Configuration

To simplify workload, SearchReplica relies on rarelly used sql COMMENTs. Defining them on table or field allows skip, rename fields, inline one table into another document, etc.

PostgreSQL COMMENT is a Text field which can store values of any length (up to 1GB)

You define comments with basic ommands

COMMENT ON TABLE your.tablename IS '...conftag values here...';
COMMENT ON COLUMN  your.tablename.columnname IS '...conftag values here...';

Syntax is slightly changed struct tags from Golang, if you familiar.

It’s a string of space separated tags, where tag starts with name followed by semicolon.
3 spacebars in a row or # stops tag parsing leaving end part as human readable comment.

tagname:"positional1,,positional3,other,whatever" tag2:"val1,val2"    Some regular human readable comment

COMMENT ON TABLE posts IS 'index:"blogposts,all" inline:"blog_comments,comments"    PLEASE DONT DELETE ROWS';
COMMENT ON COLUMN posts.id IS 'index:"-,pk" # This would use id as _id and remove it from json document';

Tag values can be positional, or Flags which follow positional values.

• index:
  • docType (- means do not index)
  • Flags:
    • all (deprecated) index all fields in this table
• inline:
  • inline name, that will be injected as a field
  • field name
  • (optional) script name for adding
  • (optional) script name for removal
• join:
  • field name
  • (optional) type name (by default docType is used)

• index:
  • rename (- means do not index)
  • Flags:
    • pk this value, prefixed by table name would be used as document _id.
    • id value would be used as document _id field
    • routing value would be used for routing
• inline:
  • inline name, where this field should be used
  • rename (within inline) NOT IMPLEMENTED
  • Flags:
    • pk inline PK
    • parent parent _id
    • (optional) routing routing value of PARENT document. (in order to find and update it)
• join:
  • Flags:
    • parent field used as parent value
    • name field is used as join type (for polymorphism)

from demo/schema.sql

COMMENT ON TABLE "main_doc" IS 'index:"main,all" join:"join,immaparent" inline:"inline_name,inlined_field"';
COMMENT ON COLUMN "main_doc".id IS 'index:",routing,id"';
COMMENT ON COLUMN "main_doc".ignore_me IS 'index:"-"';

/* Just inline and do not index as separate document.
We need to specify inline PK, parent document ID and optionally routing.
Each inlined field should be explicitly defined as such. */
COMMENT ON TABLE "inline_doc" IS 'index:"-"';
COMMENT ON COLUMN "inline_doc".parent_id  IS 'inline:"inline_name,_pk,parent,routing"';
COMMENT ON COLUMN "inline_doc".id  IS 'inline:"inline_name,pk"';
COMMENT ON COLUMN "inline_doc".value  IS 'inline:"inline_name"';


/* Be aware, child document should always be in the same shard as parent, thus routing field is used. It can be any shard key, or just a parent ID like shown below */
COMMENT ON TABLE "child_doc" IS 'index:"child,all" join:"join,immachild"';
COMMENT ON COLUMN "child_doc".id IS 'index:",id"';
COMMENT ON COLUMN "child_doc".parent_id IS 'index:",routing" join:"parent"';
COMMENT ON COLUMN "child_doc".ignore_me IS 'index:"-"';