How to hook into WWW
You still do not have a browser on your system
In that case you type
telnet info.cern.ch
A WWW session will be started.
Browse through the documentation, the answers to all your questions
are really there.
In case you prefer a manual, fetch the_www_book.tex or the_www_book.ps.Z
in pub/www/doc on info.cern.ch via anonymous ftp.
How to install the linemode browser on your system
Sources and/or binaries are available via anonymous ftp from info.cern.ch
in the directories pub/www/src and pub/www/bin.
Installing the binary is the easiest way, if the binary is not available
or does not work fetch the source, go to the appropriate directory and
type make.
Modify the homepage default.html to your needs and put the name including
the path in the environment variable WWW_HOME.
You may also code the pathname in the main program.
How to install the server on your system
Fetch the source from info.cern.ch and compile.
Create the following entry in /etc/inetd.conf
http stream tcp nowait nobody /httpd httpd -l -r
Create the following entry in /etc/services
http 80/tcp # WorldWideWeb server
Put the names of the directories to be accessed by the daemon in httpd.conf.
In the example below all html files are in the directory /user/a03/www/default.
<![CDATA[
#
map /hypertext/* //cernvax/userd/tbl/hypertext/*
map /decnet/* //cernvax/userd/tbl/denise/*
map /rpc/* //cernvax/userd/tbl/rpc/*
map //* /Net/*
pass /Net/cernvax/userd/tbl/hypertext/*
pass /Net/cernvax/userd/tbl/denise/*
pass /Net/cernvax/userd/tbl/rpc/*
pass /user/a03/www/default/*
fail *
]]>
Keyword search
The code provided by CERN will not perform a keyword search.
The following modification in HTRetrieve.c makes this possible:
<![CDATA[
if (keywords) {
/*
if (TRACE) fprint("HTHandle: can't perform search %s\n",
arg);
HTWriteASCII(soc,
"Sorry, this server does not perform searches.\n");
*/
ip = string;
command = "/user/a03/bin/WWW.sh ";
for (;*command!='\0';) *ip++ = *command++;
for (;*arg!='\0';) *ip++ = *arg++;
*ip++ = '?';
for (;*keywords!='\0';) *ip++ = *keywords++;
system (string);
return fd;
}
]]>
In this example the server will call the script WWW.sh in directory
/user/a03/bin:
<![CDATA[
name=`echo $@ | awk -F? '{print $1}'`
keys=`echo $@ | awk -F? '{print $2}'`
name=`basename $name .html`
/user/a03/bin/$name.sh $keys
]]>
A keyword search is possible for html files which contain the command
<![CDATA[<isindex>]]>
When the keywords key1 and key2 are given the server will receive the string
pathname/file.html?key1+key2.
The script WWW.sh will call the script file.sh with the parameter key1+key2.