1berry/postgres
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Give a hint, when [] is incorrectly used for a composite type in array.

Browse Source 
That used to be accepted, so let's try to give a hint to users on why
their PL/python functions no longer work.

Reviewed by Pavel Stehule.

Discussion: <CAH38_tmbqwaUyKs9yagyRra=SMaT45FPBxk1pmTYcM0TyXGG7Q@mail.gmail.com>
This commit is contained in:
Heikki Linnakangas 2016-10-26 10:38:56 +03:00
parent 94aceed317
commit 510e1b8ecf
7 changed files with 129 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
src/pl/plpython/expected/plpython_composite.out
View File

@ -579,3 +579,16 @@ SELECT * FROM composite_type_as_list();
{{"(first,1)","(second,1)"},{"(first,2)","(second,2)"},{"(first,3)","(second,3)"}}
(1 row)
-- Starting with PostgreSQL 10, a composite type in an array cannot be
-- represented as a Python list, because it's ambiguous with multi-dimensional
-- arrays. So this throws an error now. The error should contain a useful hint
-- on the issue.
CREATE FUNCTION composite_type_as_list_broken() RETURNS type_record[] AS $$
return [['first', 1]];
$$ LANGUAGE plpythonu;
SELECT * FROM composite_type_as_list_broken();
ERROR: malformed record literal: "first"
DETAIL: Missing left parenthesis.
HINT: To return a composite type in an array, return the composite type as a Python tuple, e.g. "[('foo')]"
CONTEXT: while creating return value
PL/Python function "composite_type_as_list_broken"

3
src/pl/plpython/plpy_cursorobject.c
View File

@ -240,7 +240,8 @@ PLy_cursor_plan(PyObject *ob, PyObject *args)
plan->values[j] =
plan->args[j].out.d.func(&(plan->args[j].out.d),
-1,
elem);
elem,
false);
}
PG_CATCH();
{

7
src/pl/plpython/plpy_exec.c
View File

@ -245,7 +245,7 @@ PLy_exec_function(FunctionCallInfo fcinfo, PLyProcedure *proc)
desc = lookup_rowtype_tupdesc(proc->result.out.d.typoid,
proc->result.out.d.typmod);
rv = PLyObject_ToCompositeDatum(&proc->result, desc, plrv);
rv = PLyObject_ToCompositeDatum(&proc->result, desc, plrv, false);
fcinfo->isnull = (rv == (Datum) NULL);
ReleaseTupleDesc(desc);
@ -253,7 +253,7 @@ PLy_exec_function(FunctionCallInfo fcinfo, PLyProcedure *proc)
else
{
fcinfo->isnull = false;
rv = (proc->result.out.d.func) (&proc->result.out.d, -1, plrv);
rv = (proc->result.out.d.func) (&proc->result.out.d, -1, plrv, false);
}
}
PG_CATCH();
@ -984,7 +984,8 @@ PLy_modify_tuple(PLyProcedure *proc, PyObject *pltd, TriggerData *tdata,
modvalues[i] = (att->func) (att,
tupdesc->attrs[atti]->atttypmod,
plval);
plval,
false);
modnulls[i] = ' ';
}
else

3
src/pl/plpython/plpy_spi.c
View File

@ -264,7 +264,8 @@ PLy_spi_execute_plan(PyObject *ob, PyObject *list, long limit)
plan->values[j] =
plan->args[j].out.d.func(&(plan->args[j].out.d),
-1,
elem);
elem,
false);
}
PG_CATCH();
{

