Dateiumleitung:
  STDERR und STDOUT in eine Datei umlenken (wichtig, erst Dateiname dann 2>&1
  command ...  > output.log 2>&1

Useful options for mswordview
to convert Doc97 to HTML

long descriptive options:
  mswordview --notablewidth --norevisions --mainonly --noannotations
short form:
  mswordview -k -r -m -a

Q: Wie kann ich erreichen, dass das Hintergrundbild einer HTML-Seite nur einmal
   dargestellt wird und nicht gekachelt.
A: Unter normalem HTML ist das derzeit (8/98) nicht moeglich. Mit CSS ist
   dieser Effekt allerdings durch folgende Zeilen machbar:
   body { background: url(image.gif); background-repeat: no-repeat; }

-------------------------------------------------
Important if you run psql, the postgres terminal inside emacs
In the emacs buffer you don't need the pager, in most
cases it is only annoying.
Turn it off with:
  \pset pager
or
  \pset pager tee
Or even better use command line paramters to force direct output:
  psql -P pager=tee -U run my_database

-------------------------------------------------
Compiled-in modules (for apache):

httpd -l
  http_core.c
  mod_so.c

httpd -M             to list loaded modules
/usr/sbin/apache -S  to check the configured virtual hosts

Customization of logging:
  Any http header can be included in the logfiles with the generic syntax:
  %{header-name}i   (for requests/input)
  %{header-name}o   (for response/output)
  A typical example of this technique is the commonly used: %{Referer}i

  Includding the IP-Address of the server:
  Under some circumstances it might be useful to have the ip address of
  the webserver in the logfile (if the server has multiple ip interfaces
  or uses multiple ip addresses)
  %A

  Exclude requests for images from the logfile if the HTML page, in which
  they are embedded is on the same server.
  .
  <FilesMatch \.(jpe?g|gif|png)$>
     SetEnvIf Referer "^http://www.my-own-domain.com/" local_referrer=1
  </FilesMatch>
  CustomLog logs/access.log combined env=!local_referrer

-------------------------------------------------
Commandline to extract the icons from Fvwm2:

sed -e "s/%small\.[^\.]\+\.xpm%//" .fvwm2rc

-------------------------------------------------
To set the date of pMar.dat to: April, 3rd 21:34 2000
use: 
     touch -t 0403213400 pMar.dat

-------------------------------------------------
Instead of the dos2unix wrapper of recode ibmpc..lat1
that also converts german umlauts use the following
one shot perl script if you only want to convert linebreaks

perl -e 'while (<>) { chomp; chop; print $_, "\n"; }' FILE

Of course you can also read the man/info pages and find out
that
      recode /CR-LF
or short
      recode /cl

does the same linebreak conversion. And as recode is taylored
for conversion its probably faster than the perl script.

BTW, 'recode ms-ansi/cl..h3' does convert the non 7Bit clean
characters to HTML entities. The h4 surface might be even more
useful, especially if you want to do it the other way round and
revert the entities into single (or for utf multibyte) characters.

  recode h4..latin1

   The fact `recode' is a filter makes it quite easy to use from within
GNU Emacs.  For example, recoding the whole buffer from the IBM-PC
charset to current charset (Latin-1 on Unix) is easily done with:

     C-x h C-u M-| recode ibmpc RET

The `C-x h' selects the whole buffer, and `C-u M-|' filters and
replaces the current region through the given shell command.  Here is
another example, binding the keys `C-c T' to the recoding of the
current region from Easy French to Latin-1 (on Unix) and the key
`C-u C-c T' from Latin-1 (on Unix) to Easy French:

     (global-set-key "\C-cT" 'recode-texte)
     
     (defun recode-texte (flag)
       (interactive "P")
       (shell-command-on-region
        (region-beginning) (region-end)
        (concat "recode " (if flag "..txte" "txte")) t)
       (exchange-point-and-mark))

Version Control:

  run-ediff-from-cvs-buffer can be used  during a merge


** Subversion **

If you don't want to use the standard repository layout that will be created
by cvs2svn with the directories trunk, branches and tags on the top level of
the repository

 svn  (this is the Unix-directory with subdirs conf, dav, db, lock, etc.)
  |
  +-- trunk
  |    |
  |    + myproject
  |
  +-- branches
  |
  +-- tags

but instead want to use the structure that is used in the examples of the
subversion book

  /
  |
  +-- myproject
  |   |
  |   +-- trunk
  |   +-- branches
  |   +-- tags
  |
  +-- otherproject  (might come in the future)
      |
      +-- trunk
      +-- branches
      +-- tags

use the following options

 --trunk=myproject/trunk --branches=myproject/branches --tags=myproject/tags

Instead of using the option with cvs2svn you can specify the --parent-dir
option of svnadmin.

To be precise it is not exactly what is used in the subversion book, because
cvs2svn will always create a named directory below the trunk, while the
subversion book structure will have the files directly there.

  /
  |
  +-- myproject
  |   |
  |   +-- trunk
  |   |   +-- myproject      ["myproject" is repeated]
  |   |       | fileA.c
  |   |       | fileB.c
  |   |
  |   +-- branches
  |   |   +-- one-branch
  |   |       | fileA.c
  |   |       | fileB.c


The subversion book:

  /
  |
  +-- myproject
  |   |
  |   +-- trunk
  |   |   | fileA.c
  |   |   | fileB.c
  |   |
  |   +-- branches
  |   |   +-- one-branch
  |   |       | fileA.c
  |   |       | fileB.c

Creating branches might be easier with the layout of the subversion book
as it is simply:

 svn copy myproject/trunk myproject/branches/new-branch

instead of

 svn copy myproject/trunk/myproject myproject/branches/new-branch