From 32fb6da1fc540031be6e742ec68ad80ba0c3760b Mon Sep 17 00:00:00 2001 From: Mike Boers Date: Tue, 22 Sep 2015 14:06:12 -0700 Subject: [PATCH] Notes --- TODO.md | 67 +++------------------------------------------------------ 1 file changed, 3 insertions(+), 64 deletions(-) diff --git a/TODO.md b/TODO.md index 7ed3139..e76efb6 100644 --- a/TODO.md +++ b/TODO.md @@ -1,70 +1,9 @@ -- Schema.for_url(base_url) -> static class - - - -- role assignments for columns, so that our tools that - access roles (via a special syntax) instead of actual column names - - e.g.: PublishEvent.$version -> PublishEvent.sg_version_1 - - Can have non-alnum in there, e.g.: PublishEvent.$sgpublish:publish:type - -- entity type renames, so that we can use custom entities for - whatever we want, e.g.: - - MyType: CustomEntity02 - -- Arbitrary tags/meta, e.g. if something is used by sgcache or not. - - EntityType.field: sgcache: include: true - - Could we then have xpath like expressions? - e.g.: EntityType.[sgcache.include==true] - PublishEvent.[sgpublish.is_core] - - Tags: PublishEvent.#sgsession:core -> {sg_code,sg_type,...} - -- Automatic sg_ prefix detection: - Publish.type -> PublishEvent.sg_type - - Have a "doctor" which tells us the potential problems with our schema, - such as two columns that are the same minus the prefix - -- Force a specific name, to skip the rewriting rules, e.g.: Publish.!type - This is more in SGSession (or other consumers) - - -- Public API: - - schema.resolve_entity('$Publish') -> ['PublishEvent'] - schema.resolve_field('PublishEvent', 'type') -> ['sg_type'] - - Create a standard-ish set of tags and aliases: $parent pointing to typical parent -- Include backrefs in reduced schema? Known as "inverse_association" in private - schema. - - - - - - - - - - - - - - - - - - - - - +- Include inverse_association (from private schema) +- Schema.resolve_one -> error if there isn't just one +- *type -> explicit prefix matching