document operator precedence

[punund]

I understand that tweaking operator precedence now means breaking legacy code.

But I still wish there were a table in the docs with operator precedence and associativity. Certain things, such as that precedence of >> is different from |> and <| regarding function appllication, are not obvious at all.

f a <| b    # f(a)(b)
f a >> b    # f(a >> b)

[rhendric]

There is a partial table of operator precedence here:

LiveScript/src/grammar.ls

Lines 460 to 487 in bc1c188

	# Precedence and Associativity
	# ----------------------------
	# Following these rules is what makes
	# `a + b * c` parse as `a + (b * c)` (rather than `(a + b) * c`),
	# and `x = y = z` `x = (y = z)` (not `(x = y) = z`).
	operators =
	# Listed from lower precedence.
	<[ left POST_IF ]>
	<[ right ASSIGN HURL ]>
	<[ right YIELD ]>
	<[ right BACKPIPE ]>
	<[ left PIPE ]>
	<[ right , FOR WHILE EXTENDS INDENT SWITCH CASE TO BY LABEL ]>
	<[ right LOGIC ]>
	<[ left BITWISE ]>
	<[ right COMPARE ]>
	<[ left RELATION ]>
	<[ right CONCAT ]>
	<[ left SHIFT IMPORT CLONEPORT ]>
	<[ left +- ]>
	<[ left MATH ]>
	<[ right UNARY ]>
	<[ right POWER ]>
	<[ right COMPOSE ]>
	<[ nonassoc CREMENT ]>
	<[ nonassoc ... ]>
	<[ left BACKTICK ]>

But it doesn't explain the logic for inserting implicit parentheses, which treats some tokens like <| specially:

LiveScript/src/lexer.ls

Line 1153 in bc1c188

return true if tag in <[ POST_IF PIPE BACKPIPE ]>

I agree, it would be better if all of this were actually in one place in the documentation.