Wonderful! But i fear, most of the audience here hasn't the slightest idea, what it's all about. Let me explain!
Maybe, you have already clicked into some graphics in your browser or just used the "onmouseover" functionality. In HTML, everything is repesented in a DOM, that's a tree structure, like JSON. But what do you mean, when you move your mouse over a structure, say a table? Do you mean the cell element or do you rather mean the whole table which then gets highlighted by clicking on it? Lenses here give you kind of functional tool to select or glue elements or parts of the DOM together to make it a functional unit. Internally, it's done with a set of chained getters and setters to compose DOM elements to bigger units that then can be moved together, clicked on, activated or deactivated, turned, (CSS3) animated, ... whatever. A map function here is the simplest kind of "lens". With "onmouseover" you mean whole DOM. Whereever you need kind of selector, you can do it with lenses. A more complicated example is a 5 - axis robot. Each arm can be moved independently, so i have a chain of affine transformations, that finally move the tool at the end to its position. But here 5 arms sometimes give us more freedom, than we really need and maybe that there are many ways of combining movements of arms to reach the same position. Note: Moving forward then turning is not the same as first turning and then moving forwards! And now imagine, that one arm of the 5-axis robot breaks and i have to compensate that. Mathematically, i now have to introduce co- and contravariant matrixes to compensate the broken arm. Contravariance you may also need e.g. to compensate a robot arm on a rolling ship, e.g. to keep things horizontally. Lenses here give you the universal tool for new aggregation of functions that then treat data in a different way, with a different focus on the underlying structures, be it tree, graph, linear list, whatever.... the chain of getter/setter functions get rearranged automatically. In Lisp or Picolisp it is done by reworking the code. Learn: "Data is code, code is data", which makes Lisp unique in the whole world of programming. Summary: Lenses are a general tool for functional composition or rearrangement of getter/setter functions. In non-functional programming languages lenses make no sense. They simply don't exist there. Have fun! Abel Normand <abel.ze.norm...@gmail.com> schrieb am So., 19. Mai 2019, 23:08: > Hello everyone. > > I'm happy to present you my last project: > https://gitlab.com/Abel-ze-Normand/lens.l. It is a very nice abstraction > from Haskell functional programming community and I wanted to share with > you this PicoLisp implementation of Lens pattern and utilities to work with > them. I plan to use them not only as a proof of concept but also as > powerful tool to work with different formats of data in declarative way. I > hope you will find it useful or at least interesting. > -- > Best regards, Nail >
