ROOT version and draw underflow bin


I have the following ROOT version:
ROOT 5.34/09 (v5-34-09@v5-34-09, Jun 26 2013, 17:10:36 on macosx64)
I realized that I can draw the underflow bin of a TH1 when I do a zoom.
The same happens in 5.34/11 as well. However it doesn’t happen in newer versions.
I was wondering if it’s just a drawing problem or if somehow it can affects
also other results. In the first case I would not install a new ROOT version
to avoid dependency problems.

Another thing. How can I see if the installed ROOT version is a development,
a production or a old version at the time of the installation?
For example this are two different RVersion.h files:

#define ROOT_RELEASE “5.34/09”
#define ROOT_RELEASE_DATE “Jun 26 2013”
#define ROOT_RELEASE_TIME “16:38:19”
#define ROOT_SVN_REVISION 49361
#define ROOT_GIT_COMMIT “v5-34-08-80-g1323c6f”
#define ROOT_GIT_BRANCH “heads/v5-34-00-patches”
#define ROOT_VERSION_CODE 336393
#define ROOT_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + ©)

#define ROOT_RELEASE “5.34/11”
#define ROOT_RELEASE_DATE “Oct 31 2013”
#define ROOT_RELEASE_TIME “16:40:11”
#define ROOT_SVN_REVISION 49361
#define ROOT_GIT_COMMIT “v5-34-10-229-g7a28a30”
#define ROOT_GIT_BRANCH “heads/v5-34-00-patches”
#define ROOT_VERSION_CODE 336395
#define ROOT_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + ©)


That sounds like a bug in the old ROOT version … what are you doing exactly ?
Can you provide an example showing that ?


I attached an example TH1F *h =new TH1F(“h”,“h”,10,0,10) simply filled with 3 entries.
The underflow bin is shown only when I zoom it, otherwise only the range bins=1,…,Nbins is shown.
However I think the bincontent is treated correctly, for example:
root [5] h->GetBinContent(1)
(const Double_t)1.00000000000000000e+00

I’m wondering if it’s just a drawing bug. I guess also 2D contours in bin=0 are affected, since I could
see it in the underflow bin. However, does it give any problem for bins not in the border and not drawn?
I make the contour from the histogram and I take the corresponding TGraph, which is then plotted,
but I don’t see any contour point from the underflow, so I guess it’s safe (the attached plot is a histo not a graph).

p.s. Both the ROOT versions I wrote in the previous email have this issue. Are those version reliable?


Best regards,
example_2D_contour_underflow.pdf (14.1 KB)
example_histo_underflow.pdf (13.1 KB)

can you provide an example macro ? what do you mean by “zooming it”?

I attached an example macro with the outputs.
By zooming I mean set the range from the GUI of the canvas.

Best regards,
output3.pdf (14.6 KB)
output2.pdf (13.5 KB)
output.pdf (13.2 KB)
example.C (942 Bytes)