atom-generator

Creates ATOM feeds based on the INSPIRE Technical Guidance Download Services v3.1

How to run

go run . -f=./example/inspire/xyz-example.yaml

or write to an existing directory

mkdir output

go run . -f=./example/inspire/xyz-example.yaml -o=./output

Test

go test ./... -covermode=atomic

Docker

build

docker build -t pdok/atom-generator .

run

docker run -it -v `pwd`/example/inspire:/config -v `pwd`/output:/output --name atom pdok/atom-generator ./atom -f=/config/xyz-example.yaml -o=/output

docker-compose

In the example directory is a more complex docker-compose example that will proxy some dummy files from a minio instance though a lighttpd web services as a atom feed. This example is based on the Service and Dataset Feed from the INSPIRE Technical Guidance Download Services v3.1

docker-compose up -d
docker-compose down -v

The atom feeds and files can be accessed by the following URL

file	url
Service Feed	http://localhost/download/en.xml
Dataset Feed	http://localhost/download/waternetwork.xml
Dataset Entries	http://localhost/data/abc/waternetwork_25832.gml

How to configure

The configuration is done through a YAML configuration file. The file contains a array of atom feed entries. The defined feed entries in the configuration file are basically a 1 on 1 mapping with the ATOM xml that will be generated, with some exceptions regarding namespaces, link and entry elements.

server:
 id: "http://service.feed.org/index.xml"
 title: "Service Feed XML"
 ...

dataset:
 - id: "http://dataset.feed.org/index.xml"
   title: "Dataset Feed XML"
   subtitle: "This is a Dataset Feed XML entry"
   ...

Link

Special notice needs to be take for the link elements for the ATOM configuration. The final links that are defined in the output XML are the sum of a couple of predefined link objects and a link array. The predefined link objects are:

self
describedby
search
up

When there are additional link entries needed they can be defined and parsed through the usage of the link array. The example below show the usage of the predefined links self and describedby and the link array providing the search and alternate link. The usage through the predefined link forces that these are INSPIRE compliant, where the usage of the link array is a free format for the user.

   ...
   self:
     href: "http://xyz.org/download/en.xml"
     type: "application/atom+xml"
     hreflang: en
     title: "This document"
   describedby:
     href: "http://xyz.org/metadata/iso19139_document.xml"
     rel: describedby
     type: "application/xml"
   link:
    - rel: search
      href: "http://xyz.org/search/opensearchdescription.xml"
      type: "application/opensearchdescription+xml"
      title: "Open Search Description for XYZ download service"
    - href: "http://xyz.org/download/de.xml"
      rel: alternate
      type: "application/atom+xml"
      hreflang: de
      title: "The download service information in German"
      ...

Entry

The Entries within the Dataset Feed that contain a data configuration will have their type and length values provided thought the Content-Type and Content-Length of a HEAD request to that object. This means that a ATOM Feed will only be generated when the file is reachable for the atom-generator. This can only be circumvented when the data field is not used and the fields type and length are 'manually' provided.

   entry:
    - id: "http://xyz.org/data/abc/waternetwork_25832.gml"
      title: "Water network in CRS EPSG:25832 (GML)"
      link:
       - rel: alternate
         href: "http://xyz.org/data/abc/waternetwork_25832.gml"
         type: "application/gml+xml;version=3.2"
         length: 34987
         hreflang: en
         title: "Water network dataset encoded as a GML 3.2 document in ETRS89 UTM zone 32N (http://www.opengis.net/def/crs/EPSG/0/25832)"
      ...
    - id: "http://xyz.org/data/abc/waternetwork_WGS84.gml"
      title: "Water network in CRS EPSG:4258 (GML)"
      link:
       - rel: alternate
         href: "http://xyz.org/data/abc/waternetwork_WGS84.gml"
         data: "http://backed.server.org/example/xyz.gml"
         hreflang: en
         title: "Water Network encoded as a GML 3.2 document in WGS84 geographic coordinates (http://www.opengis.net/def/crs/OGC/1.3/CRS84)"
      ...

Stylesheet

Through the yaml configuration a stylesheet can be provided, this an make the ATOM Feed more human readable. This can be done though the parameter stylesheet and can be a relative path related to the ATOM feed xml or a absolute path.

service:
   stylesheet: "./style/style.xsl"
...
dataset:
 - lang: en
   stylesheet: "http://localhost/atom/style/style.xsl"

<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet href="./style/style.xsl" type="text/xsl" media="screen"?>
<feed xmlns="http://www.w3.org/2005/Atom" xmlns:georss="http://www.georss.org/georss" xmlns:inspire_dls="http://inspire.ec.europa.eu/schemas/inspire_dls/1.0" xml:lang="nl">
...
</feed>

Validation

When an atom feed xml is generated this needs to be exposed as a web service and validated. For the validation of the atom feeds generated by this application the default validator is presumed to be the INSPIRE Reference Validator. This validator has a download service validator the Conformance Class: Download Service - Pre-defined Atom that can be used for validating the atom feeds.

PDOK/atom-generator