116
src/pl/plpython/plpy_typeio.c
View File

@ -14,6 +14,7 @@
#include "parser/parse_type.h"
#include "utils/array.h"
#include "utils/builtins.h"
#include "utils/fmgroids.h"
#include "utils/lsyscache.h"
#include "utils/memutils.h"
#include "utils/numeric.h"
@ -49,21 +50,21 @@ static PyObject *PLyList_FromArray_recurse(PLyDatumToOb *elm, int *dims, int ndi
char **dataptr_p, bits8 **bitmap_p, int *bitmask_p);
/* conversion from Python objects to Datums */
static Datum PLyObject_ToBool(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLyObject_ToBytea(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLyObject_ToComposite(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLyObject_ToDatum(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLyObject_ToTransform(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLySequence_ToArray(PLyObToDatum *arg, int32 typmod, PyObject *plrv);
static Datum PLyObject_ToBool(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static Datum PLyObject_ToBytea(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static Datum PLyObject_ToComposite(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static Datum PLyObject_ToDatum(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static Datum PLyObject_ToTransform(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static Datum PLySequence_ToArray(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray);
static void PLySequence_ToArray_recurse(PLyObToDatum *elm, PyObject *list,
int *dims, int ndim, int dim,
Datum *elems, bool *nulls, int *currelem);
/* conversion from Python objects to composite Datums (used by triggers and SRFs) */
static Datum PLyString_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *string);
static Datum PLyString_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *string, bool inarray);
static Datum PLyMapping_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *mapping);
static Datum PLySequence_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *sequence);
static Datum PLyGenericObject_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *object);
static Datum PLyGenericObject_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *object, bool inarray);
void
PLy_typeinfo_init(PLyTypeInfo *arg, MemoryContext mcxt)
@ -341,12 +342,12 @@ PLyDict_FromTuple(PLyTypeInfo *info, HeapTuple tuple, TupleDesc desc)
* as an object that has __getattr__ support.
*/
Datum
PLyObject_ToCompositeDatum(PLyTypeInfo *info, TupleDesc desc, PyObject *plrv)
PLyObject_ToCompositeDatum(PLyTypeInfo *info, TupleDesc desc, PyObject *plrv, bool inarray)
{
Datum datum;
if (PyString_Check(plrv) || PyUnicode_Check(plrv))
datum = PLyString_ToComposite(info, desc, plrv);
datum = PLyString_ToComposite(info, desc, plrv, inarray);
else if (PySequence_Check(plrv))
/* composite type as sequence (tuple, list etc) */
datum = PLySequence_ToComposite(info, desc, plrv);
@ -355,7 +356,7 @@ PLyObject_ToCompositeDatum(PLyTypeInfo *info, TupleDesc desc, PyObject *plrv)
datum = PLyMapping_ToComposite(info, desc, plrv);
else
/* returned as smth, must provide method __getattr__(name) */
datum = PLyGenericObject_ToComposite(info, desc, plrv);
datum = PLyGenericObject_ToComposite(info, desc, plrv, inarray);
return datum;
}
@ -746,7 +747,7 @@ PLyList_FromArray_recurse(PLyDatumToOb *elm, int *dims, int ndim, int dim,
* type can parse.
*/
static Datum
PLyObject_ToBool(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLyObject_ToBool(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
Datum rv;
@ -765,7 +766,7 @@ PLyObject_ToBool(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
* with embedded nulls. And it's faster this way.
*/
static Datum
PLyObject_ToBytea(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLyObject_ToBytea(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
PyObject *volatile plrv_so = NULL;
Datum rv;
@ -809,7 +810,7 @@ PLyObject_ToBytea(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
* for obtaining PostgreSQL tuples.
*/
static Datum
PLyObject_ToComposite(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLyObject_ToComposite(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
Datum rv;
PLyTypeInfo info;
@ -836,7 +837,7 @@ PLyObject_ToComposite(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
* that info instead of looking it up every time a tuple is returned from
* the function.
*/
rv = PLyObject_ToCompositeDatum(&info, desc, plrv);
rv = PLyObject_ToCompositeDatum(&info, desc, plrv, inarray);
ReleaseTupleDesc(desc);
@ -908,26 +909,70 @@ PLyObject_AsString(PyObject *plrv)
* cstring into PostgreSQL type.
*/
static Datum
PLyObject_ToDatum(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLyObject_ToDatum(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
char *str;
Assert(plrv != Py_None);
str = PLyObject_AsString(plrv);
/*
* If we are parsing a composite type within an array, and the string
