- Ifeffit - millenia.cars.aps.anl.gov

Tk804.025_beta
by Bruce Ravel 15 Oct '03

15 Oct '03

Hi everyone, Guoqiang Pan has reported problems getting perl/Tk compiled and installed on his Red Hat system. Matt and Shelly both suggested that the problem was an unset environment variable and they suggested a solution. Guoqiang has told me privately that the suggestion helped and that he was then able to compile perl/Tk. Yay! He followed Matt's suggestion of using perl's CPAN interface for installing perl Tk (i.e. do "perl -MPCAN -e shell" at the command line, then "install Tk" at the cpan> prompt). This is a fine way of installing things, but Tk is a little bit different from most perl modules in that it's purpose is to draw things on the screen using X windows. X windows has some quirks and Guoqiang ran into a common stumbling block. The installation using the CPAN interface was proceeding properly until it got to the "make test" stage of the installation, which is when a bunch of tests are run to verify that the modules were built correctly. Most of the tests involve drawing things on the screen -- unsurprising since Tk is a GUI widget set. If you are logged in as a normal user, then you, the normal user, own the X windows process and only you or those to whom you have specifically granted permission can draw on the screen. Even root cannot do so unless you allow it. So, if you became root to run the CPAN interface by typing the "su" comand or the "sudo" command, you may not be able to successfully run the "make test" stage. To allow the "make test" thing to work, you need to execute the following command BEFORE becoming root and running CPAN: xhost + localhost This allows any user logged into your computer to write to the screen under X windows when you own the X windows process. Thus, it should also allow the "make test" stage to work. Alternately, you can log out as a normal user and log back in as root. Then root will own the X windows process and the "make test" stage of the CPAN installation or Tk should proceed without a problem. The good news is that once you get Tk installed, you almost certainly will not need to update it the next time that you update Athena and Artemis. The other good news (but not for Red Hat users apparently) is that installing from an RPM sidesteps this whole issue. I use SuSE (which I have long prefered to Red Hat -- two thumbs up for German engineering!) and its perl/Tk rpm works just fine. Hope that helps, B -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

1 0

RE: [Ifeffit] Tk804.025_beta
by Kelly, Shelly D. 14 Oct '03

14 Oct '03

> So what did you do for perl/Tk? Did you install from a supplied RPM? > No, I didn't install perl/Tk from an RPM. I looked for one for RedHat and didn't find it. So I just grabbed a Tar ball from CPAN. It looks like I'm using version 800.025. Shelly

1 0

RE: [Ifeffit] Tk804.025_beta
by Kelly, Shelly D. 14 Oct '03

14 Oct '03

> On Tuesday 14 October 2003 04:02 am, Guoqiang Pan wrote: > > Dear ifeffit colleagues, I would like to install the Linux version > > for Artemis in my RedHat8.0 system, as suggested in download page it > > should install first pgplot, ifeffit, and the latest version of > > perl/Tk. The first two are OK, but for perl/Tk I had some problem. I > > downloaded Tk804.025_beta.tar.gz, and unzipped the file. Using perl > > Makefile.PL to produce the Makefile file. But after type make, I had > > the problem: [root@localhost Tk804.025_beta2]# make Makefile:95: *** > > missing separator. Stop. > > > > I had tried other versions for perl/Tk, but the problem is the > > same. Please help me to solve it. Thank you very much. I attached > > the Makefile and Makefile.PL.log , the file produced after perl > > Makefile.PL > Makefile.PL.log > On this same subject, RedHat 9.0 has a broken perl installation. The environmental language variable is set incorrectly. The result is a similar error message. Using tcsh type "echo $LANG" the result should be "lang_us". You can set this variable by typing "setenv LANG lang_us". www.linuxforum.com/forums/index.php?showtopic=1481&view=new Gives some more details. I just recently installed all of these same packages on a new RedHat 9.0 system. -This was the only bump that I ran into. Shelly

