Developer forums (C::B DEVELOPMENT STRICTLY!) > Development
Usability issues
Ceniza:
--- Quote from: mariocup on August 28, 2009, 10:41:22 pm ---Hi Ceniza,
--- Quote from: Ceniza on August 28, 2009, 07:11:13 pm ---mixed C++ & ASM debugging (it's actually ASM debugging showing you also the C++ statements), multiple memory views, ...
--- End quote ---
One of the most important features for embedded is in my eyes the mixed mode view: intermix ASM and C/C++ source code.
The GNU debugger provides already the information for mixed mode
--- Code: ---interpreter-exec mi "-data-disassemble -f main.cpp -l 2 -- 1"
--- End code ---
The passing of --0 or --1 will provide information for mixed mode if it is parsed.
The only think I do not not if scintilla would be able to display such a content or it would require modifications in the core.
--- End quote ---
I just checked the interpreter-exec mi command. I had no idea GDB had such a thing. It's just the way to go when writing a plugin that interfaces with it. It makes parsing the output a lot easier, and even more informative. Why do I think the debugger plugin does not use it?
I have the same doubt as you: would Scintilla be of any help here? What if we want syntax highlighting (there are two languages being shown)?
Anyway, it's good to know that GDB does already provide the mechanism to do it.
oBFusCATed:
Maybe the creator of the debugger plugin didn't find the GDB/MI interface or it wasn't stable at the time.
With the MI interface the parsing is easier, but the greatest benefit is that every command can have a unique id (the output of the command had it, too),
so you can send for execution a series of commands and when they finish to process the output.
Navigation
[0] Message Index
[*] Previous page
Go to full version