Imagmagick will allow you to do this. First split the original animation into the individual frames by doing

 convert anim.gif frames%02d.gif 

Then resize the frames while recombining them as

 convert frames* -delay 200 -coalesce -resize 500x500 newanim.gif

Here the value of delay indicates 2 sec delay between frames. The -coalesce argument is required to so that frames (which are generally of different sizes) are combined properly, rather than mispositioned

To get rhythmbox to play MP3's on Fedora Core 3 you will need gstreamer-plugins-mp3. However for this to install it requires a number of packages. These can be obtained from rpm.livna.org and freshrpms.net

• libid3tag
• libmad
• lame

Another aspect is that (at least on my system) rhythmbox will consistently crash when an MP3 is started giving the error: rhythmbox: pcm.c:2094: snd_pcm_wait: Assertion `err == 1' failed. This bug has been noted here. One way t get around this is to start gstreamer-properties from the command line and change the default sink from ALSA to something else. I used OSS and now rhythmbox plays fine.

The proces is similar to the above. You'll need lame and taglib which you can get from the livna repository. In case of Fedora Core 4 all you need to do is to add the livna repository to your yum configuration. This can be done by

 rpm -ivh http://rpm.livna.org/fedora/4/i386/RPMS.lvn/livna-release-4-0.lvn.5.4.noarch.rpm
 wget http://rpm.livna.org/RPM-LIVNA-GPG-KEY-i386
 rpm --import RPM-LIVNA-GPG-KEY-i386
 

Once you've done this you can get the gstreamer MP3 plugin by doing

 yum install gstreamer-plugins-mp3
 

For Fedora Core 3, since livna is not a standalone repository it has to be used in conjunction with Fedora Extras a described here

The easiest way is to set up yum to get everything from the gstreamer repository. The instructions for this are given here

To play RTSP streams (Real streams) MPlayer must be compiled with the LiveMedia libraries. This page describes how to compile MPlayer with the libraries.

However when I tried to compile MPlayer with the October and September builds of the LiveMedia libraries, I would get an error regarding a protected class in UsageEnvironment.hh. To fix this go to the LiveMedia source direcory and open up UsageEnvironment/include/UsageEnvironment.hh and comment out line 77. Recompile the LiveMedia libraries and then recompile MPlayer and it works fine.

Use the dumpstream option of mplayer as: mplayer -dumpstream rtsp://path_to_stream. The stream can then be converted to wav format by doing: mplayer -ao pcm stream.dump

The output file is audiodump.wav which can then be converted to MP3 by doing : lame -m s -q 2 --vbr-new -V 1 -B 320 audiodump.wav song.mp3

(NOTE that the MP3 encoding command line does not include tags)

To play Real streams without buffering pauses: mplayer -dumpstream -dumpfile xxx.rm rtsp://where/the/stream/is and then wait 5 minutes and from another console do: mplayer xxx.rm (Of course you loose the immediacy, 5 minutes worth at least)

cat 1.avi 2.avi | mencoder -noidx -ovc copy -oac copy -o output.avi -

here

mplayer file.wma -ao pcm -aofile file.wav

At the beginning of your data file add the line

 copy table_name from stdin with delimiter as '\t' null as 'null;
 

and at the end of your data file add

 \.
 
 

The simply do

 psql -U username -W -d dbname < datafile.txt

When configuring PostgreSQL as described here make sure that the other lines in pg_hba.conf are commented out (or at least exclude the bugs user)

In case of the bash shell you can use the following

 ~: (command   3>&1 >stdout.log 2>&3  3>&-) >stderr.log 

 # svnadmin create --fs-type fsfs /path/to/newrepos
 # svnadmin dump /path/to/oldrepos | svnadmin load /path/to/newrepos 

See here for more details

This situation can arise if you use SVN to version control binary documents (Word documents for example). In this case there are two ways we can take a look at a previous revision.

Say your current revision of text.doc is 38 and you want to take a look at revision 37. This is achieved by doing

 svn cat -r 37 text.doc > oldversion.doc

Once you're done with oldversion.doc you can get rid of it.

An alternative route, which is more applicable if you want to undo previous changes is described here. In short, to go back to revision 37 we would do

 svn merge -r 38:37 text.doc

Thus your current copy (i.e., the copy in the current working directory) is now revision 37. However the repository still contains revision 38. So if you decide you really want to keep revision 38 simply do

 svn revert text.doc

and your working copy will be updated to the latest revision in the repository.

For simply viewing older versions of a binary doc, the first approach is much easier.

If you're using CUPS then download the CUPS-PDF package from here . After it's installed restart CUPS and point your browser to http://localhost:631 and click on 'Manage Printers' and then click on 'Add Printers'. Give a name and description and then select the 'Virtual PDF' device and then for make and model select the Postscript and Color Postscript options respectively. Now applications will see the printer you named above and if you print to it you should get a PDF file under /var/spool/cups-pdf/USERNAME/ where USERNAME is your user name on the system.

