hi, am Dienstag 28 Juni 2011 (12:06) schrieb Thomas Friedrichsmeier: > I guess, having one output file per workspace would be a good default > behavior (but of course, users should also be able to start new output > files, manually, easily).
sounds great! > Besides, each output file should get a separate subdirectory, esp. to keep > plot files organized, better. right. > > a) possibility of some form of a table of contents (like named anchors in > > the HTML), e.g. for longer analysis results, and a "top" link at the end > > of each "subresult". could be implemented as two functions like > > rk.insert.TOC() and rk.insert.toplink() and then be used in plugins. the > > "top" link would only jump back to the beginning of the plugin results it > > belongs to. this would make it easier to navigate through results. > > > > b) a link to copy only the portion of results it belongs to to the > > clipboard (full HTML) > > > > c) a "delete" link to remove the portion of results it belongs to from > > the results file > > > > as for b) and c), i can't think of a way to do this in plain HTML, and i > > have never tried anything like c) before. i think b) could somehow be > > possible with <span> and some javascript, where "copy to clipboard" might > > be the tricky part. > > > > let me know if you think any of this is worth a shot. then, when i have > > some time, i could try to do some proof-of-concepts there. > > Absolutely. Each of these features would be good to have. Some comments: > a) Plugin output is expected to start with a call to rk.header(level=1). > And it is guaranteed to ends with .rk.make.hr(). So, probably, it's enough > to make these smarter, and we don't need any additional functions > (although .rk.make.hr() should probably be renamed). maybe create an "in.TOC=TRUE/FALSE" option for rk.header()? > Regarding the TOC, perhaps this should be written to a separate HTML file, > and be displayed in a frame? hm, frames are a bit out-dated, aren't they? but i think i see where you're heading. how about reserving space for the TOC as a left margin and placing it there in a <div style="position:fixed;"></div>? the downside would be, if the TOC gets longer than page height, you couldn't reach the lower end. but that should be solveable, somehow (or an incentive to reset your output at least once in a while...). i'll toy around some time soon. > b) and c) probably cannot be done in plain HTML, indeed. Rather this should > probably be implemented in C++, as a context menu. sounds interesting ;-) i'll have a look at HTML5, too, maybe there's some nice and useful new stuff. if it could be done rather cleanly in the output document itself, it would remain more portable with its functionality. but it's just a vague idea now. however, i'm often surprised what can be done in one markup file. for instance, some time ago i've written a simple jeopardy game with HTML, CSS and only a little javascript: o http://reaktanz.de/stuff/jeopardy/ircotofun.html not pretty, just a POC, too. but it's just that one file (i've written a bash script to create that file with custom questions, so you can actually play it). although this is not directly related, i find this amazing: o http://webhole.net/2010/04/24/html-5-slider-input-tutorial/ unfortunately, KHTML doesn't support the slider yet, neither does firefox, but it works in rekonq, so webkit does. viele grüße :: m.eik -- dipl. psych. meik michalke institut f"ur experimentelle psychologie abt. f"ur diagnostik und differentielle psychologie heinrich-heine-universit"at 40225 d"usseldorf
signature.asc
Description: This is a digitally signed message part.
------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2
_______________________________________________ RKWard-devel mailing list RKWard-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/rkward-devel
