Good day developers! First, forgive me if anyone has already mentioned this, I just joined the list a few days ago. Perhaps some of you are familiar with a little open source OS by the name of Syllable ( http://www.syllable.org ). It's a fork of the AtheOS project, and its doing pretty well for itself. It's aimed at home users, and its goals are to be fast, and easy to use, similar (but not replicating) BeOS. One thing Syllable could use is a web browser. The native browser is based off an old 2.x khtml port and is sort of a hack. To get a full fledged browser that is easy to use, someone would need to port Gecko, or a newer khtml. Or Dillo. :o) I read that you're planning on porting to FLTK and just read the "about porting doc". Am I to understand that Dillo (or dw, rather) is already a library, separate from the GUI? How easy (forgive the question, I'm still a young grasshopper programmer) would this be to port to a Threaded C++ native widget set? Something that doesn't have GTK+ or FLTK. Wow, this email was way longer than I intended. Forgive me for taking so much of your time. Any feedback would be greatly appreciated. Regards, Bill
Hi Bill, On Tue, Sep 21, Bill Sanders wrote:
I read that you're planning on porting to FLTK and just read the "about porting doc". Am I to understand that Dillo (or dw, rather) is already a library, separate from the GUI?
Some points to clarify: - Dw is only one module of dillo, which does the low level rendering; it is not even a rendering engine. It is the most critical case, when we port dillo to another toolkit (from Gtk+ to FLTK), so I redesigned Dw, that porting is simpler, and development can be better coordinated (This redesign has also some other benefits, look into the documentation.) - I plan to extract Dw (at least the core), and make a library of it. This is largely finished, but has a low priority, so do not expect something within the next time. The purpose is simply code reuse, other applications may find it useful to use Dw for rendering whatever. - There are currently no plans, how to deal with the other modules of dillo, which depend on Gtk+. However, they are not very critical, and any GUI toolkit should solve the problems occuring here. - We are only planning to port dillo to another toolkit, not to another OS.
How easy (forgive the question, I'm still a young grasshopper programmer) would this be to port to a Threaded C++ native widget set? Something that doesn't have GTK+ or FLTK.
The UI in dillo is single-threaded, only the networking uses threads, so this does not matter at all. For reasons stated above, no one should not hope, that dillo will soon be portable, and there may exist dozens of versions, based on dozens of toolkits. This aspect of porting is still unclear. As for Dw, you should look into the archives for my last messages, and for the prototype mentioned. Finally, porting dillo to another OS, other than Unix-like, is another topic. I don't know anything about Syllable, but there will certainly be problems, when porting the networking code. Sebastian
participants (2)
-
Bill Sanders -
Sebastian Geerken