This is useful if you want to convert documents (say from OpenOffice ) to PDF format without starting the OpenOffice GUI. Simply do

  ooffice -pt "PDFWriter" stuff.doc
  

Here "PDFWriter" is the name of the PDF printer you have set up via CUPS. In fact the above commandline will take any file format the Ooffice recognizes and dumps PDF formatted versions in the location mentioned above.

First install the Vmware RPM (or tarball) and then open up vmware-config.pl and look for the section:

  $header_page_offset = direct_command(
       shell_string($gHelper{'echo'}) . ' '
       . shell_string('#define __KERNEL__' . "\n" . '#include
  <
  asm/page.h
  >
  '
                      . "\n" . $pattern . ' __PAGE_OFFSET') . ' | '
       . shell_string($gHelper{'gcc'}) . ' ' . shell_string('-I' . $answer)
       . ' -E - | ' . shell_string($gHelper{'grep'}) . ' '
       . shell_string($pattern));
     chomp($header_page_offset);
     $header_page_offset =~ s/^$pattern \(0x([0-9a-fA-F]{8}).*$/$1/;
     if ($header_page_offset =~ /[0-9a-fA-F]{8}/) {
       # We found a valid page offset
       if (not (lc($header_page_offset) eq lc($gSystem{'page_offset'}))) {
         if ($source eq 'user') {
           print wrap('The kernel defined by this directory of header files
           does ' . 'not have the same address space size as your
           running ' . 'kernel.' . "\n\n", 0);
         }
         return '';
       }
     }
  

and comment the final return statement. Once you have that download the any.any update from here and untar it. Go into the directory and run runme.pl and follow instructions.

1. On A, 'ssh-keygen -t rsa' and press enter for everythin
2. On A, 'scp ~/.ssh/id_rsa.pub B:/home/username/'
3. ssh to B (you'll be asked for your password)
4. On B, 'ssh-keygen -t rsa' and press enter for everything
5. On B, 'cat id_rsa.pub > > ~/.ssh/authorized_keys2'
6. On B, 'scp ~/.ssh/id_rsa.pub A:/home/username'
7. Come back to A and do 'cat id_rsa.pub > > ~/.ssh/authorized_keys2'

After all this is done you should be able to ssh to B from A (and to A from B) without being asked for a password

If you need to connect to a POP/NNTP/IMAP server which is SSL enabled but your client does not support SSL (eg. pan) you'll need to setup stunnel.

After having installed stunnel you can create your own SSL certificates (here or here) but I went along with whatever came in the tarball. You'll need to edit /etc/stunnel/stunnel.conf or else copy stunnel.conf.sample that came with the tarball to this location. If you created your own certificates you should uncomment the related lines. My stunnel.conf file looks like this:

 cert = /etc/stunnel/stunnel.pem
 chroot = /var/run/stunnel/
 pid = /stunnel.pid
 setuid = nobody
 setgid = nobody
 debug = 7
 output = stunnel.log
 client = yes
 foreground = no
 [nntps]
 accept = localhost:119
 connect = news.psu.edu:563
 

This is setup so my NNTP client (pan) can acess my university NNTP server which is SSL enabled. Make sure that /var/run/stunnel exists and is writable by the user and group that stunnel will run under (in this case nobody). Once this is all done, as root do /usr/sbin/stunnel and it should start running. I start it from /etc/rc.d/rc.local so that it starts at bootup.

Finally, start up pan and in the server configuration, rather than specifying the real NNTP server, use localhost as the host and 119 for the port and you should be able to access the news server

find /var/db/pkg/ -mindepth 2 -maxdepth 2 -printf "%P\n"

find /var/db/pkg/ -mindepth 2 -maxdepth 2 -printf "%P\n" | grep < your package >

The first thing is to get Mozilla or Firefox . There are multiple ways ads can be blocked.

