Introduction to Scheme

From MemCP
Revision as of 15:24, 17 May 2024 by Carli (talk | contribs)
Jump to navigation Jump to search

When you run ./memcp, you will be dropped at a scheme shell like this:

memcp Copyright (C) 2023, 2024   Carl-Philip Hänsch
    This program comes with ABSOLUTELY NO WARRANTY;
    This is free software, and you are welcome to redistribute it
    under certain conditions;

loading storage /tmp/x/system/bdf64a22-6315-463c-bbf7-329317ad50ed-id of type 10
loading storage /tmp/x/system/bdf64a22-6315-463c-bbf7-329317ad50ed-username of type 20
loading storage /tmp/x/system/bdf64a22-6315-463c-bbf7-329317ad50ed-password of type 20

Welcome to memcp

performing unit tests ...
finished unit tests
test result: 15/15
all tests succeeded.

Initializing SQL frontend
MySQL server listening on port 3307 (connect with `mysql -P 3307 -u root -p` using password 'admin')
listening on http://localhost:4321


    Type (help) to show help

>  

Scheme is a functional programming language and a subset of LISP. Every expression is either a primitive value or a list.

To understand the semantics of Scheme, take a look at the following Scheme REPL session:

> 12
= 12

> "hi"
= "hi"

> (+ 1 2)
= 3

> (+ 1 2 3)
= 6

> (+ 2 (* 2 2))
= 6

> (+ 2 (* 2 4))
= 10

> (concat "Hello" "World")
= "HelloWorld"

Syntax

Function Call

A function call has the following format:

(functionname param1 param2 param3 ...)

Function calls start with (, contain one function and zero or more parameters separated by space .

Some basic functions are + - * / print concat. For more consult the manual with (help).

Literals

The following literals are allowed:

  • Number Literals: 1 2 3 56 7.5 4.3e20 -6.123e-3
  • String Literals: "Hello World" "First Line\nSecond Line" "he said: \"what?\" and smiled"
  • Function names: print
  • Symbol literals 'print
  • List literals '(1 2 3) '("a" "b" "c")
  • Associative Array literals '("key" "value" "size" 45 "name" "Peter" "sublist" '(1 2 3)) - associative arrays are lists with key value pairs flattened down to a one dimensional list of an even number of items.
  • Lambdas: (lambda (param1 param2 ...) body)

Lambda Functions

Scheme allows for creating lambda functions which enclose their own scope:

(define decrease-by-one (lambda (number) (- number 1)))

The function can now be called from scheme:

> (decrease-by-one 5)
= 4

Lambda functions can be passed to other functions as values. They enclose your original scope, so you can use variables from the outside inside your function.

List handling

The following list functions can be used:

> '(1 2 3) "list literal"
= (1 2 3)

> (cons 0 '(1 2 3)) "prepend item to list"
= (0 1 2 3)

> (append '(1 2 3) 4 5) "append items to list"
= (1 2 3 4 5)

> (has? '(1 2 3) 2) "check if item is in list"
= true

> (filter '(1 2 3) (lambda (x) (< x 2))) "filter items that are smaller than 2"
= (1)

> (map '(1 2 3) (lambda (x) (* x 2))) "double all numbers"
= (2 4 6)

> (reduce '(1 2 3) + 0) "build sum by starting with 0 and reducing them over (+ 0 value)"
= 6

Associative Lists (Objects)

The following functions can be used:

> (set obj '("a" 1 "b" 2 "c" 3)) /* construct object from a list with alternating key-value */
= ("a" 1 "b" 2 "c" 3)

> (obj "a") /* extract a value */
= 1

> (set_assoc obj "a" "5") /* construct a new object with another value */
= ("a" "5" "b" 2 "c" 3)

> (set obj (set_assoc obj "a" "5")) /* modify obj */
= ("a" "5" "b" 2 "c" 3)


> (filter_assoc obj (lambda (key value) (not (equal? key "c")))) /* filter an object by a condition */
= ("a" 1 "b" 2)

> (map_assoc obj (lambda (key value) (* value 2))) /* double all values */
= ("a" 2 "b" 4 "c" 6)

> (reduce_assoc obj (lambda (accumulator key value) (+ accumulator value)) 0) /* sum up all values */
= 6

As you see in set_assoc, we have a pure functional approach here. Objects are not modified in-place like in languages like Java, JavaScript, Python, C++ or such, but rather we have to construct a modified copy of that value and overwrite it to the original value. The Scheme optimizer will detect if an in-place operation would be safe, too and can optimize this corner case but the language itself prioritizes safety.

