Code Style
Introduction
This document gives the KCL code style conventions. Good code style can play a vital role in the development and maintenance of the project. We can learn the KCL code style by referring to the full text of the description and sample codes, and use KCL format and lint tools to help coding.
Source File Encoding
KCL file encoding should always use UTF-8.
Code Layout
Indentation
Use 4 spaces per indentation level such as in the schema statement and if statement.
schema PersonA:
name: str # non-recommended
age: int
schema PersonB:
name: str # recommended
age: int
if True:
a = 1 # recommended
elif True:
b = 2 # non-recommended
else:
c = 3 # non-recommended
The closing brace/bracket/parenthesis on multiline constructs should line up under first character of the line that starts the multiline construct, as in:
# valid and recommended
my_list = [
1, 2, 3,
4, 5, 6,
]
# invalid
my_list = [
1, 2, 3,
4, 5, 6,
]
Tabs or Spaces
- Spaces are the preferred indentation method.
- Tabs should be used solely to remain consistent with code that is already indented with tabs.
KCL disallows mixing the use of tabs and spaces for indentation and an error will be reported during the compile time.
Blank Lines
- Surround top-level schema definitions with one blank line.
- Keep at most one blank line between two statements and remove redundant blank lines.
- Remove extra blank characters at the end of the line
- Remove extra blank characters in a blank line.
- There is no blank line in the header of the file, start writing from the first line.
- Only one blank line will be left at the end of the KCL file.
# Remove blank lines in the file header
a = 1 # Remove white space at the end of the line
# Keep at most one blank line between two statements
b = 2
# Only leave one blank line at the end of the file
Inline Expressions
Write indentation of KCL if
, elif
, else
and other conditions on different lines.
if True: print("") # non-recommended
if True: # recommended
print("")
Line Break and Continuation lines
- For long expressions, use the line continuation symbol
\
and keep the left end of multiple expressions aligned. - The 4-space rule is optional for continuation lines.
anotherString = "Too long expression " + \
"Too long expression " # non-recommended
longString = "Too long expression " + \
"Too long expression " + \
"Too long expression " # recommended
When to Use Trailing Commas
- Always use trailing commas.
Maximum Line Length
- The general recommendation is 80 characters but not absolute.
Symbol Break White Space
Try to keep the spaces between different symbols, but not too many, usually one is good.
a = 1 # recommended
b = 1 + 2 # non-recommended
Whitespace in Expressions and Statements
Avoid extraneous whitespace in the following situations:
- The parentheses
()
, brackets[]
and braces{}
in the expression have no spaces inside.
a = (1 + 2) # recommended
b = ( 1 + 2 ) # non-recommended
c = [1, 2, 3] # recommended
d = [ 1, 2, 3 ] # non-recommended
e = {key = "value"} # recommended
f = { key = "value" } # non-recommended
spam(ham[1], {eggs = 2}) # recommended
spam( ham[ 1 ], { eggs = 2 } ) # non-recommended
- Between a trailing comma and a following close parenthesis.
foo = [0,] # recommended
bar = [0, ] # non-recommended
- Immediately before the open parenthesis that starts the argument list of a function call.
print(1) # recommended
print (1) # non-recommended
- Immediately before the open parenthesis that starts indexing or slicing.
dct = {key = "value"}
lst = [1, 2, 3]
a = dct['key'] # recommended
b = dct ['key'] # non-recommended
c = lst[0] # recommended
d = lst [1] # non-recommended
- More than one space around an assignment
=
(or other) operator to align it with another.
# recommended:
x = 1
y = 2
long_variable = 3
# non-recommended:
x = 1
y = 2
long_variable = 3
- Always surround these binary operators with a single space on either side: assignment (
=
), augmented assignment (+=
,-=
, etc.), comparisons (==
,<
,>
,!=
,<=
,>=
,in
,not in
,is
,is not
), booleans (and
,or
,not
).
