Web-based Distributed Authoring and Versioning pptx

• • no additional binaries needed, just pure PHP code • • not necessarily bound to a single server or API • • easy to extend • • all needed bits included XML parser, HTTP header and c

• Web-based Distributed Authoring and

Versioning

• • Goal: enable interoperability of tools for

distributed web authoring

• • Turns the web into a writable medium

• • Applies to all kinds of content - not just HTML and images

• • Based on extensions to HTTP

• • Uses XML for properties, control, status

• • RFC 2518 (Core) and RFC 3253 (DeltaV)

How does it work?

• • Layered on HTTP/1.1 using its extension

facilities

• • New methods, for example: COPY, MKCOL, PROPFIND, CHECKOUT

• • New headers

• • Some slight changes to existing methods

semantics

• • RFC 2518 ("core"): pretty straight-forward

• • RFC 3253 (versioning): complex

A sample request

Request all metadata for a directory and its contents

PROPFIND /container/ HTTP/1.1

Host: www.foo.bar

Depth: 1

Content-Type: text/xml; charset="utf-8"

Content-Length:

<?xml version="1.0" ?>

<D:propfind xmlns:D="DAV:">

<D:allprop/>

</D:propfind>

Shortened response

HTTP/1.1 207 Multi-Status

Content-Type: text/xml; charset="utf-8"

Content-Length:

<?xml version="1.0" encoding="utf-8" ?>

<D:multistatus xmlns:D="DAV:">

<D:response>

<D:href>http://www.foo.bar/container/</D:href>

<D:propstat>

<D:prop>

<D:getcontentlength>12345</getcontentlength>

</D:prop>

<D:prop xmlns:R="http://www.foo.bar/boxschema/">

<R:bigbox>

<R:BoxType>Box type A</R:BoxType>

</R:bigbox>

<R:author>

Why PHP?

Why PHP? Why not just use

Apache mod_dav?

• • no additional binaries needed, just pure PHP code

• • not necessarily bound to a single server or API

• • easy to extend

• • all needed bits included (XML parser, HTTP header and content access )

The WebDAV base class

The abstract HTTP_WebDAV_Server base class

The actual PHP framework for writing a customized

WebDAV server is implemented as an abstract base class The class already takes care of most of the protocol specific stuff like:

• • HTTP header parsing

• • XML parsing and generation

• • authentication

• • type conversions

• • status codes

• • lock checks

• • the OPTIONS reply

• • partial GET and PUT requests

• • known client 'issues'

• •

Extending the base class

Extending the base class requires that you add methods to it that map to WebDAV specific request methods These

methods receive and return all relevant information in the form of PHP arrays You do not have to deal with XML and header parsing / generation as this is already taken care of by the baseclass

Possible methods

• • GET()

• • PUT()

• • COPY()

• • MOVE()

• • DELETE()

• • PROPFIND()

• • PROPPATCH()

• • LOCK()

• • UNLOCK()

• • CHECKLOCK()

• • CHECKAUTH()

WebDAV conformance

For a minimal Server

you only need to implement GET(), PUT() and PROPFIND() Such a minimal server is not at all compliant to the sepcs but may already work with some clients

For a DAV level 1 compliant server

you also need COPY(), MOVE(), DELETE() and PROPPATCH()

A DAV level 2 complient server

has to implement locking, so you also need LOCK(), UNLOCK() and

CHECKLOCK()

A versioning server

as specified in RFC 3253 is beyond the current scope of the this PHP package

A WebDAV GET request can be handled similar to a normal HTTP GET request Just make sure that you always set

appropriate Content-Length: and Last-Modified: headers before

returning the actual content

The path of the requested resource is passed to GET() in the parameter element 'path':

<?php

function GET (& $param ) {

$fspath = $this -> base $param [ "path" ];

if ( file_exists ( $fspath )) {

header ( "Content-Type: " mime_content_type ( $fspath );

header ( "Last-Modified: " date ( "D, j M Y H:m:s " , file_mtime ( $fspath )) "GMT

" );

header ( "Content-Length: " filesize ( $fspath ));

readfile ( $fspath );

return true ;

} else {

return false ;

}

}

?>

GET()

As an alternative you may just return a readable PHP stream together with size and modification time information from the GET() method By doing so you do not have to deal with HTTP header generation yourself and get support for partial

GET requests without taking further action as an additional benefit

<?php

function GET (& $param )

{

$fspath = $this -> base $param [ "path" ];

if ( file_exists ( $fspath )) {

$param [ 'mimetype' ] = mime_content_type ( $fspath );

$param [ 'mtime' ] = filemtime ( $fspath );

$param [ 'size' ] = filesize ( $fspath );

$param [ 'stream' ] = fopen ( $fspath , "r" );

return true ;

} else {

return false ;

}

}

?>

Note that no further error checking is needed on calling fopen()

here The base class will take appropriate action if the

returned stream is not a PHP resource or not readable

The HTTP PUT request method is used to write data to a

resource on the server The PUT() method is called by the base class with the following elements of its parameter array

filled:

• • 'path': the resource the received data should be put into

• • 'content_length': the number of bytes transfered

• • 'content_type': the content type of the data transfered

• • 'stream': a PHP stream you can read the actual data from

• • 'ranges': an array of byte range specifications to be written on partial requests

As with the GET() method you may either handle a PUT request yourself or just return a writable stream and have the base class write to it

PUT()

<?php

function PUT (& $params ) {

$fspath = $this -> base $params [ "path" ];

if(!@ is_dir ( dirname ( $fspath ))) {

return "409 Conflict" ;

}

$new = ! file_exists ( $fspath );

$fp = fopen ( $fspath , "w" );

if( is_resource ( $fp )

&& is_resource ( $params [ "stream" ])) {

while(! feof ( $params [ "stream" ])) {

fwrite ( $fp , fread ( $params [ "stream" ], 4096 )); }

fclose ( $fp );

}

return $new ? "201 Created" : "204 No Content" ; }

?>

PUT() (cont.)

function PUT (& $params ) {

$fspath = $this -> base $params [ "path" ];

if(!@ is_dir ( dirname ( $fspath ))) {

return false ;

}

$options [ "new" ] = ! file_exists ( $fspath );

$fp = fopen ( $fspath , "w" );

return $fp ;

}

?>