https://bz.apache.org/ooo/show_bug.cgi?id=126635

--- Comment #7 from orcmid <[email protected]> ---
(In reply to j.nitschke from comment #6)
> Comment on attachment 85136 [details]
> patch to main/sw/source/core/doc/poolfmt.cxx
> 
> thank you for the patch
> but I wouldn't approve the current patch
> 1. use spaces instead of tabs
> https://wiki.openoffice.org/wiki/Cpp_Coding_Standards/FORMAT
> https://wiki.openoffice.org/wiki/Cpp_Coding_Standards/FORMAT/NoTabs
> 2. the new return path when pColl is NULL breaks functionality
> 3. as you noted in comment 2, lcl_SetNumBul is a helper of
> SwDoc::GetTextCollFromPool and the null check should be there if anywhere
> 
> @AppChecker review requests in bugzilla get automatic send to dev-list this
> has nothing to do with the issue status
> it's done to highlight patches and faster review process

If you look at the file in SVN, it employs a mix of tabs and blanks and extra
spaces at the ends of lines.  I did not cause that.

I did edit the patch to avoid doing all of that tidying at one time. I wanted
to focus on the defect.  

I am perfectly willing to remove the null check completely, since we know that
it never happens so long as the function is only executed as a helper locally.

 - - - - - - -

It happens that I had to over-ride my editor settings that replace tabs with
spaces (at 4 spaces per tab) and that trim blank ends of lines.

I will clean up that file now with just those fixes.

 - - - - - - - - - - -

I disagree about breaking functionality since I presume that pColl == NULL
breaks a pre-condition on the function, so the only way to cope with such a
defect at run-time is to get out as safely as possible.

Note that there was *never* any functionality in recent times -- it was always
returning without accomplishing anything.  How is breaking it less a
difficulty, since it has been completely broken for these past years?

-- 
You are receiving this mail because:
You are the assignee for the issue.

Reply via email to