3 2

Tk804.025_beta
by Guoqiang Pan 14 Oct '03

14 Oct '03

Dear ifeffit colleagues, I would like to install the Linux version for Artemis in my RedHat8.0 system, as suggested in download page it should install first pgplot, ifeffit, and the latest version of perl/Tk. The first two are OK, but for perl/Tk I had some problem. I downloaded Tk804.025_beta.tar.gz, and unzipped the file. Using perl Makefile.PL to produce the Makefile file. But after type make, I had the problem: [root@localhost Tk804.025_beta2]# make Makefile:95: *** missing separator. Stop. I had tried other versions for perl/Tk, but the problem is the same. Please help me to solve it. Thank you very much. I attached the Makefile and Makefile.PL.log , the file produced after perl Makefile.PL > Makefile.PL.log Besr Regards, Guoqiang Pan

2 1

odd Athena behavior on file import
by Bruce Ravel 14 Oct '03

14 Oct '03

Hi folks, An odd situation was pointed out to me this weekend that I wanted to mention here on the list. I see the problem I'll describe below on both Windows and linux and I presume it is a problem on all platforms. The "Open many files" dialog in Athena is very handy for importing large numbers of data files. However, there is a strange interaction between this dialog and folders with apostrophes in the folder name. As an example, suppose I put my data file in a folder called "Bruce's Folder". If I try to read that data file using the many-files dialog, I get a strange pop-up error message stating that the data file is a directory and refusing to import the file. Some additional and inscrutable error messages are written to standard output. Interestingly, the single data file dialog has no such problem. Thus, if you put your data in a directory that has an apostrophe (such as "Bruce's Folder") you will not be able to use the many-file import dialog. The two options are to avoid apostrophes in directory names or to read in the data one-by-one. This seems to be a bug in something from the standard perl/Tk library and not in my own code. I'll contact the appropriate person, but I cannot guarantee a fix any time soon. B -- Bruce Ravel ----------------------------------- ravel(a)phys.washington.edu Code 6134, Building 3, Room 222 Naval Research Laboratory phone: (1) 202 767 5947 Washington DC 20375, USA fax: (1) 202 767 1697 NRL Synchrotron Radiation Consortium (NRL-SRC) Beamlines X11a, X11b, X23b National Synchrotron Light Source Brookhaven National Laboratory, Upton, NY 11973 My homepage: http://feff.phys.washington.edu/~ravel EXAFS software: http://feff.phys.washington.edu/~ravel/software/exafs/

2 2

problems with deglitching in Athena
by Carlo U. Segre 09 Oct '03

09 Oct '03

Hi Bruce: I have been using the single point deglitching in Athena and I find some strange behavior. It seems that the only thing that counts is the horizontal cross hair. This needs to be right at the vertical position of the point to be selected. The vertical cross hair does not seem to matter. This is sort of counterintuitive and it took me a while to figure out that the program was not just selecting random points on me. Perhaps, if this is a feature of PGPLOT, it should be docmented. I am using the latest version of IFEFFIT and horae on Linux. Carlo -- Carlo U. Segre -- Professor of Physics Associate Dean for Special Projects, Graduate College Illinois Institute of Technology Voice: 312.567.3498 Fax: 312.567.3494 segre(a)agni.phys.iit.edu http://www.iit.edu/~segre

2 1

30 element detector data in athena
by Vivek S. Murthi 07 Oct '03

07 Oct '03

Hi Matt, Bruce.. Athena could not open files from NSLS beamline X10C vax machine which has data in the form of .XFS. So, I converted it to *.dat and made it look like a dat file from NSLS beamline X 11a. Athena doesnot give me any prompt for incompatability but it simply crashes and closes. The data I have is from a 13 element detector, so there are about 38 columns of data. I am not sure if this would make Athena crash. If So, do you have any suggestions to overcome this. thnks vivek Vivek S. Murthi Graduate Research Assistant 419 Egan Center Northeastern University Boston, MA-02115 Phone: 617-373-8949 >

