Re: Help about VteTerminal

[Shixin Zeng <shixinzeng gmail com>]

Brian wrote:

> On Sun, 2004-12-05 at 14:01 +0800, Shixin Zeng wrote:
>  
>
> > I'm writing a Gtk+ frontend for lftp. I want to create a VteTerminal,
> > and launch "lftp" with the function vte_terminal_fork_command (). But
> > I don't know how to get the output of the child process? I need the
> > output to analysis and show it with GtkTreeView.
> >
> >
> >    
> I have tried getting the output as well with the python interface
> without any reliable results.  I seemed to think it was that python was
> too slow to keep up mostly.  Some text would be duplicated, other times
> it was missing a chunk.
>
> If you enabled the docs when you installed the vte package it generates
> html pages similar to devhelp that lists and explains all options
> availabe.  Here is the docs I found on nalins site:
> http://people.redhat.com/nalin/vte/html/
>
> I connected to the signal "cursor_moved", then used the
> vte_terminal_get_text func. here is the short listing of them:
>
> char*       vte_terminal_get_text           (VteTerminal *terminal,
>                                             gboolean (*is_selected) (VteTerminal *terminal,glong column,glong 
> row,gpointer data),
>                                             gpointer data,
>                                             GArray *attributes);
> char*       vte_terminal_get_text_range     (VteTerminal *terminal,
>                                             glong start_row,
>                                             glong start_col,
>                                             glong end_row,
>                                             glong end_col,
>                                             gboolean (*is_selected) (VteTerminal *terminal,glong column,glong 
> row,gpointer data),
>                                             gpointer data,
>                                             GArray *attributes);
> void        vte_terminal_get_cursor_position
>                                            (VteTerminal *terminal,
>                                             glong *column,
>                                             glong *row);
>
>  
Thanks.
With this method, I managed to get the output, but also the input from 
user. So I have to distinguish the input and output, but how can I 
distinguish them?

> I so far have been unable to get a signal back when a shell run command
> has finished.  I can get the "child_exited" and "eof" signals when the
> fork() terminates, but I want to be able to keep the primary fork
> running with a normal shell.  Then feed the shell the commands.  So far
> I have not been able to get feedback when the command has finished.  Any
> ideas?
>
> The only thing I have thought of so far is to chain another command to
> it that somehow signals my monitoring function that it has completed.
>
> e.g.   my_command = "some_command & my_feedback_command"
>       vte_terminal_feed_child(my_command, len(my_command))
>