Thread: Misinformation in 31.18 chapter.
Hey,
According to 31.18,
http://www.postgresql.org/docs/9.1/static/libpq-threading.html
"PGresult objects are read-only after creation, and so can be passed around freely between threads."
But PGresult objects are *not* read-only because of existence of
PQsetvalue() described in
http://www.postgresql.org/docs/9.1/static/libpq-misc.html
--
// Dmitriy.
According to 31.18,
http://www.postgresql.org/docs/9.1/static/libpq-threading.html
"PGresult objects are read-only after creation, and so can be passed around freely between threads."
But PGresult objects are *not* read-only because of existence of
PQsetvalue() described in
http://www.postgresql.org/docs/9.1/static/libpq-misc.html
--
// Dmitriy.
Dmitriy Igrishin <dmitigr@gmail.com> writes: > According to 31.18, > http://www.postgresql.org/docs/9.1/static/libpq-threading.html > "PGresult objects are read-only after creation, and so can be passed around > freely between threads." > But PGresult objects are *not* read-only because of existence of > PQsetvalue() described in > http://www.postgresql.org/docs/9.1/static/libpq-misc.html Good point ... and the "event" stuff can modify the state of a PGresult, too. I've updated that statement accordingly. Thanks for the note. regards, tom lane