soml-parser/README.md
2015-09-15 18:57:21 +03:00

134 lines
5.9 KiB
Markdown

[![Build Status](https://travis-ci.org/salama/salama-reader.svg?branch=master)](https://travis-ci.org/salama/salama-reader)
[![Gem Version](https://badge.fury.io/rb/salama-reader.svg)](http://badge.fury.io/rb/salama-reader)
[![Code Climate](https://codeclimate.com/github/salama/salama-reader/badges/gpa.svg)](https://codeclimate.com/github/salama/salama-reader)
[![Test Coverage](https://codeclimate.com/github/salama/salama-reader/badges/coverage.svg)](https://codeclimate.com/github/salama/salama-reader)
## Salama Reader
The parser part of salama is now a standalone gem. It parses bosl using Parslet and no other dependencies.
Also it is very educational, as it is very readable code, and not too much of it.
## Bosl Basic Object System Language
Bosl is just forming after a major insight. I used to nag about C quite randomly before, but now i
found the two main things that make it (as a system language)unsuitable for implementing an Object
system:
- C has inherrent non object features. But one could just use structs to get around that.
One would have to (unlike c++ eg) forbid the usage of large parts of the language
- The calling convention is not object based, ie not upward compatible in an oo system.
Contrary to what i thought before, other features of c are actually needed. Programming
an oo vm without a system language is like programming an os in assembler. All right for some, but
not most.
Specifically a static language is not an obstacle, or even a good thing. One pretends the world
is closed until run-time. Then one needs to have the same compiling capabilities.
Types, or a static type system, is also quite necessary to stay sane. It is "just" a matter of
extending that for oo later. Luckily i have found a system to do that.
### Syntax
Syntax (and semantics) of bosl are just forming, but some things are clear:
- statically typed (in the beginning with just two types) meaning all variable declarations,
functions and arguments shall be typed.
- objects but without data hiding
- probably nil objects
- static blocks (a bit ala crystal)
- call syntax as already discussed, ie message based
Some things we shall leave behind from the ruby approach are a lot of sugar, like missing brackets,
random code everywhere, expressions galore . . .
### Parser
The main parser per se is in parser/salama , but it just pulls in all the parts.
All the other files are ruby modules representing aspects of the parser.
Most names are quite self explanatory, but here is a list:
- basic_type defines just that. Strings, symbols, integers, floats , also comments and space
- call_site is a function call. May be qualified, but currently must still have braches
- compound types are hash and array definitions. Hashes still need curlies
- control is if statement which still must have an else
- expression is a helper for all code allowed in a function
- function definition must have braces too
- keywords is just a list of them
- operator expression are binary operators (see also below). There's a surprising amount
- return statement are straightforward
- while still needs a do, though i think in ruby a newline is sufficient
**Transform** defines how the rules map to Ast objects.
### Ast
The ast layer now uses the ast gem. That approach is to use a single class to represent all
types of node and use a type symbol (instead of different classes)
This works well, and is much less work.
The following step of compiling use the same kind of visitor approach as before
### Parslet
Parslet is really great in that it:
- does not generate code but instead gives a clean dsl to define a grammar
- uses ruby modules so one can split the grammars up
- has support for binary operators with precedence and binding
- has a separate transform stage to generate an ast layer
Especially the last point is great. Since it is separate it does not clutter up the actual grammar.
And it can generate a layer that has no links to the actual parser anymore, thus saving/automating
a complete transformation process.
### Operators
Parslets operator support is **outstanding** and such it was a breeze to implement most operators
very simply. See the operators.rb for details. Below is a list from the web of how it should be.
Operator list from http://stackoverflow.com/questions/21060234/ruby-operator-precedence-table
N A M Operator(s) Description
- - - ----------- -----------
1 R Y ! ~ + boolean NOT, bitwise complement, unary plus
(unary plus may be redefined from Ruby 1.9 with +@)
2 R Y ** exponentiation
1 R Y - unary minus (redefine with -@)
2 L Y * / % multiplication, division, modulo (remainder)
2 L Y + - addition (or concatenation), subtraction
2 L Y << >> bitwise shift-left (or append), bitwise shift-right
2 L Y & bitwise AND
2 L Y | ^ bitwise OR, bitwise XOR (exclusive OR)
2 L Y < <= >= > ordering
2 N Y == === != =~ !~ <=> equality, pattern matching, comparison
(!= and !~ may not be redefined prior to Ruby 1.9)
2 L N && boolean AND
2 L N || boolean OR
2 N N .. ... range creation (inclusive and exclusive)
and boolean flip-flops
3 R N ? : ternary if-then-else (conditional)
2 L N rescue exception-handling modifier
2 R N = assignment
2 R N **= *= /= %= += -= assignment
2 R N <<= >>= assignment
2 R N &&= &= ||= |= ^= assignment
1 N N defined? test variable definition and type
1 R N not boolean NOT (low precedence)
2 L N and or boolean AND, boolean OR (low precedence)
2 N N if unless while until conditional and loop modifiers