[Libreoffice-bugs] [Bug 144294] CALC *URGENT* probably fatal failure of the internal file structure =IFERROR(IF(SEARCH(F$1; $C5)<1; ""; 1); "") now returns incorrect values

[bugzilla-daemon at bugs.documentfoundation.org]

https://bugs.documentfoundation.org/show_bug.cgi?id=144294

--- Comment #9 from Colin <that.man.colin at gmail.com> ---
(In reply to Mike Kaganski from comment #7)
> I open the two completely different attachments. The "Version 6 Save" has
> plain "4" in Stats.M4, while "Version 7 Save" has "5" there. That cell is
> used in the formula in 'Version 7 Save'.AllHits.E1; but 'Version 6
> Save'.AllHits.E1 refers to Stats.N4, which is a formula returning a string
> "04". The AllHits.E1 is used in AllHits.E6... how could one compare?

They're a three week old version that worked fine with "formulae follow me"
applied and a today version that fails and the only difference I can identify
is that I originally forgot to set "follow me" with v.7 and then realised my
omission and subsequently set it .
If you look, even though the underlying data is different, the test is still to
discover whether the "zero padded number" in the absolute source cell appears
anywhere in the relative string of the comparison cell.
It Doesn't
Look at the two images, the numbers on that sheet - regardless of their
original source are not found in the target string for which they are compared.
In the v.6 sheet the object cell returns a null value but in v.7 it claims to
have found the string and returns 1.
The problem being, the string doesn't match in v.7 but returns a positive test
value.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice-bugs/attachments/20210904/427a3f4f/attachment.htm>