• Get a list of hosts that serve ads and make it your /etc/hosts where each hostname is set to 127.0.0.1. You can get such a list here
• Use a stylesheet which blocks various HTML elements responsible for showing ads. You can get one such CSS file here . Its for Galeon but will probably work for Mozilla and Firefox .
• Get the Adblock plugin.

to allow a user foo to connect to a database bar with password eek start MySQL as root and then do

grant all privileges on bar.* to foo@localhost identified by 'eek' with grant option;

Use the listings package. Also see here for a good collection of colorings and style definitions for some languages

There are a number of packages that can do this (floatflt, picins and wrapfig). I use wrapfig and an example is

 \begin{wrapfigure}{l}{.48\linewidth}
     \begin{center}
         \includegraphics[width=\linewidth]{picturefilename}
         \caption{Some caption}
         \label{fig:somelabel}
     \end{center}
 \end{wrapfigure}

The 'pound' bug in dvips: When converting a DVI file to a PDF file you need to use dvips to generate a PS file first. However, if you use

 dvips -Ppdf file.dvi -o file.ps

to generate the PS file, anywhere the string fi occurs in the original TeX file it will be converted into a pound symbol in the final PDF. To get around this append the -G0 flag to the above dvips command line

Sharper images in PDF files using LaTeX: When using ps2pdf to convert a Postscript file to PDF, eps images in the original document are downsampled. Thus they look fuzzy in the final document (and crappy when you view the page full screen). To get around this add

 -dPDFSETTINGS=/prepress -dAutoFilterColorImages=false -dColorImageFilter=/FlateEncode

to the psp2df command line

To get ticks or crosses use the Zapf Dingbats font.

 \usepackage{pifont}
 \ding{51} %% tick 
 \ding{55} %% cross

http://www.th.physik.uni-frankfurt.de/~tauth/uni/poster.html
http://fuse.pha.jhu.edu/~wolven/posters.html
http://www.phys.ufl.edu/~pjh/posters/poster_howto_UF.html
http://www.astro.gla.ac.uk/users/norman/docs/posters/
http://www.met.ed.ac.uk/~hcp/index.shtml
http://www.robots.ox.ac.uk/~awf/latex-posters/
http://andreas.welcomes-you.com/projects/a0poster/
http://www-aix.gsi.de/~tbaumann/tex_poster.html
http://www.physik.uni-augsburg.de/theo3/Comp/hp750c/computing_hp750c_A0.en.shtml

 \usepackage{amsmath}
 
 \begin{equation}
   x_{\substack{a\\b}}
 \end{equation}
  

Also see The short math guide

 \usepackage{amsmath}
 \makeatletter
   \def\rightharpoonupfill@{\arrowfill@\relbar\relbar\rightharpoonup}
   \newcommand{\overrightharpoonup}{%
   \mathpalette{\overarrow@\rightharpoonupfill@}}
 \makeatother 

and then use \overrightharpoonup{abcd}. Alternatively

 \documentclass{article}
 \usepackage{esvect}
 \begin{document}
   $\vv{abcd}$
 \end{document} 

To prevent a single paragraph from being indented, place the command \noindent before it. To prevent indentation of all paragraphs set \parindent to 0 by doing: \setlength{\parindent}{0em}

The closest you can get is Adobe Helvetica and can be had by doing:

 
 \renewcommand{\rmdefault}{phv}\renewcommand{\sfdefault}{phv}

Alternatively,

 \usepackage{helvet} \renewcommand{\familydefault}{\sfdefault} 

See here. The macro that does it is reproduced here:

  \newcommand{\query}[1]{\marginpar{%
      \vskip-\baselineskip
      \raggedright\footnotesize
      \itshape\hrule\smallskip#1\par\smallskip\hrule}} 

So an example would be

      Do some stuff with the code and go 
      for dinner and come back\query{Should we come back?}
      later and finish it off 

If you're using the \emph{setspace} package then you probably want the margin notes to be single spaced, whatever the spacing of the document as a whole. In that case the firt line above would look like

 \newcommand{\query}[1]{\marginpar{\singlespacing% 

By default, horizontal rules in a longtable are generated using \hrule. However you might want the horizontal rules to be created like in a ctable (first line at the head of the table is thick and all subsequent lines thin and the footer line is again thick - all with proper spacing).

The solution is to use the booktabs package and then use \toprule, \midrule and \botomrule (equivalent to \FL, \ML and \LL from the ctable package). A snippet that shows how to get horizontal rules in a longtable look like those in a ctable

 \usepackage{longtable,booktabs}
 ...
 \begin{longtable}{lrr}
     \caption[A Caption]{A Caption}\label{lab} \\
     \toprule
     Head 1 & Head 2 & Head 3 \\
     \midrule
     \endfirsthead
     \caption[]{(continued)} \\
     \toprule
     Head 1 & Head 2 & Head 3 \\
     \midrule
     \endhead
     \bottomrule
     \endfoot
     %% Put the rows of your table here
 \end{longtable}

FTP to ftp://ftp.tex.ac.uk/tex-archive/ and check out FILES.bydate, FILES.byname, FILES.last07days

The solution to this is to open up epstopdf in an editor and add a line. Around line 195 you should see the following code

 ### open output file
 if ($::opt_gs) {
   my $pipe = "$GS -q -sDEVICE=pdfwrite $GSOPTS " .
           "-sOutputFile=$OutputFilename - -c quit";
   debug "Ghostscript pipe:", $pipe;
   open(OUT,"|$pipe") or error "Cannot open Ghostscript for piped input";
 }

