Difference between revisions of "GUI terminal widget"

From Octave
Jump to navigation Jump to search
(Insert content of https://lists.gnu.org/archive/html/octave-maintainers/2019-05/msg00115.html.)
 
(Wiki markup.)
Line 1: Line 1:
Way back in January 2012, I posted the following message about changing
+
== Background ==
the terminal widget in the GUI to handle input differently.
 
  
 +
Way back in January 2012, jwe [https://lists.gnu.org/archive/html/octave-maintainers/2012-01/msg00416.html posted the following message] about changing the terminal widget in the GUI to handle input differently.
  
http://lists.gnu.org/archive/html/octave-maintainers/2012-01/msg00416.html
+
Jwe included a simple example written with gtkmm to illustrate the idea of having the terminal widget in the GUI be in control of input and output and feed lines of text to the interpreter.
 +
To properly handle input that spans multiple lines, a push parser is used instead of the current pull parser.
 +
It worked as an example, but was not much help for Octave since we are using Qt.
 +
So now jwe has reworked it using Qt and you can find the sources here:
  
I included a simple example written with gtkmm to illustrate the idea of
+
https://hg.octave.org/jwe-qt-gui-with-push-parser
having the terminal widget in the GUI be in control of input and output
 
and feed lines of text to the interpreter.  To properly handle input
 
that spans multiple lines, a push parser is used instead of the current
 
pull parser.  It worked as an example, but was not much help for Octave
 
since we are using Qt.  So now I've reworked it using Qt and you can
 
find my sources here:
 
  
  http://hg.octave.org/jwe-qt-gui-with-push-parser
+
See the [https://hg.octave.org/jwe-qt-gui-with-push-parser/file/tip/NOTES NOTES file] for build instructions.
 +
That file also contains the following list of open questions that will need to be resolved if we are going to attempt a switch.
  
See the NOTES file for build instructions.  That file also contains the
 
following list of open questions that will need to be resolved if we are
 
going to attempt a switch.
 
  
  * With this arrangement, would the interpreter have to run in a
+
== Open questions ==
separate thread?  As the example shows, it's not absolutely necessary.
 
It could offer some advantages, but only if it is possible for the GUI
 
to do useful things while the interpreter is busy.
 
  
  * If the interpreter is not running in a separate thread but but the  
+
# With this arrangement, would the interpreter have to run in a separate thread? As the example shows, it's not absolutely necessary.  It could offer some advantages, but only if it is possible for the GUI to do useful things while the interpreter is busy.
graphics engine is, then what happens to graphics callbacks when the  
+
# If the interpreter is not running in a separate thread but but the graphics engine is, then what happens to graphics callbacks when the parser is in the middle of parsing an expression?  Or is this not an issue because separate parsers can be used even if there is only one evaluator?  Could it ultimately be possible to have the evaluator running in multiple threads?
parser is in the middle of parsing an expression?  Or is this not an  
+
# If the interpreter does run in a separate thread, we still must wait for it to calculate and return a result so we can synchronize input and output.  Otherwise, we may print the next prompt before the output from the previous expression is evaluated and printed.  You'll see this behavior if you build the example program with CALC_USE_INTERPRETER_THREAD defined.
issue because separate parsers can be used even if there is only one  
+
# The example parser currently also performs evaluations and computes results immediately so it doesn't properly handle expression lists that are split across multiple lines.  Octave wouldn't have this problem because we already build a parse tree then execute it once it is complete.
evaluator?  Could it ultimately be possible to have the evaluator  
+
# Do we need text position markers to keep track of the prompt position (beginning of current line) when inserting or clearing text? This doesn't seem necessary in the current example, but it doesn't have functions that can clear the screen or otherwise redraw prior output that would cause the position of the cursor in the window to change.
running in multiple threads?
+
# The example program doesn't attempt handle multi-line prompts or prompts with invisible characters (color specifications, for example). Fixing that will make the redisplay function significantly more complex. See, for example, how complicated the default rl_redisplay function is in the readline library.  Unless we actually write a terminal emulator (like the current terminal widgets) then it is not possible to use readline's rl_redisplay function directly.
 
+
# We'll need to implement a pager ourselves, since "less" won't work in this simplified terminal widget.
  * If the interpreter does run in a separate thread, we still must  
+
# The system function may need to be modified so that external programs that expect to be running in a terminal will continue to work properly.  On Unixy systems, this job can be done with ptys.  I guess Windows systems can use a hidden console?  But if these things are required, are we more or less back to were we were before since we used a pty and hidden console to implement the terminal widgets?  I believe the Emacs start-process function must do similar things, so we might be able to reuse that code.
wait for it to calculate and return a result so we can synchronize input  
+
# If readline runs in the terminal widget, who owns the command-line history?  Either way, if the GUI is in control of keyboard input, it will need access to the history list and Octave will also need access for the history functions.
and output.  Otherwise, we may print the next prompt before the output  
 
from the previous expression is evaluated and printed.  You'll see this  
 
behavior if you build the example program with  
 
CALC_USE_INTERPRETER_THREAD defined.
 
