Anton, You made a good point and I noticed it already for years drawing. The drawing algorithm is far from perfect and I have a few ideas to make it better. There are also some problems when the start time is not a simple time (some multiple of 10000). Nevertheless, it is not completely obvious to make the change you propose because it has to stay consistent with all the other modes (seconds, days, etc...). For the time being, I'm relaying on standard C time functions to parse the time format. If the behaviour has to change depending on the scale, I have to do some work and it's hard to know now what is the extent of this work. I hope to have time (!) to clean this up a bit in september. I'll think about your request and try to do something. Cheers Damir KOSU_FOKIN@garbo.lucas.lu.se wrote: > > Hi Rene, > > One note about present implementation of time axis labels. > Frankly speaking its drawing algorithm has to be improved and > the most important is to set a right behavior of time label > appearance. For example if now you want to see months on the > axis, a label for a new month will appear on the axis in a place > where it fits axis division proportions but not in the beginning > of the month. The latter is very important if one wants to work > with real world time series. The same goes for years, hours etc. > A simple picture illustrates what I am saying: > > 1) Current time label implementation > > October, November, > 10th 12 > > -------|------------|-------> > Oct Nov > > 2) Required > > October, November, > 1st 1st > > -------|------------|-------> > Oct Nov > > I understand that some work around has to be done and I hope this > work can coinside with Rene's work around string labels. But again, > picture 1) is what happens in ROOT now and it is very bad because > it brings a lot of misunderstanding, i.e. if a user sees label > named "Oct 98", the user thinks of course that October 98 data > start at this label but this is not true since it can be any day > in October, perhaps October 31, depending on how current algorithm > divides axis to draw labels. > > Damir, can you also comment this since you are the author of the > current implementation? > > Regards, > Anton -- ===================================================================== | Damir Buskulic | Universite de Savoie/LAPP | | | Chemin de Bellevue, B.P. 110 | | Tel : +33 (0)450091600 | F-74941 Annecy-le-Vieux Cedex | | e-mail: buskulic@lapp.in2p3.fr | FRANCE | ===================================================================== mailto:buskulic@lapp.in2p3.fr
This archive was generated by hypermail 2b29 : Tue Jan 02 2001 - 11:50:30 MET