* Add gzip to API.py
* pytest for gzip
* fix pytest
* Update test_api.py
* Add gzip to server block of configuration files
* Update api.py to include gzip opts from config
Update logic behind gzip compression such that compression will only happen when gzip in the Accept-Encoding request header and config.server.gzip is True. Use server charset for encoding/decoding and include charset in Content-Type when compressed.
* Update pytest to use gzip config options
Update pytest to test with (gzip in Accept-Encoding & server.gzip = False), (gzip in Accept-Encoding & server.gzip = True), (gzip not in Accept-Encoding & server.gzip = True)
* Better Content Negotiation (#7)
* Change content negotiation in _get_format()
* Add to pytest and force h to always be string
* Force F_JSON to default response
* Update test_api.py for utf-16 encoding
* better content negotiation pytest.
* Add comments to pytest
* Fix flake 8
* Improve docstring for gzip decorator function
* - Added missing parameters on the openapi definition of get an observations tile request
* - changed default background map to osm
* Revert "- changed default background map to osm"
This reverts commit 554c065c6699ad6c0585f5c09ce34e746d5b3dc2.
* - Added example with docker containers, for MongoDB data provider
* - attempts
* - switched to use ne_110m_populated_places_simple.geojson as data source
* - updated README
* - updated README
Co-authored-by: doublebyte1 <info@doublebyte.net>
* - Added missing parameters on the openapi definition of get an observations tile request
* - changed default background map to osm
* Revert "- changed default background map to osm"
This reverts commit 554c065c6699ad6c0585f5c09ce34e746d5b3dc2.
* - added discovery.type to elasticsearch configuration, to prevent failed bootstrap checks
Co-authored-by: doublebyte1 <info@doublebyte.net>
* return correct limits & faster hits query
* Fix hits behavior (and break pytest)
* fix pytest for sensorthings provider
* check before unpacking properties
* add q string
* enclose id in single quotes for entities with non-literal id values
* Undo accidental branch merging
* Undo other part of accident branch merging
* add uri_field to config schema
* Remove uri_field from schema for config file
Fixes the STA provider when there is a stac-collection or process in the resources block. Also iterates through to return as many hits as specified by pygeoapi. This happened when the sta provider had an internal limit lower than the requested number of items.
On the job detail page (`/processes/PROCESS/jobs/JOB`), the link to "Jobs" used to be `./`. This
creates the url `/processes/PROCESS/jobs/`.
On that page, the link to "Jobs" is `./jobs`, which then leads you to
`/processes/PROCESS/jobs/jobs`. This however is the URL for the job with
the id `jobs`.
This patch fixes this by changing the link to jobs on the job detail page to `../jobs`.
I'm not sure if this is the best fix. Just using `.` as link doesn't seem to lead to the
desired outcome, it also generates the link `/processes/PROCESS/jobs/`.
Tom Kralidis