BioCollect documents being downloaded from MERIT

[temi]

BioCollect documents sometime have fieldcapture.ala.org.au as hostname. This is due to the way URL is constructed Document.groovy in Ecodata. Sometime ago hostname was added based on the server sending the request. However, when reindexing is done, it is triggered by merit or BioCollect. Hostname that triggered the reindex is then added to all document URLs.

The current behaviour was introduced by this ticket

Solution

1. Keep old behaviour of adding relative path.
2. Add hostname before pdfgen requests are issued in BioCollect and MERIT.
3. Add a new method to get the fully resolved path. This will required for during download and darwin core archive generation.

Make sure document ecodata queries adds the hostname after doing the search in GSON.