parser

Documentation

Overview

Package parser finds the pages in a slice of tokens from tokenizer.

A normal page is /t+(co+)*c/, although technically the last clear is not part of the page. There are many corner cases, but those are documented only in the tests.

This package expects 'a' to be tokenized as 't', in contrast to what the tokenizer package produces by default.

Index

• type Page
  • func Parse(tokens []tokenizer.Token) (pages []Page)
• type Part
• type TokenError
  • func (err TokenError) Error() string

Types

type Page

type Page struct {
	Head, Body Part
}

Page contains the parts for a page. In the case of empty headers or bodies, the empty part is set to zero length (low == high), with the value chosen so head.low:body.high slices all lines inside the page.

func Parse

func Parse(tokens []tokenizer.Token) (pages []Page)

Parse finds the pages in tokens.

type Part

type Part struct {
	Low, High int
}

Part represents a header or body. It contains the values that would slice the part's lines out of a slice of all lines in the file.

type TokenError

type TokenError struct {
	Token tokenizer.Token
	Index int
}

TokenError represents an invalid token given for a certain line. It is used by Parse when panicking.

func (TokenError) Error

func (err TokenError) Error() string

Error satisfies the error interface.