nameGUI - an RPC frontend GUI
Forum rules
Warning !
Avoid using binary softwares from untrusted users.
Prefer compiling it yourself and verify sources.
Warning !
Avoid using binary softwares from untrusted users.
Prefer compiling it yourself and verify sources.
Re: nameGUI - an RPC frontend GUI
I like it. While I agree that Tk is probably quite a strange UI framework, I'm also not really a big fan of Qt, either. I think that it is fine to use whatever seems to be best.
BTC: 1domobKsPZ5cWk2kXssD8p8ES1qffGUCm | NMC: NCdomobcmcmVdxC5yxMitojQ4tvAtv99pY
BM-GtQnWM3vcdorfqpKXsmfHQ4rVYPG5pKS
Use your Namecoin identity as OpenID: https://nameid.org/
BM-GtQnWM3vcdorfqpKXsmfHQ4rVYPG5pKS
Use your Namecoin identity as OpenID: https://nameid.org/
Re: nameGUI - an RPC frontend GUI
My opinion in Namecoin software discussions are also based on this idea, usually. In other words, keep as close as possible to Bitcoin code.biolizard89 wrote:Well, I'd really like to see it use the same GUI toolkit that a Bitcoin wallet uses
In this case, however, Tk has two significant advantages over Qt:
1) Tk is much easier to learn which means there are a lot more potential developers who can contribute.
2) No compile/build headaches.
-
- Posts: 2001
- Joined: Tue Jun 05, 2012 6:25 am
- os: linux
Re: nameGUI - an RPC frontend GUI
Well, if it ends up being part of Armory or Electrum, then the build headaches are handled by their respective build systems. If I were building a Bitcoin wallet from scratch, I wouldn't have much opinion -- but since we would strongly benefit from having Armory or Electrum support Namecoin, it seems worth the trouble to match their GUI toolkit.cassini wrote:My opinion in Namecoin software discussions are also based on this idea, usually. In other words, keep as close as possible to Bitcoin code.biolizard89 wrote:Well, I'd really like to see it use the same GUI toolkit that a Bitcoin wallet uses
In this case, however, Tk has two significant advantages over Qt:
1) Tk is much easier to learn which means there are a lot more potential developers who can contribute.
2) No compile/build headaches.
-
- Posts: 2001
- Joined: Tue Jun 05, 2012 6:25 am
- os: linux
Re: nameGUI - an RPC frontend GUI
FYI, I just Startpaged "python tkinter vs qt". The first two results were:
https://stackoverflow.com/questions/109 ... er-vs-pyqt
https://www.reddit.com/r/Python/comment ... caught_up/
In both of these, the top 2 answers (by far) were that PyQt is much preferable to Tkinter.
https://stackoverflow.com/questions/109 ... er-vs-pyqt
https://www.reddit.com/r/Python/comment ... caught_up/
In both of these, the top 2 answers (by far) were that PyQt is much preferable to Tkinter.
-
- Posts: 69
- Joined: Sun Nov 23, 2014 3:34 pm
- os: linux
Re: nameGUI - an RPC frontend GUI
I haven't tried nameGUI, but it seems like it is pretty far along in terms of development. It would be a shame to toss it out and rewrite the whole thing using PyQt.
Maybe nameGUI doesn't have to be part of Armory.
Or maybe it can be kept standalone using Tkinter and we could do something like call os.system("python namegui.py") when a button is pressed in Armory? It would introduce another dependency to running Armory, but we might be able to provide an option to the configure script to disable Tkinter. The option could maybe hide the button that brings up the nameGUI program.
Maybe nameGUI doesn't have to be part of Armory.
Or maybe it can be kept standalone using Tkinter and we could do something like call os.system("python namegui.py") when a button is pressed in Armory? It would introduce another dependency to running Armory, but we might be able to provide an option to the configure script to disable Tkinter. The option could maybe hide the button that brings up the nameGUI program.
Re: nameGUI - an RPC frontend GUI
don't believe everything you read on the internetsbiolizard89 wrote:FYI, I just Startpaged "python tkinter vs qt". The first two results were:
https://stackoverflow.com/questions/109 ... er-vs-pyqt
https://www.reddit.com/r/Python/comment ... caught_up/
In both of these, the top 2 answers (by far) were that PyQt is much preferable to Tkinter.
![Razz :P](./images/smilies/icon_razz.gif)
Certainly ttk/tkinter lacks in some areas. E.g. it does not support creating a tray icon, which Qt5 supports. It has it's strengths, though, too, for example it has zero dependencies and works out of the box on various systems. It's sad that development is going slow these days but it has been around for ages and is very stable.
What somewhat bothers me about PyQt is the semi commercial status.
The GUI overhead is relatively small so far and I am considering some restructuring to even more separate the GUI stuff. For the name management pages I also plan keeping the GUI as separate as possible, so it should be quite easy to switch the GUI framework or reuse code for other GUI frameworks.
Hopefully there will be a proper pure Python GUI some day with native widgets for all systems.
Re: nameGUI - an RPC frontend GUI
i just hope nmcontrol and namegui are compatible together with a possible future light client .
Re: nameGUI - an RPC frontend GUI
Should the GUI use it's own config directory or a subfolder in the namecoin config directory (where the wallet is).
I tend towards a separate folder but both options have their advantages...
I tend towards a separate folder but both options have their advantages...
Re: nameGUI - an RPC frontend GUI
Hmm it works fine for me. Are you possibly using an old Python version?cassini wrote: I found one serious bug, so far. Enter any combining character (e.g. grave, acute, tilde, circumflex) in a text box, and the program crashes with
Python[1989:5194914] -[__NSCFConstantString characterAtIndex:]: Range or index out of bounds
https://bugs.python.org/issue22566
Re: nameGUI - an RPC frontend GUI
Out now: v0.2 with locked wallet handling and instant name lookup from the "register" field.
I like how fast and responsive the name lookup is - which somewhat surprised as there is no optimization whatsoever.
One of the next work packets will be a Windows installer and config tabs for id/ and d/ ip config.
I like how fast and responsive the name lookup is - which somewhat surprised as there is no optimization whatsoever.
One of the next work packets will be a Windows installer and config tabs for id/ and d/ ip config.