* isn't a valid record literal, there's a high chance that the function
* did something like:
*
* CREATE FUNCTION .. RETURNS comptype[] AS $$ return [['foo', 'bar']] $$
* LANGUAGE plpython;
*
* Before PostgreSQL 10, that was interpreted as a single-dimensional
* array, containing record ('foo', 'bar'). PostgreSQL 10 added support
* for multi-dimensional arrays, and it is now interpreted as a
* two-dimensional array, containing two records, 'foo', and 'bar'.
* record_in() will throw an error, because "foo" is not a valid record
* literal.
*
* To make that less confusing to users who are upgrading from older
* versions, try to give a hint in the typical instances of that. If we are
* parsing an array of composite types, and we see a string literal that
* is not a valid record literal, give a hint. We only want to give the
* hint in the narrow case of a malformed string literal, not any error
* from record_in(), so check for that case here specifically.
*
* This check better match the one in record_in(), so that we don't forbid
* literals that are actually valid!
*/
if (inarray && arg->typfunc.fn_oid == F_RECORD_IN)
{
char *ptr = str;
/* Allow leading whitespace */
while (*ptr && isspace((unsigned char) *ptr))
ptr++;
if (*ptr++ != '(')
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("malformed record literal: \"%s\"", str),
errdetail("Missing left parenthesis."),
errhint("To return a composite type in an array, return the composite type as a Python tuple, e.g. \"[('foo')]\"")));
}
return InputFunctionCall(&arg->typfunc,
PLyObject_AsString(plrv),
str,
arg->typioparam,
typmod);
}
static Datum
PLyObject_ToTransform(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLyObject_ToTransform(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
return FunctionCall1(&arg->typtransform, PointerGetDatum(plrv));
}
static Datum
PLySequence_ToArray(PLyObToDatum *arg, int32 typmod, PyObject *plrv)
PLySequence_ToArray(PLyObToDatum *arg, int32 typmod, PyObject *plrv, bool inarray)
{
ArrayType *array;
int i;
@ -1085,7 +1130,7 @@ PLySequence_ToArray_recurse(PLyObToDatum *elm, PyObject *list,
else
{
nulls[*currelem] = false;
elems[*currelem] = elm->func(elm, -1, obj);
elems[*currelem] = elm->func(elm, -1, obj, true);
}
Py_XDECREF(obj);
(*currelem)++;
@ -1095,7 +1140,7 @@ PLySequence_ToArray_recurse(PLyObToDatum *elm, PyObject *list,
static Datum
PLyString_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *string)
PLyString_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *string, bool inarray)
{
Datum result;
HeapTuple typeTup;
@ -1120,7 +1165,7 @@ PLyString_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *string)
ReleaseSysCache(typeTup);
result = PLyObject_ToDatum(&locinfo.out.d, desc->tdtypmod, string);
result = PLyObject_ToDatum(&locinfo.out.d, desc->tdtypmod, string, inarray);
MemoryContextDelete(cxt);
@ -1172,7 +1217,7 @@ PLyMapping_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *mapping)
}
else if (value)
{
values[i] = (att->func) (att, -1, value);
values[i] = (att->func) (att, -1, value, false);
nulls[i] = false;
}
else
@ -1265,7 +1310,7 @@ PLySequence_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *sequence)
}
else if (value)
{
values[i] = (att->func) (att, -1, value);
values[i] = (att->func) (att, -1, value, false);
nulls[i] = false;
}
@ -1294,7 +1339,7 @@ PLySequence_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *sequence)
static Datum
PLyGenericObject_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *object)
PLyGenericObject_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *object, bool inarray)
{
Datum result;
HeapTuple tuple;
@ -1335,16 +1380,29 @@ PLyGenericObject_ToComposite(PLyTypeInfo *info, TupleDesc desc, PyObject *object
}
else if (value)
{
values[i] = (att->func) (att, -1, value);
values[i] = (att->func) (att, -1, value, false);
nulls[i] = false;
}
else
{
/*
* No attribute for this column in the object.
*
* If we are parsing a composite type in an array, a likely
* cause is that the function contained something like "[[123,
* 'foo']]". Before PostgreSQL 10, that was interpreted as an
* array, with a composite type (123, 'foo') in it. But now
* it's interpreted as a two-dimensional array, and we try to
* interpret "123" as the composite type. See also similar
* heuristic in PLyObject_ToDatum().
*/
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_COLUMN),
errmsg("attribute \"%s\" does not exist in Python object", key),
errhint("To return null in a column, "
"let the returned object have an attribute named "
"after column with value None.")));
inarray ?
errhint("To return a composite type in an array, return the composite type as a Python tuple, e.g. \"[('foo')]\"") :
errhint("To return null in a column, let the returned object have an attribute named after column with value None.")));
}
Py_XDECREF(value);
value = NULL;

