Open studentX opened 8 years ago
Ok
Well, it is weird, same thing repeats when I do similar operations on postgres shell as well. I am also checking it.
There is some issue either with index(b-tree index code) or table creation(create table command), after I changed the btree index code to do a full scan in the key space, this particular error is not showing up again. Need to resolve this after the presentation.
Modification done:
src/postgres/interfaces/libpq/fe-exec.cpp : Line 3203 - Changed macro arguments from true to false, i.e. avoid escaping backslash as memcached already escapes it for us. And return value expects unescaped .
btree_index.cpp: Line 155: commented break
, search goes through the entire key space now, need to look into it
Nice! You can reference this issue in commit message with #72.
Di, can you check this?
After insertion of the second key, it fails to be retrieved, benchmark is failing because of this.