| 1 |
RDF::LinkedData - A Linked Data server implementation |
| 2 |
|
| 3 |
DESCRIPTION |
| 4 |
|
| 5 |
This module is used to create a Linked Data server that can |
| 6 |
serve RDF data out of an RDF::Trine::Model. It will look up URIs in |
| 7 |
the model and do the right thing (known as the 303 dance) and mint |
| 8 |
URLs for that, as well as content negotiation. Thus, you can |
| 9 |
concentrate on URIs for your things, you need not be concerned about |
| 10 |
minting URLs for the pages to serve it. In addition, optional modules |
| 11 |
can provide other important functionality: Cross-origin resource |
| 12 |
sharing, VoID description, cache headers, SPARQL Endpoint, Triple |
| 13 |
Pattern Fragments, etc. As such, it encompasses a fair share of |
| 14 |
Semantic Web best practices, but possibly not in a very flexible Big |
| 15 |
Data manner. |
| 16 |
|
| 17 |
INSTALLATION |
| 18 |
|
| 19 |
On Debian and derivatives, such as Ubuntu, this module can be |
| 20 |
installed with all its dependencies using |
| 21 |
|
| 22 |
apt-get install librdf-linkeddata-perl |
| 23 |
|
| 24 |
as root or using sudo. |
| 25 |
|
| 26 |
To install the most recent module, it is likely that you already have |
| 27 |
the cpan tool installed. Then just run it on the command line. If you |
| 28 |
don't have it, see http://www.cpan.org/modules/INSTALL.html |
| 29 |
|
| 30 |
Then, in the cpan tool, type |
| 31 |
|
| 32 |
install RDF::LinkedData |
| 33 |
|
| 34 |
The relevant scripts and modules will be install to different paths |
| 35 |
depending on your system. To use it, you need to find the script |
| 36 |
linked_data.psgi, e.g. using locate. |
| 37 |
|
| 38 |
CONFIGURATION |
| 39 |
|
| 40 |
*Quick setup for a demo* |
| 41 |
|
| 42 |
One-liner |
| 43 |
|
| 44 |
It is possible to make it run with a single command line, e.g.: |
| 45 |
|
| 46 |
PERLRDF_STORE="Memory;path/to/some/data.ttl" plackup -host localhost script/linked_data.psgi |
| 47 |
|
| 48 |
This will start a server with the default config on localhost on port |
| 49 |
5000, so the URIs you're going serve from the file data.ttl will have |
| 50 |
to have a base URI http://localhost:5000/. |
| 51 |
|
| 52 |
Using perlrdf command line tool |
| 53 |
|
| 54 |
A slightly longer example requires App::perlrdf, but sets up a |
| 55 |
persistent SQLite-based triple store, parses a file and gets the |
| 56 |
server with the default config running: |
| 57 |
|
| 58 |
export PERLRDF_STORE="DBI;mymodel;DBI:SQLite:database=rdf.db" |
| 59 |
perlrdf make_store |
| 60 |
perlrdf store_load path/to/some/data.ttl |
| 61 |
plackup -host localhost script/linked_data.psgi |
| 62 |
|
| 63 |
*Configuration* |
| 64 |
|
| 65 |
To configure the system for production use, create a configuration |
| 66 |
file rdf_linkeddata.json that looks something like: |
| 67 |
|
| 68 |
{ |
| 69 |
"base_uri" : "http://localhost:5000/", |
| 70 |
"store" : { |
| 71 |
"storetype" : "Memory", |
| 72 |
"sources" : [ { |
| 73 |
"file" : "/path/to/your/data.ttl", |
| 74 |
"syntax" : "turtle" |
| 75 |
} ] |
| 76 |
|
| 77 |
}, |
| 78 |
"endpoint": { |
| 79 |
"html": { |
| 80 |
"resource_links": true |
| 81 |
} |
| 82 |
}, |
| 83 |
"cors": { |
| 84 |
"origins": "*" |
| 85 |
}, |
| 86 |
"void": { |
| 87 |
"pagetitle": "VoID Description for my dataset" |
| 88 |
}, |
| 89 |
"expires" : "A86400" , |
| 90 |
"fragments" : { |
| 91 |
"fragments_path" : "/fragments" , |
| 92 |
"allow_dump_dataset" : 0 |
| 93 |
} |
| 94 |
} |
| 95 |
|
| 96 |
In your shell set |
| 97 |
|
| 98 |
export RDF_LINKEDDATA_CONFIG=/to/where/you/put/rdf_linkeddata.json |
| 99 |
|
| 100 |
If the linked_data.psgi script was installed in /usr/local/bin, go: |
| 101 |
|
| 102 |
plackup /usr/local/bin/linked_data.psgi --host localhost --port 5000 |
| 103 |
|
| 104 |
The endpoint-part of the config sets up a SPARQL Endpoint. This requires |
| 105 |
the RDF::Endpoint module, which is recommended by this module. To |
| 106 |
use it, it needs to have some config, but will use defaults. |
| 107 |
|
| 108 |
It is also possible to set an expires time. This needs |
| 109 |
Plack::Middleware::Expires and uses Apache mod_expires syntax, in the |
| 110 |
example above, it will set an expires header for all resources to |
| 111 |
expire after 1 day of access. It is strongly recommended that this is |
| 112 |
used, as it can potentially speed up access to resources that aren't |
| 113 |
