Omer Zak wrote on Fri, Aug 21, 2015 at 07:51:51 +0300: > Hello Daniel, > Three years ago, I developed an Android application which does the same > thing [1]. > > I hoped to get more volunteers to polish and improve it, however there > was very little interest in it.
Haven't seen it, but in my case I needed a desktop app, not a smartphone one. > Nevertheless, it was a good way to tell the world that yes, I am an > Android developer. I got other projects thanks to it (such as [2]). > > So I suggest that you release your application but without expecting > anything from it beyond being a showcase of your software development > skills. That's good advice. However, I don't think the code showcases anything: it's a chunk of glue code that any first-year student could write. It saves time but is not enlightening to read. The only value I see in it is that it might save someone else the need to redo the work I have done. > While you are it, I strongly urge you to split off the part of the code, > which deals with Hebrew terminals, and make it a library that people can > take and use in their own projects. Maybe you'll be able to contribute > to terminfo/ncurses projects. All I have is a chunk of code that decides whether or not to insert fribidi into the output filter chain based on whether $(ps -o comm= -p $(xprop -id $WINDOWID _NET_WM_PID | sed 's/.*= *//g')) contains "mlterm" or "konsole". While that chunk of code does solve a problem that is in scope for ncurses, I believe the approach it takes is unsuitable for inclusion therein, for several reasons: (a) it would require ncurses to link libx11 (to get window properties); (b) it hardcodes terminal names that behave differently — that is a duct-tape solution, not a long-term one; (c) it is sometimes wrong: konsole can be configured to behave either way, but the code hardcodes a specific behaviour for it. In short, what I have is a duct-tape solution, not a long-term one. That was fine for the context I wrote it in (I needed to get my app to work in today's terminal emulators without changing them), but it's not suitable for upstreaming. The proper long-term fix would be either to make terminal emulators all behave the same way with respect to directionality, or to invent a terminfo capability that terminals would set in their terminfo descriptors and ncurses could query. Cheers, Daniel > --- Omer > > > > [1] https://github.com/Hamakor/teuria > [2] > https://play.google.com/store/apps/details?id=com.heliconbooks.epub.epubreader > > > On Fri, 2015-08-21 at 01:24 +0000, Daniel Shahaf wrote: > > I've got here a desktop app that quizzes the user with questions from > <מבחן התיאוריה (the one people take when they learn driving). It's > > basically a self-test/study app, using the questionset from the > > Ministry's web site. (They publish questionsets in six languages, > > I only tried the Hebrew set.) > > > > By "app" I mean a python script that prints questions to stdout, reads > > answers from stdin, and displays images by invoking display(1) [a > > minimal image viewer from imagemagick]. It doesn't have a GUI (beyond > > the image displayer) since my target audience didn't need one. > > > > I can't release it as-is because $LEGAL_REASONS, but I could clean it up > > to make it releaseable. Before I spend too much time on that, is that > > something anybody would be interested in? > > > > [feel free to reply offlist] > > > > Cheers, > > > > Daniel > > > > P.S. It wasn't fun to get Hebrew to print correctly on all terminals: > > there are differently-behaving terminals that use the same value of > > $TERM (undermining terminfo-based solutions). I ended up using > > $WINDOWID to get the terminal emulator's argv[0], and hardcoding > > exceptions based on that. That's so 1990... > -- > QA People of Curse. > My own blog is at http://www.zak.co.il/tddpirate/ > > My opinions, as expressed in this E-mail message, are mine alone. > They do not represent the official policy of any organization with which > I may be affiliated in any way. > WARNING TO SPAMMERS: at http://www.zak.co.il/spamwarning.html > > > _______________________________________________ > Linux-il mailing list > Linux-il@cs.huji.ac.il > http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il _______________________________________________ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
