[haiku-development] Re: Practical questions from a new developer

  • From: John Scipione <jscipione@xxxxxxxxx>
  • To: haiku-development@xxxxxxxxxxxxx
  • Date: Fri, 23 Apr 2010 15:01:26 -0400

I created three tiny patches and posted them to ticket #5203


Each of these patches can be applied separately. The first patch increases
the precision of pi and e by using the MAPM version of these constants
instead of the one from math.h. This patch is definitely needed. The next
two are optional, just little enhancements. The first one changes log to ln
and log10 to just log. I think this is what most people expect from a
calculator. The third patch just prints "error" on an error instead of
printing a detailed error message. I don't really know if this third one is
really useful or not but I'll leave it up to you guys.

John Scipione

On Fri, Apr 23, 2010 at 12:38 PM, Wim van der Meer

> On Sat, Apr 24, 2010 at 12:45 AM, John Scipione <jscipione@xxxxxxxxx>wrote:
>> As the author of the original Deskcalc patch I'd like to make a few
>> comments. First of all, I like your approach because you don't have to
>> recalculate the result to convert to scientific notation. I hope that my
>> patch was at least helpful in giving you guidance in creating your patch.
>> Congratulations on getting your patch accepted. However, I still think that
>> there are some really nice aspects to my patch that got overlooked. For
>> instance my patch allows you to specify the number of significant digits
>> that you want calculated instead of just using 32. Also, I made a few fixes
>> that allow you to get more precise results for tranendental functions (like
>> pi, sin(rad), etc) and I don't think those changes got put in to your patch.
>> I'll apply your patch when I get a chance and see what other things got left
>> behind. However, I would like to say, good job, I think your approach is
>> overall much cleaner than mine.
>> John Scipione
> Thank you for your comments. Yes, I looked at your patch and that gave me a
> good starting point. I also agree with you that your patch has some good
> points, but I believe these did not really relate to the problem. It is hard
> to calculate the number of characters that can fit in a certain width,
> especially when proportional fonts are used, that is why I decided on my
> approach.
> It may be a good idea to create a new patch with the other improvements you
> propose and submit it.
> Regards,
> --
> Wim van der Meer
> \/\///\/\

Other related posts: