problems with ichart.yahoo.com ?

[Pumplerod]

Have people often found ichart.yahoo.com to be unavailable?


This has happened to me a few times using load_from_yahoo where I get an error like this...

IOError: after 3 tries, Yahoo! did not return a 200 for url 'http://ichart.yahoo.com/table.csv?s=SPY&a=0&b=1&c=2012&d=11&e=1&f=2013&g=d&ignore=.csv'

yet I try sometime a good while later and it does work.

[Pumplerod]

It seems that all of a sudden 'finance' needs to be included in the ichart subdomain name.

This works...

  http://ichart.finance.yahoo.com/table.csv?s=SPY&a=0&b=1&c=2012&d=11&e=1&f=2013&g=d&ignore=.csv

but this does not... (Though it used to)
  http://ichart.yahoo.com/table.csv?s=SPY&a=0&b=1&c=2012&d=11&e=1&f=2013&g=d&ignore=.csv

how would I make load_from_yahoo return the proper domain address?

[Thomas Wiecki]

Don't you love it when they just change the domains from under you?

Since behind the scenes we are just using pandas this would be the place to check. Or email yahoo finance to not break their API.

--
You received this message because you are subscribed to the Google Groups "Zipline Python Opensource Backtester" group.
To unsubscribe from this group and stop receiving emails from it, send an email to zipline+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Pumplerod]

I did find and change the pandas data.py file.  However the error pops up again here...

/Library/Python/2.7/site-packages/requests/adapters.pyc

where it is still somehow getting the host 'ichart.yahoo.com' 

I'm not too keen on changing any of the package files.  

I would expect things are broken for a lot of people everywhere and I thought I'd see more posts
regarding the situation.  Most of the time I'm doing something wrong so I'm naturally prone to 
believe that I've screwed something up somewhere.

On Tuesday, December 31, 2013 10:51:34 AM UTC-8, Pumplerod wrote:

[Thomas Wiecki]

Found this:
http://stackoverflow.com/questions/20870956/fixing-pandas-datareader-from-yahoo-finance-with-enthought-canopy

--

[Eddie Hebert]

Thomas, Pumplerod:

It looks like pandas has already merged in a fix.

Opened an issue here for us to update Zipline accordingly,
https://github.com/quantopian/zipline/issues/255

Pumplerod, in the meantime installing pandas with at least the
following commitish,
https://github.com/pydata/pandas/commit/4bb199c0a6033e6512f9b23d7df2e7a9d8994ca6
, should unblock you.

- Eddie

[Pumplerod]

I did have to also make a change to one of the zipline files...

zipline/data/benchmarks.py

On Tuesday, December 31, 2013 10:51:34 AM UTC-8, Pumplerod wrote:

[Eddie Hebert]

The ichart.yahoo.com domain seems to have been restored.

- Eddie

[S]

I'm having problems with the ichart.yahoo.com domain again. Anyone else? ichart.finance.yahoo.com also seems to not work.

[Germán Alfaro]

  I have the same issue, I cannot run any of the tutorials,

[Alexander Myltsev]

Hi all,

I’m trying to reproduce https://blog.quantopian.com/markowitz-portfolio-optimization-2/ on latest zipline v1.1.1. 

`from zipline.utils.factory import load_bars_from_yahoo` is not anymore in v1.1.1. I found package `from zipline.data.loader import load_from_yahoo`. But it crashes with error:

ConnectionError: HTTPConnectionPool(host='ichart.finance.yahoo.com', port=80): Max retries exceeded with url: /table.csv?a=11&ignore=.csv&s=IBM&b=7&e=7&d=6&g=d&f=2017&c=2007 (Caused by NewConnectionError('<requests.packages.urllib3.connection.HTTPConnection object at 0x7fa32b6b5a10>: Failed to establish a new connection: [Errno -2] Name or service not known',))

The problem seems with pandas url to Yahoo. I found that PR: https://github.com/pandas-dev/pandas/commit/4bb199c0a6033e6512f9b23d7df2e7a9d8994ca6 , and manually fixed URL to Yahoo finance in site-packages since zipline requires pandas to be <0.19. But Yahoo seems just closed that URL: https://forums.yahoo.net/t5/Yahoo-Finance-help/Is-Yahoo-Finance-API-broken/m-p/251312#M3123 .

Has anybody managed to fix the problem?