[Kmymoney-devel] Review Request: Add profiles to csvplugin to retain user's settings

[Allan]

On 27/01/12 06:08, Thomas Baumgart wrote:
> Hi,
>
> on Friday 27 January 2012 01:01:43 Allan Anderson wrote:
>
>>> On Jan. 26, 2012, 6:16 p.m., Cristian Oneț wrote:
>>>> kmymoney/plugins/csvimport/convdate.cpp, line 47
>>>> <http://git.reviewboard.kde.org/r/103492/diff/2/?file=46695#file46695li
>>>> ne47>
>>>>
>>>>      What happened with this file?
>>>>      All spaces between 'if' and '(' were removed.
>>
>> This is down to astyle, I believe.  I've been suspecting that there is a
>> problem, as sometimes it takes the spaces out, and sometimes it puts the
>> space in.  In the past I have always included a space, but, because of
>> astyle, I have in the last week or so started to omit the space.  I'm
>> happy to leave them in.  When I run astyle, as you know, its output
>> indicates files that have been changed.  If I then run it again, some of
>> the same files are flagged again, although not as many.  Last time, I ran
>> it a third time, and the output was the same as the second time.
>>
>> There definitely is some inconsistency with it, in my experience at least.
>> I have to say, though, that I modified it to run only on the csvimport
>> folder, to avoid reformatting the whole project, but that's all I've done.
>>   However, it's possible that the 'official' KMM astyle has been modified
>> subsequently and that I am missing something.  I'll look at that, too.
>
> Be aware that there is a second step that is run after the astyle command did
> its thing. The comment in astyle.sh says:
>
>      # process the same set of files to replace "foreach(" with "foreach ("
>
> which is exactly what is done. Now if you run the script again, astyle changes
> "foreach (" into "foreach(" and the wrapper script changes it back. That is
> why you see a change each time you run the script.
>
> BTW, running the script before checkin is good enough. I don't even do it
> before each checkin, but now and then I simply run it and checkin the result
> without any other change.
>


Thanks for the explanation Thomas.  I'd only glanced at the code before, 
but I see what you mean.  It looks like I have to find some other 
explanation, anyway as it's not astyle causing it.

Allan