On Wed, 02 Nov 2011 15:11:24 +0200, Vladimir Panteleev wrote: > I guess such things are really subjective, but personally to me it does > seem a bit silly, and out-of-place from the documentation of a language > that aims to be the successor of C++.
I'll make it a pop culture reference... kidding. > Regardless > of that, IMO the naming of "csvText" is not ideal. What does the "text" > part imply - isn't CSV inherently a text format? How about "csvReader", > which would also disambiguate it from a CSV formatter? Sounds reasonable to me. CSV could be applied to other things, not that it ever is. I started making the implementation very generic and had to cut back especially since I couldn't test it all. I had even played with the idea of being able to just return slices of a range of things https://github.com/he-the-great/JPDLibs/blob/separator/csv/csv.d but stopped its development in favor of something that work with input ranges and could get into Phobos. > Have you decided about whether you'll be adding a formatter? Yes I'll give this a try, but I'm going to concentrate on documentation of what is there.
