Thanks for your comment, Tim. You provided a good example and comments on my blog.
On Friday, May 18, 2012 4:35:26 PM UTC-4, Tim Visher wrote: > > On Fri, May 18, 2012 at 11:03 AM, octopusgrabbus > <octopusgrab...@gmail.com> wrote: > > In our production development environment, we perform a lot of data > > transfers between diverse systems, and most of those transfers involve > > comma-delimited (.csv) data. So my first small Clojure applications have > > revolved around the clojure-csv library. > > > > While learning Clojure I have seen the comment that using nth "stinks", > > because it creates dependencies. Wanting to do things in a Clojure way, > I > > have a question. > > > > If I need to extract a number of columns of a spreadsheet to minimize > the > > dataset and this happens as the application is reading in and > initializing > > its data, what should I use to extract those columns other than nth? > And, if > > I use a series first and rest, isn't that also positional? > > > > I've thought of ways to re-position the data initially, so comparison > > columns between two different spreadsheets that have one unique key > column > > in common would be accessible with first, for example. However, I would > > still need to get at that data by column in order to reposition it, > hence > > the need for nth. > > I think the point of the parts of the discussion that I've been privy > to so far is mostly that `nth` is something that a novice clojurist > might reach for because they don't have a grasp of the sequence > abstraction as well as stream operations and laziness, in the same way > that many people immediately reach for loop/recur when some form of > list comprehension would've solved the problem better. > > Once you have your mind around these other concepts (or perhaps even > before, this is a practical language after all) there's absolutely > nothing wrong with `nth`. It just probably shouldn't be everywhere in > your code unless your domain (like, say, a domain heavy in csv data, > perhaps?) supports the need for it. > > It does tie you to ordering so it may be worth the effort to pour your > non-tagged csv data into maps so that your implementation code can be > happily ignorant if the order of the fields ever changes. You may know > things about your domain that would be good counterarguments to that. > > Anyway, good to know you're still soldiering on. ;) > > -- > > Timmy V. > > http://twonegatives.com/ > http://five.sentenc.es/ -- Spend less time on mail. > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
