Fixing Issue #38

[weshatheleopard]

No description provided.

[marksweston]

Hi @weshatheleopard

As per issue #34 I'm taking over maintenance of finance. I'll test and respond to this PR by Friday this week.

In the meantime, are there any other useful bugfixes or updates worth pushing upstream from your own fork, now that finance will be actively maintained again?

Mark W

[weshatheleopard]

Not at the moment. I will provide more fixes if I have them. :)

[marksweston]

Firstly, sorry for being later than promised with this review.

Your diagnosis of the problem and your fix make sense to me, within my limited understanding of the algorithm.

However, one of your new tests is failing for me, repeatedly.

finance @ Marks-Mac-mini (eligoenergy-master)$ ruby test/test_cashflows.rb
Run options: --seed 38870

Running:

...F..

Finished in 0.601733s, 9.9712 runs/s, 111.3451 assertions/s.

1. Failure:
   Cashflows::an array of Transactions#test_0003_should properly calculate IRR when Complex numbers >arise from calculations [test/test_cashflows.rb:36]:
   Expected: DecNum('0.085677')
   Actual: DecNum('0.085675')

6 runs, 67 assertions, 1 failures, 0 errors, 0 skips

Could you take a look at this please?

[weshatheleopard]

Hmmm, strange. On my frozen branch:

$ ruby test/test_cashflows.rb
Run options: --seed 49990

# Running:

......

Finished in 0.404069s, 14.8490 runs/s, 165.8134 assertions/s.

6 runs, 67 assertions, 0 failures, 0 errors, 0 skips

Could it be that something else that you did broke it?

[marksweston]

I haven't really done much yet :) But I'll take another look.

[marksweston]

OK, so I tried again checking out commit ce3bd6f , in a clean ruby version with a clean gemset and I'm still seeing the same error.

I'm running on Mac OSX 10.10.5 on Ruby 2.2.0 and Ruby 2.0.0. How about you?

[weshatheleopard]

Problem solved. DST and system's local time settings were to blame. Please integrate this CL

[yanowitz]

what's the next step with this pull request? i too am running into this bug.

[yanowitz]

Bump?!

[mike-sandler]

@marksweston any chance to get this merged in?

[weshatheleopard]

@marksweston Can you please incorportate this PR? It has been a year!

[weshatheleopard]

@marksweston Can we please have some progress with this PR?