Hi Patrick,
I fixed the problem you are reporting with fitting in August in version 3.02.
This was an unfortunate problem with the TF1 copy constructor called
when making a copy of the fitted function to be stored in the histogram
list of functions.
When you specify the option "N" (do not store the fit), the problem
disappears. However, in this case, you should not call, eg
TF1 *ffi=(TF1 *)h1->GetFunction("lastShape")
the pointer ffi, in this case, will be null and you will get a crash
when calling GetChisquare.
The only thing that I can suggest is to use version 3.02.
We are currently at version 3.02/02 and expect to make a production
release by the end of November (+- one week).
Rene Brun
Patrick Jarry wrote:
>
> --------------------------------------------------------------------------------
>
> Name: brun.txt
> brun.txt Type: Plain Text (text/plain)
> Encoding: 7bit
>
> Name: patrick.jarry.vcf
> patrick.jarry.vcf Type: VCard (text/x-vcard)
> Encoding: 7bit
> Description: Card for Patrick Jarry
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:51:05 MET