> Hi Valeri et al, > > Uh. Who would have thought such a simple mail could spawn some much > communique. It proved to be useful the real bug has been detected and fixed. > I also tried using ROOT 2.25/03, and it doesn't work either. Same > thing - hangs. I should also mention the rest of my setup (sorry for > not doing it in my first mail): > > OS: Linux 2.2.16 > Arch: i386 > GCC: egcs-2.91.66 > > > > > > How deep is your ("/home/group/user") ? > > Very VERY deep (at least 10 level if you include symlinks). However, > that should not be a problem. My guess it is working it did not hang. 10 levels depth means may involved 1000's real remote files. Then the symbolic link may cause the real dead loop too. Are you sure there is no dead loop some where ? Just you have some link like "../newDir" and you are done. Can you check this with some small artificial directory tree. In fact what do you anticipate if such browser finds link ? I think I made the wrong advice. TFileSet class is not what you need since it first reads the entire directory tree into the memory resided object. It was designed to facilitate the database application,. There we need fast navigation and it is where the memory resided snapshot of the file system helps. Valeri > > > > > Are you sure it is not running around the globe follow the > > > > > symbolic links for example.It may take a while I think. > > > > > I have found one has to protect this class against of "black > > > > > hole deep" directories. > > I'd probably be a good idea to have some guard against expanding to > arbitrary depths. In fact, I believe the class should only expand the > current directory, and if the user requests a sub-directory, it should > then expand that one ("just in time"). > > Anyway, thanks for looking in to it. > > Yours, > > Christian ----------------------------------------------------------- > Holm Christensen Phone: (+45) 35 35 96 91 > Sankt Hansgade 23, 1. th. Office: (+45) 353 25 305 > DK-2200 Copenhagen N Web: www.nbi.dk/~cholm > Denmark Email: cholm@nbi.dk > >
This archive was generated by hypermail 2b29 : Tue Jan 01 2002 - 17:50:39 MET