# recommended:
i = i + 1
submitted += 1
x = x * 2 - 1
hypot2 = x * x + y * y
c = (a + b) * (a - b)
# non-recommended:
i = i+1
submitted+=1
x = x*2 - 1
hypot2 = x*x + y*y
c = (a+b) * (a-b)
- Break one blank line between different statements e.g., import, schema and expression statements.
import math
import net
schema Person:
name: str
person = Person {
name = "Alice"
}
- Compound statements (multiple statements on the same line) are generally discouraged
# recommended:
if foo == 'blah':
do_blah_thing()
do_one()
do_two()
do_three()
# non-recommended:
if foo == 'blah': do_blah_thing()
do_one(); do_two(); do_three()
Naming Conventions
Naming Styles
The following naming styles are commonly distinguished:
b
(single lowercase letter)B
(single uppercase letter)lowercase
lower_case_with_underscores
UPPERCASE
UPPER_CASE_WITH_UNDERSCORES
CapitalizedWords
(capitalize all letters of the acronym inCapitalizedWords
e.g.,HTTPServer
.)mixedCase
(differs fromCapitalizedWords
by initial lowercase character)Capitalized_Words_With_Underscores
(ugly and non-recommended)
Names to Avoid
Never use the characters 'l' (lowercase letter el), 'O' (uppercase letter oh), or 'I' (uppercase letter eye) as single-character variable names.
Package and Module Names
Package and module names should have short, all-lowercase names.
Schema Names
Schema names should normally use the CapWords
convention.
Constants
Constants are usually defined on a module level and written in all capital letters with underscores separating words such as MAX_OVERFLOW
and TOTAL
.
Import
- Imports should usually be on separate lines.
- Imports are always put at the top of the file, just after any module comments and docstrings, and before module globals and constants.
- Imports should be grouped in the following order and we should put a blank line between each group of imports.
- Standard library imports.
- Related third party plugin imports.
- Local application/library specific imports.
- Use an alias when we import a package name with a relatively long path.
- Leave only one space between the Import keyword and the package name.
import net # recommended
import math # non-recommended
import ..pkg.internal_pkg as alias_pkg # recommended
Comments
- Comments should be complete sentences. The first word should be capitalized unless it is an identifier that begins with a lower-case letter (never alter the case of identifiers!).
- Block comments generally consist of one or more paragraphs built out of complete sentences, with each sentence ending in a period.
- Use two spaces after a sentence-ending period in multi-sentence comments, except after the final sentence.
Block Comments
Block comments generally apply to some (or all) code that follows them, and are indented to the same level as that code. Each line of a block comment starts with a #
and a single space(unless it is indented text inside the comment).
Paragraphs inside a block comment are separated by a line containing a single #
.
# This is a block comment
a = 1
Inline Comments
Use inline comments sparingly.
An inline comment is a comment on the same line as a statement. Inline comments should be separated by at least two spaces from the statement. They should start with a #
and a single space.
a = 1 # This is an inline comment
Documentation Strings
Write doc strings for all public schema and schema attributes.
schema Person:
"""
Person schema doc string
"""
name: str = "Alice"
String
- Single-quoted strings and double-quoted strings are the same in KCL.
- Use double-quoted string with lowercase prefix
- For triple-quoted strings, always use double quote characters to be consistent with the docstring convention.
- When a string contains single or double quote characters, use the other one to avoid backslashes in the string.
strC = "'123'" # recommended
strD = "\"123\"" # non-recommended
Number
- Use lowercase for the prefix of non-decimal numbers, and use uppercase for the number itself.
foo = 0xAB # recommended
bar = 0Xab # non-recommended
Operators
Binary Operators
- Leave only one space before and after the assignment
=
. - Leave only one space before and after the binary operator in the expression.
a = 1 # recommended
b=2 # non-recommended
c= 3 # non-recommended
d =4 # non-recommended
_value = (1 + 2 * 3) # recommended
_value = (1+2*3) # non-recommended
Unary Operators
- There is only no space after unary operators e.g.,
~
,+
and-
.
_value = 1 + -2 * ~3 # recommended
_value = 1+ - 2 * ~ 3 # non-recommended
- There is no space after
**
and*
in the dict/list deduction expressions and argument expressions.
_list = [1, 2, 3]
_list = [*_list, [4, 5 ,6]] # recommended
_list = [* _list, [4, 5 ,6]] # non-recommended
_dict = {**{k = "v"}, **{k = "v"}} # recommended
_dict = {** {k = "v"}, ** {k = "v"}} # non-recommended
- Use
is not
operator rather thannot ... is
.