Further Help and Documentation of all Functions

If you type (help) into the console, you will get the following overview:

> (help)
Available scm functions:

-- SCM Builtins --
  quote: returns a symbol or list without evaluating it
  eval: executes the given scheme program in the current environment
  optimize: optimize the given scheme program
  if: checks a condition and then conditionally evaluates code branches; there might be multiple condition+true-branch clauses
  and: returns true if all conditions evaluate to true
  or: returns true if at least one condition evaluates to true
  coalesce: returns the first value that has a non-zero value
  coalesceNil: returns the first value that has a non-nil value
  define: defines or sets a variable in the current environment
  set: defines or sets a variable in the current environment
  error: halts the whole execution thread and throws an error message
  try: tries to execute a function and returns its result. In case of a failure, the error is fed to the second function and its result value will be used
  apply: runs the function with its arguments
  apply_assoc: runs the function with its arguments but arguments is a assoc list
  symbol: returns a symbol built from that string
  list: returns a list containing the parameters as alements
  string: converts the given value into string
  match: takes a value evaluates the branch that first matches the given pattern
  lambda: returns a function (func) constructed from the given code
  begin: creates a own variable scope, evaluates all sub expressions and returns the result of the last one
  parallel: executes all parameters in parallel and returns nil if they are finished
  source: annotates the node with filename and line information for better backtraces

-- Arithmetic / Logic --
  number?: tells if the value is a number
  +: adds two or more numbers
  -: subtracts two or more numbers from the first one
  *: multiplies two or more numbers
  /: divides two or more numbers from the first one
  <=: compares two numbers
  <: compares two numbers
  >: compares two numbers
  >=: compares two numbers
  equal?: deep-compares two values of the same type
  equal??: performs a sloppy equality check on primitive values (number, string, bool. nil), strings are compared case insensitive
  !: negates the boolean value
  not: negates the boolean value
  nil?: returns true if value is nil
  min: returns the smallest value
  max: returns the highest value
  floor: rounds the number down
  ceil: rounds the number up
  round: rounds the number

-- Strings --
  string?: tells if the value is a string
  concat: concatenates stringable values and returns a string
  substr: returns a substring
  simplify: turns a stringable input value in the easiest-most value (e.g. turn strings into numbers if they are numeric
  strlen: returns the length of a string
  strlike: matches the string against a wildcard pattern (SQL compliant)
  toLower: turns a string into lower case
  toUpper: turns a string into upper case
  replace: replaces all occurances in a string with another string
  split: splits a string using a separator or space
  htmlentities: escapes the string for use in HTML
  urlencode: encodes a string according to URI coding schema
  urldecode: decodes a string according to URI coding schema

-- Lists --
  append: appends items to a list and return the extended list.
  append_unique: appends items to a list but only if they are new.
  cons: constructs a list from a head and a tail list
  car: extracts the head of a list
  cdr: extracts the tail of a list
  merge: flattens a list of lists into a list containing all the subitems. If one parameter is given, it is a list of lists that is flattened. If multiple parameters are given, they are treated as lists that will be merged into one
  merge_unique: flattens a list of lists into a list containing all the subitems. Duplicates are filtered out.
  has?: checks if a list has a certain item (equal?)
  filter: returns a list that only contains elements that pass the filter function
  map: returns a list that contains the results of a map function that is applied to the list
  mapIndex: returns a list that contains the results of a map function that is applied to the list
  reduce: returns a list that contains the result of a map function
  produce: returns a list that contains produced items - it works like for(state = startstate, condition(state), state = iterator(state)) {yield state}
  produceN: returns a list with numbers from 0..n-1
  list?: checks if a value is a list
  contains?: checks if a value is in a list; uses the equal?? operator

-- Associative Lists / Dictionaries --
  filter_assoc: returns a filtered dictionary according to a filter function
  map_assoc: returns a mapped dictionary according to a map function
  reduce_assoc: reduces a dictionary according to a reduce function
  has_assoc?: checks if a dictionary has a key present
  extract_assoc: applies a function (key value) on the dictionary and returns the results as a flat list
  set_assoc: returns a dictionary where a single value has been changed.
  merge_assoc: returns a dictionary where all keys from dict1 and all keys from dict2 are present.

-- Parsers --
  parser: creates a parser

-- Sync --
  newsession: Creates a new session which is a threadsafe key-value store represented as a function that can be either called as a getter (session key) or setter (session key value) or list all keys with (session)
  once: Creates a function wrapper that you can call multiple times but only gets executed once. The result value is cached and returned on a second call. You can add parameters to that resulting function that will be passed to the first run of the wrapped function.
  mutex: Creates a mutex. The return value is a function that takes one parameter which is a parameterless function. The mutex is guaranteed that all calls to that mutex get serialized.

-- IO --
  print: Prints values to stdout (only in IO environment)
  help: Lists all functions or print help for a specific function
  import: Imports a file .scm file into current namespace
  load: Loads a file and returns the string
  serve: Opens a HTTP server at a given port
  mysql: Imports a file .scm file into current namespace
  password: Hashes a password with sha1 (for mysql user authentication)

-- Storage --
  scan: does an unordered parallel filter-map-reduce pass on a single table and returns the reduced result
  scan_order: does an ordered parallel filter and serial map-reduce pass on a single table and returns the reduced result
  createdatabase: creates a new database
  dropdatabase: creates a new database
  createtable: creates a new database
  createcolumn: creates a new column in table
  shardcolumn: tells us how it would partition a column according to their values. Returns a list of pivot elements.
  altertable: alters a table
  droptable: removes a table
  insert: inserts a new dataset into table
  stat: return memory statistics
  show: show databases/tables/columns
  rebuild: rebuilds all main storages and returns the amount of time it took
  loadCSV: loads a CSV file into a table and returns the amount of time it took.
  loadJSON: loads a .jsonl file from disk into a database and returns the amount of time it took.
  settings: reads or writes a global settings value. This modifies your data/settings.json.

get further information by typing (help "functionname") to get more info