front is an experimental interpreter built with PHP. This interpreter will take your input, tokenize it, built an AST and execute the AST. It is a package built to experiment with the concept of parser and interpreted languages.

DO NOT USE THIS PACKAGE IN PRODUCTION!

USE ONLY FOR EDUCATIONAL PURPOSES!

I'm serious. This whole thing is experimental. It doesn't even respect basic mathematical rules.

1. git clone git@github.com:vertexvaar/front.git
2. cd front

-> Continue at the tag specific instructions (under alternative attempts)

The tag tokenizer tags a commit where the input on the CLI is split into single tokens which will then be ordered in a tree structure (AST) which then will be executed.

The interpreter will try to make sense out of every operation given to it. Even mathematical operations on strings.

1. ./front
2. Type in some calculations.

Each parameter is delimited by whitespace.

Examples:

• 1 + 1
• 53628749 * 7453
• 7 / 5
• foo + bar
• abc * fff
• 1 + ( 7 * 2 )
• 2 * ( 4 + 1 )
• 2 * 4 + 1
• 2 + 4 * 3 (ATTENTION: This should return 16 but returns 18 because it is calculated left to right, not point before line)

The tag functional marks a state of this repository with a different approach to tokenizing/parsing than on the youtube video "function parsing" by computerphile: https://www.youtube.com/watch?v=dDtZLm7HIJs The functional branch contains an approach that has hit its limits and will be discontinued.

This approach does not actually calculate any output. It will just return the parsed input, which could be the foundation for calculations and other operations.

1. ./front
2. Type in anything you like.

The Backus-Naur-Form is a special text that defines formal language. The BNF-Definition is written in the grammar.bnf.txt file. The searchtypes.txt file containes all tags the input should be scanned for.

The aim of this code is to write and understand a parser or rather an interpreter for the BNF.

The interpreter will not do any calculations but it will classify your input based on the rules and tags given in the input files grammar.bnf.txt and searchtypes.txt

Fixed strings and whitespace are not yet supported

1. Run bash ctrl.sh install once
2. Run bash ctrl.sh run to start the interpreter
3. Type in anything you want to classify

You should try to invent your own language ;) Open grammar.bnf.txt and type in your own rules. E.g. you can define that every word must begin with an "A".

You can add the -x flag to the run command to initiate a xdebug session.