17
src/pl/plpython/plpy_typeio.h
View File

@ -10,8 +10,11 @@
#include "fmgr.h"
#include "storage/itemptr.h"
/*
* Conversion from PostgreSQL Datum to a Python object.
*/
struct PLyDatumToOb;
typedef PyObject *(*PLyDatumToObFunc) (struct PLyDatumToOb *, Datum);
typedef PyObject *(*PLyDatumToObFunc) (struct PLyDatumToOb *arg, Datum val);
typedef struct PLyDatumToOb
{
@ -39,11 +42,15 @@ typedef union PLyTypeInput
PLyTupleToOb r;
} PLyTypeInput;
/* convert PyObject to a Postgresql Datum or tuple.
* output from Python
/*
* Conversion from Python object to a Postgresql Datum.
*
* The 'inarray' argument to the conversion function is true, if the
* converted value was in an array (Python list). It is used to give a
* better error message in some cases.
*/
struct PLyObToDatum;
typedef Datum (*PLyObToDatumFunc) (struct PLyObToDatum *, int32, PyObject *);
typedef Datum (*PLyObToDatumFunc) (struct PLyObToDatum *arg, int32 typmod, PyObject *val, bool inarray);
typedef struct PLyObToDatum
{
@ -104,7 +111,7 @@ extern void PLy_output_tuple_funcs(PLyTypeInfo *arg, TupleDesc desc);
extern void PLy_output_record_funcs(PLyTypeInfo *arg, TupleDesc desc);
/* conversion from Python objects to composite Datums */
extern Datum PLyObject_ToCompositeDatum(PLyTypeInfo *info, TupleDesc desc, PyObject *plrv);
extern Datum PLyObject_ToCompositeDatum(PLyTypeInfo *info, TupleDesc desc, PyObject *plrv, bool isarray);
/* conversion from heap tuples to Python dictionaries */
extern PyObject *PLyDict_FromTuple(PLyTypeInfo *info, HeapTuple tuple, TupleDesc desc);

9
src/pl/plpython/sql/plpython_composite.sql
View File

@ -213,3 +213,12 @@ CREATE FUNCTION composite_type_as_list() RETURNS type_record[] AS $$
return [[('first', 1), ('second', 1)], [('first', 2), ('second', 2)], [('first', 3), ('second', 3)]];
$$ LANGUAGE plpythonu;
SELECT * FROM composite_type_as_list();
-- Starting with PostgreSQL 10, a composite type in an array cannot be
-- represented as a Python list, because it's ambiguous with multi-dimensional
-- arrays. So this throws an error now. The error should contain a useful hint
-- on the issue.
CREATE FUNCTION composite_type_as_list_broken() RETURNS type_record[] AS $$
return [['first', 1]];
$$ LANGUAGE plpythonu;
SELECT * FROM composite_type_as_list_broken();