ltd-beget / dns-zone-tokenizer Goto Github PK

View Code? Open in Web Editor NEW

6.0 24.0 6.0 856 KB

Dnszone file lexical parser

License: MIT License

PHP 93.90% Ruby 1.78% DIGITAL Command Language 4.32%

dns-zone-tokenizer's Introduction

Php dns-zone-tokenizer

Tokenize dns zone files and that's all, folks.

Installation

composer require ltd-beget/dns-zone-tokenizer

Usage

<?php
    use LTDBeget\dns\Tokenizer;
    
    require(__DIR__ . '/vendor/autoload.php');
    
    $config_path = realpath(__DIR__."/zone/exampleZone"); // path to your dns zone file
    $plain_config = file_get_contents($config_path);
    
    $tokenized = Tokenizer::tokenize($plain_config); // that's all, folks. All is done =)

Developers

Regenerate documentation

$ ./vendor/bin/phpdox

Run tests

$ wget https://phar.phpunit.de/phpunit.phar

$ php phpunit.phar --coverage-html coverage

$ php phpunit.phar --coverage-clover coverage.xml

License

released under the MIT License. See the bundled LICENSE file for details.

dns-zone-tokenizer's People

Contributors

Stargazers

Watchers

Forkers

radiowandafm westie samuelwilliams mtorromeo fizzyduck tito10047

dns-zone-tokenizer's Issues

Information about tokeniser (use of root zone character, etc)

Hi!

I'd like to ask a quick question about your tokeniser.

Imagine that I have something like this (simplified) example:

$ORIGIN typefish.co.uk.

@                                   3600 IN A     127.0.0.1
a                                   3600 IN A     127.0.0.2
a.typefish.co.uk                    3600 IN A     127.0.0.3
a.typefish.co.uk.                   3600 IN A     127.0.0.4

We end up with output from named-compilezone similar to:

typefish.co.uk.                     3600 IN A     127.0.0.1
a.typefish.co.uk.                   3600 IN A     127.0.0.2
a.typefish.co.uk.a.typefish.co.uk.  3600 IN A     127.0.0.3
a.typefish.co.uk.                   3600 IN A     127.0.0.4

However, the output from your tokeniser is:

@                                   3600 IN A     127.0.0.1
a.typefish.co.uk.                   3600 IN A     127.0.0.2
a.typefish.co.uk.a.typefish.co.uk.  3600 IN A     127.0.0.3
a.typefish.co.uk.                   3600 IN A     127.0.0.4

Now, the code - and its tests, imply that the root zone is tokenised as @ as opposed to typefish.co.uk. - should this be the case?

(I personally think it should match named-compilezone)

I've changed the code on my machine to match the output of named-compilezone but I don't want to change any tests nor proceed further with any merging of code until I know why this has been done in this manner.

Regards,
David Weston

Add all resource record types (AXFR, RRSIG, DNSKEY, etc.)

You can find the whole list here: http://www.iana.org/assignments/dns-parameters/dns-parameters.xhtml#dns-parameters-4

Падает в бесконечный цикл

;
; BIND reverse data file for BEGET-NEW-NET1 5.101.159.0/24 zone
;
$TTL 3600
@   IN  SOA ns1.beget.ru.   hostmaster.beget.ru. (
    2014070701  ; Serial
    3600        ; Refresh
    600         ; Retry
    86400       ; Expire
    3600        ; Negative Cache TTL
)

;; NS
@   IN  NS      ns1.beget.ru.
@   IN  NS      ns2.beget.ru.

;; SYSTEM
1   IN  PTR     rt01.beget.ru.
3   IN  PTR     ns3.beget.ru.
5   IN  PTR     ns5.beget.ru.
11  IN  PTR     ns1.beget.ru.
12  IN  PTR     ns2.beget.com.ua.

не зацикливается без комментариев в начале

Parsing a zone file without a trailing newline results in an infinite loop

dns-zone-tokenizer/src/LTDBeget/dns/record/RData.php

Line 205 in 69d6498

$this->tokens[$tokenName] = trim($this->tokens[$tokenName]);

In RData.php, line 205, we have:

                    if($this->stream->isEnd()) {
                        $this->tokens[$tokenName] = trim($this->tokens[$tokenName]);
                    }
                    goto start;

The problem is that isEnd() returns true, but the code still goes back to start:.

On line 155, the code should check for isEnd() as well as AsciiChar::NULL to prevent the infinite loop.

Implement phpunit tests

Subj. For dns parser we need a lot of unit tests

Parser nodename

WTF
@ 600 IN MX 10 mx1.beget.ru.

This will be parsed as WTF\n@ in nodename

Tabs in whitespace of multiline txt record

If there are tabs at the beginning of the second (or subsequent) lines in a multi line TXT record, the tokenizer only returns the first line of text. It correctly ignores space and newline, but currently doesn't ignore tab.

While this may not be too widespread, the functions that DirectAdmin (server management software) uses to create TXT records inserts tabs at the beginning of the second (and subsequent) lines of a multi line TXT record, and as DirectAdmin is widely used it would be nice to be able to support this.

I've created a fix and a pull request for this: #23

I appreciate your fantastic script - thanks!