Comments on: siunitx 2: more numbers, more tables, SVN? http://www.texdev.net/2009/01/10/siunitx-2-more-numbers-more-tables-svn/ Coding in the TeX world Tue, 07 Feb 2012 20:07:12 +0000 hourly 1 By: Joseph Wright http://www.texdev.net/2009/01/10/siunitx-2-more-numbers-more-tables-svn/#comment-25 Joseph Wright Sun, 11 Jan 2009 22:17:19 +0000 http://www.texdev.net/?p=91#comment-25 No problems. Much of what is in siunitx is things that people have suggested to me. I have one or two "regulars", who have provided a lot of the ideas implemented. I've already gone quite a long way in separating input and output, particularly at the low level for version 2. In v1 things are a bit inter-mixed (this is one of the reasons for a new version). On pgfplotstable, it gives me ideas not only for siunitx v2 but also where I thing tables in LaTeX3 need to go. At present nothing is even planned there (that I know), but I have a feeling I'll need to take a keen interest once I start on siunitx for LaTeX3 (probably in 2010). No problems. Much of what is in siunitx is things that people have suggested to me. I have one or two “regulars”, who have provided a lot of the ideas implemented. I’ve already gone quite a long way in separating input and output, particularly at the low level for version 2. In v1 things are a bit inter-mixed (this is one of the reasons for a new version).

On pgfplotstable, it gives me ideas not only for siunitx v2 but also where I thing tables in LaTeX3 need to go. At present nothing is even planned there (that I know), but I have a feeling I’ll need to take a keen interest once I start on siunitx for LaTeX3 (probably in 2010).

]]> By: Tomek http://www.texdev.net/2009/01/10/siunitx-2-more-numbers-more-tables-svn/#comment-24 Tomek Sun, 11 Jan 2009 20:12:14 +0000 http://www.texdev.net/?p=91#comment-24 I hope my comments didn't put any pressure on you - that wasn't the idea. I was merely curious about the design decisions you are going to make with siunitx because they will shape up the future of this package (at least for ver. 2). From that point of view it occurred to me that it might be a good idea to make input format of numbers independent form the output format. The reason I mentioned pgfplotstable is because of its design rather than functionality. One thing I like about it is that a number of hooks is provided at key points in the processing chain. That allows for a lot of customization with relatively little effort. To give an example, one can pre- and postprocess cell content before and after formating. Combined with the possibility to pass the content unformated this allows to use fully custom formating but still benefit from other functionality in that package. In principle it should be even possible to use siunitx for number formating in pgfplotstable but I haven't tried that. I hope my comments didn’t put any pressure on you – that wasn’t the idea. I was merely curious about the design decisions you are going to make with siunitx because they will shape up the future of this package (at least for ver. 2). From that point of view it occurred to me that it might be a good idea to make input format of numbers independent form the output format.

The reason I mentioned pgfplotstable is because of its design rather than functionality. One thing I like about it is that a number of hooks is provided at key points in the processing chain. That allows for a lot of customization with relatively little effort. To give an example, one can pre- and postprocess cell content before and after formating. Combined with the possibility to pass the content unformated this allows to use fully custom formating but still benefit from other functionality in that package. In principle it should be even possible to use siunitx for number formating in pgfplotstable but I haven’t tried that.

]]>