 
 
  * The example parser currently also performs evaluations and computes  
 
results immediately so it doesn't properly handle expression lists that  
 
are split across multiple lines.  Octave wouldn't have this problem  
 
because we already build a parse tree then execute it once it is complete.
 
 
 
  * Do we need text position markers to keep track of the prompt  
 
position (beginning of current line) when inserting or clearing text?  
 
This doesn't seem necessary in the current example, but it doesn't have  
 
functions that can clear the screen or otherwise redraw prior output  
 
that would cause the position of the cursor in the window to change.
 
 
 
  * The example program doesn't attempt handle multi-line prompts or  
 
prompts with invisible characters (color specifications, for example).  
 
Fixing that will make the redisplay function significantly more complex.  
 
  See, for example, how complicated the default rl_redisplay function is  
 
in the readline library.  Unless we actually write a terminal emulator  
 
(like the current terminal widgets) then it is not possible to use  
 
readline's rl_redisplay function directly.
 
 
 
  * We'll need to implement a pager ourselves, since "less" won't work  
 
in this simplified terminal widget.
 
 
 
  * The system function may need to be modified so that external  
 
programs that expect to be running in a terminal will continue to work  
 
properly.  On Unixy systems, this job can be done with ptys.  I guess  
 
Windows systems can use a hidden console?  But if these things are  
 
required, are we more or less back to were we were before since we used  
 
a pty and hidden console to implement the terminal widgets?  I believe  
 
the Emacs start-process function must do similar things, so we might be  
 
able to reuse that code.
 
 
 
  * If readline runs in the terminal widget, who owns the command-line  
 
history?  Either way, if the GUI is in control of keyboard input, it  
 
will need access to the history list and Octave will also need access  
 
for the history functions.
 
 
 
Now that I have an almost working (if quite simplistic) example in Qt, I
 
will attempt to modify Octave to use this approach.  From that, I expect
 
many more questions to come up.
 

Revision as of 08:22, 29 May 2019

Background

Way back in January 2012, jwe posted the following message about changing the terminal widget in the GUI to handle input differently.

Jwe included a simple example written with gtkmm to illustrate the idea of having the terminal widget in the GUI be in control of input and output and feed lines of text to the interpreter. To properly handle input that spans multiple lines, a push parser is used instead of the current pull parser. It worked as an example, but was not much help for Octave since we are using Qt. So now jwe has reworked it using Qt and you can find the sources here:

https://hg.octave.org/jwe-qt-gui-with-push-parser

See the NOTES file for build instructions. That file also contains the following list of open questions that will need to be resolved if we are going to attempt a switch.


Open questions

  1. With this arrangement, would the interpreter have to run in a separate thread? As the example shows, it's not absolutely necessary. It could offer some advantages, but only if it is possible for the GUI to do useful things while the interpreter is busy.
  2. If the interpreter is not running in a separate thread but but the graphics engine is, then what happens to graphics callbacks when the parser is in the middle of parsing an expression? Or is this not an issue because separate parsers can be used even if there is only one evaluator? Could it ultimately be possible to have the evaluator running in multiple threads?
  3. If the interpreter does run in a separate thread, we still must wait for it to calculate and return a result so we can synchronize input and output. Otherwise, we may print the next prompt before the output from the previous expression is evaluated and printed. You'll see this behavior if you build the example program with CALC_USE_INTERPRETER_THREAD defined.
  4. The example parser currently also performs evaluations and computes results immediately so it doesn't properly handle expression lists that are split across multiple lines. Octave wouldn't have this problem because we already build a parse tree then execute it once it is complete.
  5. Do we need text position markers to keep track of the prompt position (beginning of current line) when inserting or clearing text? This doesn't seem necessary in the current example, but it doesn't have functions that can clear the screen or otherwise redraw prior output that would cause the position of the cursor in the window to change.
  6. The example program doesn't attempt handle multi-line prompts or prompts with invisible characters (color specifications, for example). Fixing that will make the redisplay function significantly more complex. See, for example, how complicated the default rl_redisplay function is in the readline library. Unless we actually write a terminal emulator (like the current terminal widgets) then it is not possible to use readline's rl_redisplay function directly.
  7. We'll need to implement a pager ourselves, since "less" won't work in this simplified terminal widget.
  8. The system function may need to be modified so that external programs that expect to be running in a terminal will continue to work properly. On Unixy systems, this job can be done with ptys. I guess Windows systems can use a hidden console? But if these things are required, are we more or less back to were we were before since we used a pty and hidden console to implement the terminal widgets? I believe the Emacs start-process function must do similar things, so we might be able to reuse that code.
  9. If readline runs in the terminal widget, who owns the command-line history? Either way, if the GUI is in control of keyboard input, it will need access to the history list and Octave will also need access for the history functions.