An implementation of clustering without locking would start by comparing the
index to the table from the beginning to find the first mismatch. Rows
before the mismatch are fine, and can be left alone. From here on, go
through the index and rewrite each row in order. This will put the rows at
Why doesn't the PL/Tcl (and PL/Python) implementation use the SPI functions?
For example pltcl_set_tuple_values() calls NameStr() instead of SPI_fname()
and heap_getattr() instead of SPI_getbinval(). Why? This makes the code
impossible to follow for someone who is not familiar with the
