You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Since this API is a compact software and does not too much stuff at all, it seems preferable for its application together with Discourse to re-implement it as an extension of the Discourse API. This will have major performance improvements when updating the pre-generated files ("seconds, not hours") as a direct connection to the Discourse database and all of its internal caching in Redis is available. This would also make it possible to always analyze the real-time Discourse data.
In turn, the "import from Discourse" feature will no longer be needed in this software and could be removed. This software would then be a standalone implementation of the Graphryder API that can be used when one wants to analyze data not coming from Discourse but from some sort or generic, other data source.
After this, there would be some name changes:
This software would be renamed to graphryder-api-standalone.
The Discourse plugin, providing another implementation of the API, would be called graphryder-api-discourse.
The text was updated successfully, but these errors were encountered:
Actually there is no need to remove the Discourse importer code from this software. It may still be useful for anyone wanting to analyze Discourse forum content of a forum they don't have admin control over (such as their own forum hosted with a SaaS solution, which usually means that they cannot install plugins there).
tanius
changed the title
Move the "import from Discourse" part to Discourse
Re-implement the "import from Discourse" part in Discourse
Oct 23, 2019
Since this API is a compact software and does not too much stuff at all, it seems preferable for its application together with Discourse to re-implement it as an extension of the Discourse API. This will have major performance improvements when updating the pre-generated files ("seconds, not hours") as a direct connection to the Discourse database and all of its internal caching in Redis is available. This would also make it possible to always analyze the real-time Discourse data.
In turn, the "import from Discourse" feature will no longer be needed in this software and could be removed. This software would then be a standalone implementation of the Graphryder API that can be used when one wants to analyze data not coming from Discourse but from some sort or generic, other data source.
After this, there would be some name changes:
graphryder-api-standalone
.graphryder-api-discourse
.The text was updated successfully, but these errors were encountered: