Serializers in the index_v2 of history_contents were re-
querying the database for mapped model attributes.
- eagerload collection on hdcas
- undefer _metadata on hdas
- eaglerload tags and annotations for both
- rework annotatable manager mixin to cycle through annotations
bypassing the get_item_annotation_str call (which would bypass
item.annotations even if already loaded)
Anecdotally provides significant speed up.
Renames serialization attribute contents_shown to contents_active to
better match original terminology.
Adds both state_counts and active_counts to HistoryContentsManager and
then has serialize_contents_states and serialize_contents_active call
them from the HistorySerializer.
Adds key server_starttime to the JSON returned from api/configuration.
This is the epoch time since the last server restart and may be
convenient for use in client-side caching.
This moves server_starttime from web/framework/helpers/__init__.py
(for use in script cache-busting) to web/framework/base.py.
On the client, this should be available to all pages as:
Galaxy.config.server_starttime.
- Fixes numerous problems mostly notably fixes external metadata generation and job metrics collection for recent Pulsar. This was broken when tools were given clean working directory.
- Produce a new message if an older Pulsar version is targetted via REST of library (should help with upgrading to Galaxy 16.04).
- Newer pulsar lib allows more flexible configuration of managers - update documntation to reflect this.
- Ensure all web controller history json is using view='current'
- Fix rendering of already fetched contents when
searching and/or progressively fetching details
- minor doc cleanup
Followed through with @nsoranzo's comment:
1. overwrite CSV.sniff() and CSV.set_meta() methods with the ones from BaseCSV
2. remove BaseCSV class and rename CSV to BaseCSV
3. rename ExcelCSV class to CSV
Also renamed ExcelTSV to TSV and corrected another bug in sample datatypes noticed by @nsoranzo.
carlfeberhard