if i ever ever get round to properly implementing telnet properly in talon.
07 Apr, 2009, David Haley wrote in the 3rd comment:
Votes: 0
It would be pretty easy to hack in at the line-processing step. Just replace ".^H" with ""; any ^H at the beginning of the line is simply discarded. Since you process input on a line-by-line basis, you don't have to care about packet fragmentation because you know you already have the whole packet. You can't backspace a newline you've already sent.
It would be pretty easy to hack in at the line-processing step. Just replace ".^H" with ""; any ^H at the beginning of the line is simply discarded. Since you process input on a line-by-line basis, you don't have to care about packet fragmentation because you know you already have the whole packet. You can't backspace a newline you've already sent.
yep, i know
07 Apr, 2009, David Haley wrote in the 5th comment:
Votes: 0
So then it's a lot easier than implementing telnet in Talon :smile:
It has been in a usable state since the first moment I pushed it to a public repository, though I believe it's been incorrectly stated as otherwise by others. There has been an odd bug or two during that time (mostly only in code for specific features), but no show stoppers. The current code is released as version 0.9, and I felt confident enough that it's stable to announce it on freshmeat.net.
The only thing I've done since that first public git push has been adding more features to it beyond the core TELNET handling. At this point it's feature complete and can be used for "real" TELNET servers, clients, and proxies. The only thing stopping me from releasing it as 1.0 is that I'm integrating it into some of my own projects and just double checking that the API is as easy to use as I want it to be (and I have identified a few warts with the TTYPE/ENVIRON/MSSP API, which I tried to shoehorn into a single API to save me some work, but they're all three just slightly different enough from each other that that isn't working too well). I come from the school of thought that 1.0 for projects like this one should mean "it's totally done and I currently have absolutely no reason to make another release." Which will be soon; work slowed a bit since I've been partying with friends, given that I'm permanently moving to the other side of the continent in a few months.
If you prefer Scandum's code, knock yourself out. My feelings on it are totally unchanged, but I'm not going to keep degenerating every topic into a flamewar with him; it doesn't do anybody any good at all. You know what I think of his stuff, you know what he thinks of my stuff, make your own decisions based on your own experience which is best for your needs. If you have specific questions or concerns about my implementation vs his then I will address them fully and frankly, but otherwise I'll try not to be a dick. ;)