftime('%m/%d/%y %H:%M:%S') + ', MessageType: ' + message.MessageTypeAsString
    def getStatus(self, event):
        iterator = event.CreateMessageIterator()
        while iterator.Next():
            message = iterator.Message
            if message.AsElement.HasElement('reason'):
                element = message.AsElement.GetElement('reason')
                print strftime('%m/%d/%y %H:%M:%S') + ', MessageType: ' + message.MessageTypeAsString + ', CorrelationId: ' + str(message.CorrelationId) + ', Category: ' + element.GetElement('category').Value + ', Description: ' + element.GetElement('description').Value 
            if message.AsElement.HasElement('exceptions'):
                element = message.AsElement.GetElement('exceptions')
                exceptionString = ''
                for n in range(element.NumValues):
                    exceptionInfo = element.GetValue(n)
                    fieldId = exceptionInfo.GetElement('fieldId')
                    reason = exceptionInfo.GetElement('reason')
                    exceptionString = exceptionString + ', (Field: ' + fieldId.Value + ', Category: ' + reason.GetElement('category').Value + ', Description: ' + reason.GetElement('description').Value + ') ' 
                print strftime('%m/%d/%y %H:%M:%S') + ', MessageType: ' + message.MessageTypeAsString + ', CorrelationId: ' + str(message.CorrelationId) + exceptionString

class bloombergSource:
    def __init__(self):
        session = win32com.client.DispatchWithEvents('blpapicom.Session' , EventHandler)
        session.Start()
        started = session.OpenService('//blp/mktdata')
        subscriptions = session.CreateSubscriptionList()
        for tickerIndex, ticker in enumerate(constants.tickers):
            if len(constants.interval) > 0:
                subscriptions.AddEx(ticker, constants.fields, constants.interval, session.CreateCorrelationId(tickerIndex))
            else:
                subscriptions.Add(ticker, constants.fields, session.CreateCorrelationId(tickerIndex))   
        session.Subscribe(subscriptions)
        endTime = time() + 2
        while True:
            PumpWaitingMessages()
            if endTime < time():                
                break               

if __name__ == "__main__":
    aBloombergSource = bloombergSource()

constants.py:

ADMIN = 1
AUTHORIZATION_STATUS = 11
BLPSERVICE_STATUS = 9
PARTIAL_RESPONSE = 6
PUBLISHING_DATA = 13
REQUEST_STATUS = 4
RESOLUTION_STATUS = 12
RESPONSE = 5
SESSION_STATUS = 2
SUBSCRIPTION_DATA = 8
SUBSCRIPTION_STATUS = 3
TIMEOUT = 10
TOKEN_STATUS = 15
TOPIC_STATUS = 14
UNKNOWN = -1
fields = ['BID']
tickers = ['AUD Curncy']
interval = '' #'interval=5.0'

For historical data I used this simple script:

import win32com.client

session = win32com.client.Dispatch('blpapicom.Session')
session.QueueEvents = True
session.Start()
started = session.OpenService('//blp/refdata')
dataService = session.GetService('//blp/refdata')
request = dataService.CreateRequest('HistoricalDataRequest')
request.GetElement('securities').AppendValue('5 HK Equity')
request.GetElement('fields').AppendValue('PX_LAST')
request.Set('periodicitySelection', 'DAILY')
request.Set('startDate', '20090119')
request.Set('endDate', '20090130')
cid = session.SendRequest(request)
ADMIN = 1
AUTHORIZATION_STATUS = 11
BLPSERVICE_STATUS = 9
PARTIAL_RESPONSE = 6
PUBLISHING_DATA = 13
REQUEST_STATUS = 4
RESOLUTION_STATUS = 12
RESPONSE = 5
SESSION_STATUS = 2
SUBSCRIPTION_DATA = 8
SUBSCRIPTION_STATUS = 3
TIMEOUT = 10
TOKEN_STATUS = 15
TOPIC_STATUS = 14
UNKNOWN = -1
stayHere = True
while stayHere:
    event = session.NextEvent();
    if event.EventType == PARTIAL_RESPONSE or event.EventType == RESPONSE:
        iterator = event.CreateMessageIterator()
        iterator.Next()
        message = iterator.Message
        securityData = message.GetElement('securityData')
        securityName = securityData.GetElement('security')
        fieldData = securityData.GetElement('fieldData')
        returnList = [[0 for col in range(fieldData.GetValue(row).NumValues+1)] for row in range(fieldData.NumValues)]
        for row in range(fieldData.NumValues):
            rowField = fieldData.GetValue(row)
            for col in range(rowField.NumValues+1):
                colField = rowField.GetElement(col)
                returnList[row][col] = colField.Value
        stayHere = False
        break
element = None
iterator = None
message = None
event = None
session = None
print returnList

Bloomberg API request timing out

I didn't really ever get around to solving this question, but we did find a workaround.

Based on a small, apparently throwaway, comment in the Server API documentation, we opted to create a second session. One session is responsible for static requests, the other for real-time. e.g.

_marketDataSession.OpenService("//blp/mktdata"); 
_staticSession.OpenService("//blp/refdata");

The means one session operates in subscription mode, the other more synchronously - I think it was this duality which was at the root of our problems.

Since making that change, we've not had any problems.

Is it possible to access (read: query) an xml data service provider (REST or SOAP) directly from the command line? If so, how?

The curl command takes a URL as its argument and issues an HTTP GET to get the resource at that address. If that HTTP request accesses a REST web service, then it displays the response from that service. You can of course build on this as your starting point.

---

Related Topics