accessed frequently considerably, and take load off your server. |
| 114 |
|
| 115 |
The cors-part of the config enables Cross-Origin Resource |
| 116 |
Sharing, which is a W3C Recommendation for relaxing security |
| 117 |
constraints to allow data to be shared across domains. In most cases, |
| 118 |
this is what you want when you are serving open data, but in some |
| 119 |
cases, notably intranets, this should be turned off by removing this |
| 120 |
part. |
| 121 |
|
| 122 |
The void-part generates some statistics and a description of the |
| 123 |
dataset, using RDF::Generator::Void. It is strongly recommended to |
| 124 |
install and run that, but it can take some time to generate, so you |
| 125 |
may have to set the detail level. |
| 126 |
|
| 127 |
Finally, fragments add support for Triple Pattern Fragments, a |
| 128 |
work-in-progress, initiated by http://linkeddatafragments.org/ |
| 129 |
It is a more lightweight but less powerful way to query RDF data than |
| 130 |
SPARQL. If you have this, it is recommended to have CORS enabled and |
| 131 |
required to have at least a minimal VoID setup. |
| 132 |
|
| 133 |
It is also worth noting that an environment variable LOG_ADAPTER can |
| 134 |
be set to send log statements to the console. I recommend installing |
| 135 |
Log::Any::Adapter::Screen, which can be used by setting this variable |
| 136 |
like LOG_ADAPTER=Screen. Its documentation details more environment |
| 137 |
variables that can be used to control log levels, etc. |
| 138 |
|
| 139 |
This module now also contains some facilities to support read-write |
| 140 |
hypermedia RDF, as well as the possibility to subclass RDF::LinkedData |
| 141 |
itself. The read-write functionality is provided by a separate module, |
| 142 |
RDF::LinkedData::RWHypermedia, but is highly experimental at this |
| 143 |
point. |
| 144 |
|
| 145 |
*Production server setup* |
| 146 |
|
| 147 |
In addition to the configuration above, a production system should set |
| 148 |
up a real Web server to run the Plack script. There are many ways to |
| 149 |
do this (as Plack provides an elegant separation of concerns between |
| 150 |
developers and system administrators). |
| 151 |
|
| 152 |
To set this up under Apache, put this in the host configuration: |
| 153 |
|
| 154 |
<Location /> |
| 155 |
SetHandler perl-script |
| 156 |
PerlResponseHandler Plack::Handler::Apache2 |
| 157 |
SetEnv RDF_LINKEDDATA_CONFIG /to/where/you/put/rdf_linkeddata.json |
| 158 |
PerlSetVar psgi_app /usr/local/bin/linked_data.psgi |
| 159 |
</Location> |
| 160 |
|
| 161 |
<Perl> |
| 162 |
use Plack::Handler::Apache2; |
| 163 |
$ENV{RDF_LINKEDDATA_CONFIG}='/to/where/you/put/rdf_linkeddata.json'; |
| 164 |
Plack::Handler::Apache2->preload("/usr/local/bin/linked_data.psgi"); |
| 165 |
</Perl> |
| 166 |
|
| 167 |
<Location ~ "^/(dumps|js/|css/|favicon.ico)"> |
| 168 |
SetHandler default-handler |
| 169 |
</Location> |
| 170 |
|
| 171 |
Note that in some environments, for example if the Plack server |
| 172 |
is dynamically configured and/or behind a proxy server, the server |
| 173 |
may fail to bind to the address you give it as hostname. In this case, |
| 174 |
it is wise to allow the server to bind to any public IP address, |
| 175 |
i.e. set the host name to 0.0.0.0. |
| 176 |
|
| 177 |
|
| 178 |
|
| 179 |
AUTHOR |
| 180 |
Kjetil Kjernsmo, "<[email protected]>" |
| 181 |
|
| 182 |
BUGS |
| 183 |
Please report any bugs using github |
| 184 |
<https://github.com/kjetilk/RDF-LinkedData/issues> |
| 185 |
|
| 186 |
SUPPORT |
| 187 |
You can find documentation for this module with the perldoc command. |
| 188 |
|
| 189 |
perldoc RDF::LinkedData |
| 190 |
|
| 191 |
The perlrdf IRC channel is the right place to seek help and discuss this module: |
| 192 |
|
| 193 |
irc://irc.perl.org/#perlrdf |
| 194 |
|
| 195 |
ACKNOWLEDGEMENTS |
| 196 |
This module was started by Gregory Todd Williams "<[email protected]>" |
| 197 |
for RDF::LinkedData::Apache, but has been almost totally rewritten. |
| 198 |
|
| 199 |
COPYRIGHT & LICENSE |
| 200 |
Copyright 2010 Gregory Todd Williams |
| 201 |
|
| 202 |
Copyright 2010 ABC Startsiden AS |
| 203 |
|
| 204 |
Copyright 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018 Kjetil Kjernsmo |
| 205 |
|
| 206 |
This program is free software; you can redistribute it and/or modify it |
| 207 |
under the same terms as Perl itself. |
| 208 |
|