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
In mappings.json we exclude the context fields from the stored source document
"_source": {
"excludes": [
"context.*"
]
},
As the context fields are populated by the importer and updater, this implies that we can't reindex in situ (aka any change in the mappings or similar requires a reimport). I can see the space savings in doing this, but on the other hand it means that we can't evolve the mappings and related aspects of Photon in a "user"-friendly fashion.
The text was updated successfully, but these errors were encountered:
I would guess that the majority of users downloads the precomputed database provided by Graphhopper and rather repeat the download than reindexing the database. So I'd consider size a much more important argument.
That is true now given that the choice isn't actually available, but given a lightweight way to update, I suspect that will change essentially immediately.
I still doubt that the average user will do that. If you want to provide a development mode with a reindexable database, now that would be a different story.
In mappings.json we exclude the context fields from the stored source document
As the context fields are populated by the importer and updater, this implies that we can't reindex in situ (aka any change in the mappings or similar requires a reimport). I can see the space savings in doing this, but on the other hand it means that we can't evolve the mappings and related aspects of Photon in a "user"-friendly fashion.
The text was updated successfully, but these errors were encountered: