1Introduction

We often see a process that applies some operation or transformation on multiple data stored in lists and then put the results into another list. Among them are includes plotting results of a mathematical function fed with sequence of numbers as its parameter and tansforming multiple records extracted from some database into a specific format.

For such a process, many programming language provides sequence control syntax for repeating, with which you can pick up elements from a list subsequently and then create another list that contains result values. Or, if you're a programmer of a functional language, it might be a familiar approach that you prepare a higher-order function with which you apply a certain function on elements in a list.

Either way, you've had to explicitly program "repeat" operation with existing languages. However, when you provide n number of values to a function taking one argument and returning one result, it's obvious that you want n number of answers from it. If a programming language itself has a feature to repeat a function automatically when it's given with a list or an iterator as its arguments, there's no need to explicitly describe repeating syntax any more.

I calls this feature Implicit Mapping since it implicitly does mapping process.

This may look similar with a feature called "vectorization" that has already been adopted by languages and libraries such as MATLAB and NumPy. A different point is that Implicit Mapping is not limited to mathematic operations with number values, but it can work with various types of value like string, image and even user-defined one. And I've found out Implicit Mapping would be much efficient when it cooperates with more sophisticated iterator operations such as Member Mapping that can access members of objects coming from an iterator or a list, and repeat functions capable of generating iterators. These ideas have motivated me to create a brand-new language.

Before the creation of a new language, I made guidelines described below:

• Inherit Familiar Syntax

  I don't think it's a good idea to bother creating an original syntax if it has same effects as that of existing languages. I decided to follow other popular languages as much as possible when I need to make syntax and name variables and functions. In fact, as the new language uses a pair of curly brackets to embrace a block, an overwhole appearance of the code may look like one in C or Java.

• Be Practical

  Any programming languages are expected to solve problems that actually exist around us. For such purposes, capabilities of reading/writing files and processing text data are still important. However, these days, having such functions is far from enough because various technologies like Internet, graphic image files, database and GUI become so common that most users of computer expect any programs to be capable of handling them. To be practical, the new language should be shipped with these capabilities as standard.

Following these guidelines, I've developed a script language Gura that comes with functions and methods that are aware of Implicit Mapping policy, and published its first version on March 15, 2011.

I found it amazing to develop a new programming language since creating a language doesn't instantly mean that the creater is an expert programmer of it. This may be similar to a situation that you try to come up with an idea of a new game: even if you make its rule, you have to actually play it to know tricks and tactics so that you get a victory on the rule. I also had to create and try a lot of scripts for myself to know how to make programs of Gura. Throughout the process, I've learned that Gura's various features including Implicit Mapping are really practical in actual programming fields.

As one user, I can recommend this script language for you.

Yutaka Saito
March 6th, 2014