# recommended:
if foo is not None:
a = 1
# non-recommended:
if not foo is None:
a = 1
Dict
- There is no space before the colon
:
at the instantiation of KCL dict and schema config, and a space after the colon:
.
d1 = {labels: {k1 = "v1"}} # recommended
d2 = {labels : {k1 = "v1"}} # non-recommended
d3 = {labels :{k1 = "v1"}} # non-recommended
- Always surround the override attribute operator
=
and the insert attribute operator+=
with a single space on either sid.
d1 = {key = "value"} # recommended
d2 = {key= "value"} # non-recommended
d3 = {key ="value"} # non-recommended
d1 = {key += [0, 1, 2]} # recommended
d2 = {key+= [0, 1, 2]} # non-recommended
d3 = {key +=[0, 1, 2]} # non-recommended
- Remove all commas at the end of the line in the KCL multiline dict because the end commas of each line are optional.
d = {
key1 = "value1"
key2 = "value2"
key3 = "value3"
key4 = "value4"
}
List
- Keep only one space after the comma
,
separating elements in the list
a = [1, 2, 3] # recommended
b = [1,2,3] # non-recommended
- Keep only one space before and after the comprehension expression token
for
andin
in the dict and list.
a = [i for i in range(10)] # recommended
b = [i for i in range(10)] # non-recommended
Slice
- Keep the same number of spaces before and after the colon
:
of the list slice.
l = [1, 2, 3]
a = l[0:2] # recommended
b = l[0 : 2] # non-recommended
c = l[0: 2] # non-recommended
d = l[0 + 0 : 1 + 1] # recommended
e = l[0 + 0:1 + 1] # non-recommended
Schema
- Leave only one space before and after the schema attribute assignment
=
. - Always add a doc string to a schema, which is a good programming habit.
schema Person:
"""
Schema doc string
"""
name: str = "Alice" # recommended
age : int=12 # non-recommended
person = Person {}
- Keep no spaces around the schema inheritance operator
()
schema Base:
name: str
schema Person(Base): # recommended
age: int
schema Schema ( Base ): # non-recommended
age: int
- Keep only one space between the brackets and the schema name of the config at schema instantiation.
schema Base:
name: str
schema Person(Base):
age: int
personA = Person{} # non-recommended
personB = Person {} # recommended
- Keep only one space between the mixin keyword and the following
[]
operator
schema NameMixin:
name: str = "name"
schema Person:
mixin [NameMixin] # non-recommended
age: int
schema Parent:
mixin [NameMixin] # recommended
age: int
Attribute Annotations
- Annotations for schema attributes should have a single space after the colon
:
and no space before the colon:
.
# recommended:
schema Person:
name: str # No space before the colon `:`
age: int = 18 # Spaces around assignment`=`
# non-recommended:
schema Person:
codeA:int # No space after the colon `:`
codeB : int # Space before the colon `:`
name: str="Alice" # No spaces around assignment`=`
- There are no spaces around the colon
:
in the dict type annotation.
schema Person:
labels: {str:str} # recommended
keyValues: {str : str} # non-recommended
Arguments
- There are no spaces around the assignment
=
in the function/schema/decorator keyword arguments (kwargs).
schema Person[nameVar]:
# Decorator kwargs
@deprecated(strict=False) # recommended
name: str = nameVar
@deprecated(strict = False) # non-recommended
age: int
# Schema kwargs
personA = Person(nameVar="Alice") {} # recommended
personB = Person(nameVar = "Bob") {} # non-recommended
# Function kwargs
print("", end='') # recommended
print("", end = '') # non-recommended
Index Signature
- It is recommended to place the schema index signature before the schema attribute instead of mixed placement.
schema ConfigReCommended:
[...str]: str # recommended
name: str
image: str
schema ConfigNonReCommended:
name: str
[...str]: str # non-recommended
image: str
Keywords
- Only one space is usually reserved around the keyword, such as
schema
,mixin
,is
andnot
, etc.
schema NameMixin:
check:
name not None
schema Person:
"""
Person schema definition
"""
mixin [NameMixin]
name: str = "Alice"
age: int
person = Person {
age = 18
}
Function
- There are no spaces around the function/package select operator
.
- There are no spaces between the function name and the parentheses
()
.
import math
print(math.log(10)) # recommended
print( math . log (10)) # non-recommended