Replace this with

 ### open output file
 if ($::opt_gs) {
   my $pipe = "$GS -q -sDEVICE=pdfwrite $GSOPTS " .
 	  "-dAutoRotatePages=/None " .
           "-sOutputFile=$OutputFilename - -c quit";
   debug "Ghostscript pipe:", $pipe;
   open(OUT,"|$pipe") or error "Cannot open Ghostscript for piped input";
 }

 plot(1:10,1:10, main=substitute( R^{2} == var, list(var=var)))

First install the SSOAP, RCurl and XML packages. The next step is to have a web service and methods to call. For this example we use the ToxTree web service.

 library(SSOAP)
 s <- SOAPServer('http://blue.chem.psu.edu','/tox/services/toxTreeWS', 8080)
 ret <- .SOAP(s, 'getCramerClass', smiles='CCN', action="urn:xmethods-getCramerClass')
 

The return value of this web services' method is a string which has two parts seperated by a '#'. You can use strsplit to get at the individual components

 plot(x,y, xlab='HN Output', ylab=expression(paste('Observed ', log,' (',K[p],')'))

Recently I had tried installing SJava on my Fedora Core 4 system which has GCC 4.0. The install process reported the following error

 CtoJava.cweb:215: error: static declaration of 'std_env' follows non-static declaration
 CtoJava.cweb:195: error: previous declaration of 'std_env' was here 

The simple way to fix this is to untar the SJava package resulting in an SJava directory. Then open up SJava/src/RSJava/CtoJava.h and line 14 will read as

 extern JNIEnv *std_env;  /* initialized in create_Java_vm */ 

Change this to

 static JNIEnv *std_env;  /* initialized in create_Java_vm */ 

After patching the source I also had to edit the configure file since it was not finding my R shared library. The diff between my version and the original version is

 2809c2809
 < if test -f ${R_HOME}/lib/libR.so ; then
 ---
 > if test -f ${R_HOME}/bin/libR.so ; then
 2811c2811
 <   R_SHARED_LIB="-L${R_HOME}/lib -lR"
 ---
 >   R_SHARED_LIB="-L${R_HOME}/bin -lR"

In addition if you are trying to get SJava running on an AMD64 system you will need to add -fPIC to Makefile_rules.in in the SJava package. (Thanks to Egon Willighagen for pointing this out).

After all the edits are done go to the directory containing the SJava sources and then from the command line do

 #: R CMD build SJava 
 #: R CMD INSTALL -c SJava_0.68.tar.gz 

After installation I had to make two other changes. In my ~/.bash_profile I added

 LD_LIBRARY_PATH=/usr/local/lib/R/library/SJava/libs:\ 
 /usr/local/share/java/jdk1.5.0/jre/lib/i386/client:\
 /usr/local/share/java/jdk1.5.0/jre/lib/i386:\
 /usr/local/share/java/jdk1.5.0/jre/../lib/i386:$LD_LIBRARY_PATH

Also in /etc/ld.so.conf I added the line

 /usr/local/lib/R/lib

(I could have put this line in my ~/.bash_profile).

You can use ":set list" (which defaults to "nolist"), which shows tabs as ^I and puts a "$" at the end of lines. You can also define highlighting regions with "syn match" to define a single tab. You can then link this to whatever highlighting you like perhaps using the following:

 :syn match TAB_CHAR "\t"
 :hi link TAB_CHAR Error

(Thanks to Tom Chase)

To wrap a visually selected piece of the text by say, a \mathbf{} tag, use

vmap ;bf "xc\mathbf{}<esc>"xP

Note that this doesn't appear to work if you have the LaTeX-VIM package loaded

Assuming the database is called db.mdb and the structures are the column title mol we can do

db_ExportSD[ 'db.mdb', 'structures.sdf', ['mol'], [] ]

This exports all the structures to the file called structures.sdf

Assuming you have a MDB file and you want to dump the contents of a column containing MACSS fingerprints. Let the column be called FP:MACCS. Then we can do

 entry_keys = db_Entries 'somedb.mdb'
 db_ExportASCII[ 'somedb.mdb', 'fp.txt', ['FP:MACCS'], entry_keys ]

Here somedb.mdb is the name of the MOE database that you want to extract from

In case you want to extract a set of fields from a MOE DB then use the following

 entry_keys = db_Entries 'somedb.mdb'
 fields = dbv_Fields 'somedb.db'
 mask = m_id igen length fields
 mask[ [1,2,3] ] = 0
 db_ExportASCII[ 'somedb.mdb', 'vars.txt', fields|mask, entry_keys ]

Here we specify that the first three fields are not to be exported.