| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
| |
It makes more sense this way, because otherwise it must be passed to every call
of `read_message()`.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
This change exposes lower level functions for operating the
(logical) replication protocol, while keeping the high-level
start_replication function that does all the job for you in
case of a synchronous connection.
A number of other changes and fixes are put into this commit.
|
| |
|
|
|
|
| |
Also, don't start an implicit transaction when fetching with
named with hold cursor, since it already returns results
from a previously committed transaction.
|
| |
|
|
| |
PQclear already guards against NULL, so the extra check is redundant
|
| | |
|
| |
|
|
|
| |
Also more return values checked for values < 0 for errors, instead of
checking == 0 and leaving the positive side unchecked
|
| | |
|
| |
|
|
|
| |
Be more consistent in using 0 for success, <0 for failure, and to check
for values < 0 instead of specific -1.
|
| |
|
|
|
|
|
| |
The aim of these function is to allow the connection to make a better use
of the pqpath functions instead of using PQexec for these small things.
Also, the functions are to be called with the connection lock: this makes
composing higher level functions using them easier.
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
.c files only need to import psycopg.h: it will in turn import
dependencies from Python and libpq and configure.h. psycopg.h should be
the first to be imported, so the basic imports are not required in
the .h's
As a guideline I'm trying to import from the most specific to the most
generic to detect missing imports in the .h's.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
Dropped calls to PQisBusy/PQconsumeInput in the connection code.
|
| |
|
|
| |
The function reads the last result after an asynchronous query.
|
| |
|
|
|
|
|
| |
The function is called without holding the GIL. Because it is necessary
to execute the Python callback if set, we need to re-acquire the GIL and
tnen release it again. In order to correctly bookkeep the thread state,
the pointer of the _save variable is passed to the function.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
| |
that should not be exported from the module. This results in a 5%
reduction in code size and shortens the dynamic symbol table.
* psycopg/config.h: If GCC >= 4.0 is installed, define the HIDDEN
symbol to apply the "hidden" visibility attribute.
|
| |
|
|
|
|
| |
function, and simplify.
(pq_resolve_critical): make function static, since it isn't being
used outside of pqpath.c any more.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
pq_abort_locked() prototype.
(conn_switch_isolation_level): fix for new pq_abort_locked()
prototype, and use pq_complete_error() to show error message.
(conn_set_client_encoding): same here.
* psycopg/pqpath.c (pq_execute_command_locked): remove static
modifier.
(pq_complete_error): same here.
(pq_abort_locked): add pgres and error arguments.
(pq_abort): call pq_abort_locked() to reduce code duplication.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* psycopg/pqpath.c (pq_execute_command_locked): add an error
argument to hold an error when no PGresult is returned by PQexec,
rather than using pq_set_critical().
(pq_complete_error): new function that converts the error returned
by pq_execute_command_locked() to a Python exception.
(pq_begin_locked): add error argument.
(pq_commit): use pq_complete_error().
(pq_abort): use pq_complete_error().
(pq_abort_locked): always call pq_set_critical() on error, and
clear the error message from pq_execute_command_locked().
(pq_execute): use pq_complete_error() to handle the error from
pq_begin_locked().
* psycopg/pqpath.c (pq_begin): remove unused function.
* psycopg/connection_type.c (psyco_conn_commit): if conn_commit()
raises an error, just return NULL, since it is now setting an
exception itself.
(psyco_conn_rollback): same here.
* psycopg/connection_int.c (conn_commit): don't drop GIL and lock
connection before calling pq_commit().
(conn_rollback): same here.
(conn_close): use pq_abort_locked().
(conn_switch_isolation_level): same here.
(conn_set_client_encoding): same here.
* psycopg/pqpath.h: add prototype for pq_abort_locked().
* psycopg/pqpath.c (pq_commit): convert function to run with GIL
held, and handle errors appropriately.
(pq_abort): same here.
(pq_abort_locked): new function to abort a locked connection.
2007-12-22 James Henstridge <james@jamesh.id.au>
* psycopg/pqpath.c (pq_raise): add a "pgres" argument so we can
generate nice errors not related to a particular cursor.
(pq_execute): use pq_begin_locked() rather than pq_begin(). Use
pq_raise() to handle any errors from it.
* psycopg/pqpath.c (pq_execute_command_locked): helper function
used to execute a command-style query on a locked connection.
(pq_begin_locked): a variant of pq_begin() that uses
pq_execute_command_locked().
(pq_begin): rewrite to use pq_begin_locked().
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
of the exception message if it actually gives the severity.
* psycopg/pqpath.h (pq_resolve_critical): add prototype, since
this function is being used from connection_int.c.
* psycopg/psycopg.h: update psyco_set_error() prototype.
* psycopg/psycopgmodule.c (psyco_errors_init): set pgerror, pgcode
and cursor class attributes to None on psycopg2.Error so that the
attributes will always be available (simplifies error handling).
(psyco_set_error): add const qualifiers to msg, pgerror and pgcode
arguments.
Don't bother setting pgerror, pgcode or cursor to None if they are
not provided -- the class defaults take care of this.
|
| |
|
