Show MRU files into menuItems defined by fileStackMax.
Jérôme LAFORGE
"The box said 'Requires Windows 95, NT, or better,' so I installed Linux."
Jérôme LAFORGE
Neil Hodgson
> For me, it's annoying when you change the value of fileStackMax, the menu doesn't take care about this value.
> I think that that's helpful when you change this value, that the menuItem adjusts automatically to this value.
> I propose you this patch, that should add this feature unfortunately for GTK only
That code looks too complex and difficult to maintain for what
seems to me to be a minor benefit. The amount of explicit memory
management is particularly unwieldy and may be possible to avoid using
containers from the C++ standard library.
> In the case case, you accept it. I plan to add property to defined this value by configuration (e.g. mruFileSize or another name of your choice).
> (If you think that's worth, I can propose you only one patch included property directly)
I don't really see an extra long MRU menu being useful. Maybe
others want this.
Neil
Jérôme LAFORGE
> seems to me to be a minor benefit. The amount of explicit memory
> management is particularly unwieldy and may be possible to avoid using
> containers from the C++ standard library.
why I implemented this code with explicit memory managment.
So, I changed it, and as you told me I replaced all this code with
containers from the C++ standard library (deque) and I removed all
explicit memory managment.
I hope that this new patch is acceptable for you in term of complexity
> I don't really see an extra long MRU menu being useful. Maybe
> others want this.
mru.file.list.size.
Please find the new patch :
http://pastebin.com/jiDnQ9cT
Jérôme
Neil Hodgson
> Originally, I didn't want modify the design of the code. That's reason
> why I implemented this code with explicit memory managment.
> So, I changed it, and as you told me I replaced all this code with
> containers from the C++ standard library (deque) and I removed all
> explicit memory managment.
> I hope that this new patch is acceptable for you in term of complexity
Instead of being simpler, the patch has grown from 164 to 414
lines. I won't be applying it in its current form.
Neil
Jérôme LAFORGE
> lines. I won't be applying it in its current form.
managment of the size of mru file.
I remove it into this patch. This patch manages only the mru dynamic
menu without explicit memory.
http://pastebin.com/gX6qCZv7
The patch is now only 237 lines (with majority is due to use of
iterator)
Jérôme
Neil Hodgson
> The patch is now only 237 lines (with majority is due to use of
> iterator)
My suggestion about using standard containers was an attempt to
simplify the original code. If it doesn't help simplify then a
different approach should be used. This feature should not require a
lot of code. If it does then I won't be merging it.
Neil
Jérôme LAFORGE
> different approach should be used.
code.
Jérôme
Jérôme LAFORGE
Jérôme LAFORGE
this patch where the user can define the number of mru files and
whether the user wants use submenu for hiding the very long mru files
list. For now, it's just a mock up.
pastebin.com/CvBmY4DV
Jérôme
Jérôme LAFORGE
Neil Hodgson
I see many warnings like:
SciTEGTK.cxx: In member function ‘void SciTEGTK::CreateMenu()’:
SciTEGTK.cxx:3326:36: warning: passing NULL to non-pointer argument 1
of ‘SString::SString(int)’
Neil
Frank Wunderlich
> In the case where my previous patch is suitable for you, I propose you
> this patch where the user can define the number of mru files and
> whether the user wants use submenu for hiding the very long mru files
> list. For now, it's just a mock up.
> pastebin.com/CvBmY4DV
>
thank you for your work...is it working on gtk+ (linux) also? please
make with submenu...mru-lists direct in file-menu are to
big...especially on small displays (netbooks).
do you use dynamic creation of menus/submenus? maybe its useful to
enahnce the tools-menu like in scite-ru also for gtk+
regards Frank
Jérôme LAFORGE
Do you use some particular environment for having this kind of
warning?
Please find this new patch where replace NULL by empty string ("")
for accelerator and also replace 0 by empty string for item_type.
http://pastebin.com/MAa5dbMD
I hope this patch fix all warning, since my gcc doesn't produce the
warning,
Frank :
> do you use dynamic creation of menus/submenus? maybe its useful to
> enahnce the tools-menu like in scite-ru also for gtk+
patch into scite-ru, if useful.
Jérôme
Frank
dynamic tools-menu.
it will be great if your functions can also be used to build the tools-
menu with submenus.
On 1 Nov., 17:58, Jérôme LAFORGE <jerome.lafo...@gmail.com> wrote:
> Neil :
>
> > I see many warnings like:> SciTEGTK.cxx: In member function ‘void SciTEGTK::CreateMenu()’:> SciTEGTK.cxx:3326:36: warning: passing NULL to non-pointer argument 1> of ‘SString::SString(int)’
>
> I have no warnings. I use gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5).
> Do you use some particular environment for having this kind of
> warning?
>
> Please find this new patch where replace NULL by empty string ("")