4 8

Slight Problem with the new ARTEMIS
by Norbert Weiher 07 Oct '03

07 Oct '03

Dear Friends, I found a slight problem with the new ARTEMIS - I use the latest package for ifeffit (1.2.3) and HORAE (0.24). When you do a fit and look at the fit results for the current fit it seems that the program does not update all fit statistics the first time you do a fit. For example, when I switch from a single-shell fit to a dual-shell mode (from 4 to 8 variables), I get still 4 variables in the first run. When I re-calculate, it updates everything correctly. It's just for the number of variables - the rest seems to be updated in the first run. By the way: The chi-square maps are doing well. I am at the state of applying them to my systems. The GUI is horrible still, but the algorithm works. Thanks again to Matt for coding the library which does this work for me :) Cheers, Norbert -- Dr. rer. nat. Norbert Weiher (norbertweiher(a)yahoo.de) Institute for Chemical and Bioengineering - ETH Hönggerberg HCI E 117 - 8093 Zürich - Phone: +41 1 63 3 48 32

4 5

Feff6l on Windows
by Matt Newville 29 Sep '03

29 Sep '03

Hi Everyone, The Feff6l executable included with the Ifeffit 1.2.3 windows installer is broken. The source is OK (and I expect the version of feff6l in Carlo's new debian packages too), but it looks like an older, broken version of the binary got included with the 1.2.2 windows distribution in late August. Anyway, I have a fixed version of feff6l.exe that runs correctly. But since running feff on Windows has always been a pain -- requiring you to use the DOS command shell -- so I also added a program (runfeff6l.exe) that pops up a file browser pops up so that you can select a feff.inp file (doesn't need to be called feff.inp, either) and will run feff6l for you. This is not a complete Feff GUI, but it is easier to use than the older executable. The feff6l.bat file and the 'feff' shortcut now point to this program. Since there were questions about (and problems with) the automatic updater, I have not yet made this an 'update'. I'd like to hear some feedback (either here or by private email) on how to deal with such small changes. Forcing people to downlod an install another 28Mb setup.exe seems sort of silly. Some of the options I see are: 1. assume the updater is working for most people, and just make this an automatic update. 2. scrap the automatic updater altogether, and just make this a new Ifeffit version. 3. make a separate installer for Feff6L. Any other suggestions are welcome. If you have Ifeffit 1.2.3 installed and want to try the new Feff6l, follow these instructions: 1. download http://cars9.uchicago.edu/ifeffit/updates/ifup_002.zip 2. unzip this into your main Ifeffit directory (probably C:\Program Files\Ifeffit or C:\Ifeffit ) 3. IMPORTANT: after unzipping this file, double-click on make_bats.exe This will re-make the batch files and shortcuts that run the programs. 4. If you're interested, more details can be found in IFEFFIT_DIR\updates\readme_002.txt For now, I'll leave this version of Feff6 in this zip file, so that you can test it out if you want but aren't forced to do so. If the consensus is to make this a regular update, I can do that easily. If having feff6l is important to you, please let me know whether these binaries work for you. Any feedback will be greatly appreciated. Thanks, --Matt

2 1

new ifeffit and horae packages
by Carlo U. Segre 28 Sep '03

28 Sep '03

Hello All: After some time being unable to work on this (I skipped a few releases), I have Debian/Linux binary and source packages for ifeffit 1.2.3 and horae 024 for both the stable and testing releases of Debian. I can also make rpms of the two even though I have not tested these. The Debian packages can be accessed with the following lines in the /etc/apt/sources.list file. # X-Ray Software Archive deb http://fermi.phys.iit.edu/debian/woody binary/ #deb http://fermi.phys.iit.edu/debian/sarge binary/ deb-src http://fermi.phys.iit.edu/debian/woody source/ #deb-src http://fermi.phys.iit.edu/debian/sarge source/ Note, the testing archive has been commented out. Just uncomment it and comment the stable (woody) lines. Cheers, -- Carlo U. Segre -- Professor of Physics Associate Dean for Special Projects, Graduate College Illinois Institute of Technology Voice: 312.567.3498 Fax: 312.567.3494 segre(a)agni.phys.iit.edu http://www.iit.edu/~segre

1 0