On Thu, 2006-01-19 at 08:50 -0600, Mike Klinke wrote: > On Thursday 19 January 2006 08:33, brouwers roland lx wrote: > > I think I agree to. > > I will try again: > > In a program or script, I would like to call a routine, written > > in PHP, or something else that would start a browser, in which I > > can look for a file. Once the file is found, I will enter on this > > file or click on it, which will result in a returnvariable in the > > calling program. This returnvariable contains the name of the > > file,including his path, p.e. "/home/roland/Docs/NameOfFile". > > > > Suppose you have a program, which let you enter all data about a > > car. One of the fields is the name of picture of the car, with > > full path. So, wenn one enters in this field, it will start the > > calling routine, opening a browser, enabling one to look up the > > file and transfer the path/filename. > > > > This routine is done on your linux desktop when you start a new > > launcher. You can enter the full path + filename or start a > > browser. The same thing is done when you print to a file, instead > > of a printer. > > > > I hope this is clear enough. > > How would this differ from the http browser you already have? For > example, start your http browser, select: > > File -> > Open File -> > Navigate to: /usr/share/doc/coreutils-5.2.1/README > > > The browser will display the file and place the name of the file in > the navigation bar. You can also do this from the command line > like so: > > $ mozilla /usr/share/doc/coreutils-5.2.1/README > > Is this not what you've described? > > Regards, Mike Klinke > No, it is not. Let us take the example of program that lets you enter Data of a car. Once you are in the field where the name, including path, should be entered, you do not know exactly what the name of the picture is. So you push p.e. a function key or click on a button and then I would like to start the browser. Once the browser is opened, I will walk to the directory where the pictures of cars are memorized and click on the related picture. The browser will close and put the name,with full path, into a variable, $1 or whatever. In my program I will get this variable and leave the content in this field, for later use. You would do the same thing if you link a command to a launcher. Mostly if you would like to create a launcher on your desktop, you will not know exactly the name of the program or script to launch on the icon. So you open a small browser, lookup the script or program and leave it on the properties of the launcher, so that he will know in the future what program to start when you push or click the icon. A simple command like "mozilla" would do, but once you click on the wanted file, he will open it. I don't him to open it, but give it to me in a variable. I hope you follow me. Thanks Roland
