Re: Cleaning up array_in() - Mailing list pgsql-hackers
| From | Tom Lane |
|---|---|
| Subject | Re: Cleaning up array_in() |
| Date | |
| Msg-id | 557218.1699401129@sss.pgh.pa.us Whole thread Raw |
| In response to | Re: Cleaning up array_in() (jian he <jian.universality@gmail.com>) |
| Responses |
Re: Cleaning up array_in()
Re: Cleaning up array_in() |
| List | pgsql-hackers |
I got back to looking at this today (sorry for delay), and did a pass
of code review. I think we are getting pretty close to something
committable. The one loose end IMO is this bit in ReadArrayToken:
+ case '"':
+
+ /*
+ * XXX "Unexpected %c character" would be more apropos, but
+ * this message is what the pre-v17 implementation produced,
+ * so we'll keep it for now.
+ */
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected array element.")));
+ return ATOK_ERROR;
This comes out when you write something like '{foo"bar"}', and I'd
say the choice of message is not great. On the other hand, it's
consistent with what you get from '{"foo""bar"}', and if we wanted
to change that too then some tweaking of the state machine in
ReadArrayStr would be required (or else modify ReadArrayToken so
it doesn't return instantly upon seeing the second quote mark).
I'm not sure that this is worth messing with.
Anyway, I think we are well past the point where splitting the patch
into multiple parts is worth doing, because we've rewritten pretty
much all of this code, and the intermediate versions are not terribly
helpful. So I just folded it all into one patch.
Some notes about specific points:
* Per previous discussion, I undid the change to allow "[1:0]"
dimensions, but I left a comment behind about that.
* Removing the ArrayGetNItemsSafe/ArrayCheckBoundsSafe calls
seems OK, but then we need to be more careful about detecting
overflows and disallowed cases in ReadArrayDimensions.
* I don't think the ARRAYDEBUG code is of any value whatever.
The fact that nobody bothered to improve it to print more than
the dim[] values proves it hasn't been used in decades.
Let's just nuke it.
* We can simplify the state machine in ReadArrayStr some more: it
seems to me it's sufficient to track "expect_delim", as long as you
realize that that really means "expect typdelim or right brace".
(Maybe another name would be better? I couldn't think of anything
short though.)
* I switched to using a StringInfo instead of a fixed-size elembuf,
as Heikki speculated about.
* I added some more test cases to cover things that evidently weren't
sufficiently tested, like the has_escapes business which was flat
out broken in v10, and to improve the code coverage report.
Comments?
regards, tom lane
diff --git a/src/backend/utils/adt/arrayfuncs.c b/src/backend/utils/adt/arrayfuncs.c
index 3ff13eb419..638e4f0382 100644
--- a/src/backend/utils/adt/arrayfuncs.c
+++ b/src/backend/utils/adt/arrayfuncs.c
@@ -54,18 +54,16 @@ bool Array_nulls = true;
PG_FREE_IF_COPY(array, n); \
} while (0)
+/* ReadArrayToken return type */
typedef enum
{
- ARRAY_NO_LEVEL,
- ARRAY_LEVEL_STARTED,
- ARRAY_ELEM_STARTED,
- ARRAY_ELEM_COMPLETED,
- ARRAY_QUOTED_ELEM_STARTED,
- ARRAY_QUOTED_ELEM_COMPLETED,
- ARRAY_ELEM_DELIMITED,
- ARRAY_LEVEL_COMPLETED,
- ARRAY_LEVEL_DELIMITED,
-} ArrayParseState;
+ ATOK_LEVEL_START,
+ ATOK_LEVEL_END,
+ ATOK_DELIM,
+ ATOK_ELEM,
+ ATOK_ELEM_NULL,
+ ATOK_ERROR,
+} ArrayToken;
/* Working state for array_iterate() */
typedef struct ArrayIteratorData
@@ -91,15 +89,21 @@ typedef struct ArrayIteratorData
int current_item; /* the item # we're at in the array */
} ArrayIteratorData;
-static int ArrayCount(const char *str, int *dim, char typdelim,
- Node *escontext);
-static bool ReadArrayStr(char *arrayStr, const char *origStr,
- int nitems, int ndim, int *dim,
+static bool ReadArrayDimensions(char **srcptr, int *ndim_p,
+ int *dim, int *lBound,
+ const char *origStr, Node *escontext);
+static bool ReadDimensionInt(char **srcptr, int *result,
+ const char *origStr, Node *escontext);
+static bool ReadArrayStr(char **srcptr,
FmgrInfo *inputproc, Oid typioparam, int32 typmod,
char typdelim,
int typlen, bool typbyval, char typalign,
- Datum *values, bool *nulls,
- bool *hasnulls, int32 *nbytes, Node *escontext);
+ int *ndim_p, int *dim,
+ int *nitems_p,
+ Datum **values_p, bool **nulls_p,
+ const char *origStr, Node *escontext);
+static ArrayToken ReadArrayToken(char **srcptr, StringInfo elembuf, char typdelim,
+ const char *origStr, Node *escontext);
static void ReadArrayBinary(StringInfo buf, int nitems,
FmgrInfo *receiveproc, Oid typioparam, int32 typmod,
int typlen, bool typbyval, char typalign,
@@ -185,12 +189,10 @@ array_in(PG_FUNCTION_ARGS)
char typalign;
char typdelim;
Oid typioparam;
- char *string_save,
- *p;
- int i,
- nitems;
- Datum *dataPtr;
- bool *nullsPtr;
+ char *p;
+ int nitems;
+ Datum *values;
+ bool *nulls;
bool hasnulls;
int32 nbytes;
int32 dataoffset;
@@ -233,104 +235,34 @@ array_in(PG_FUNCTION_ARGS)
typdelim = my_extra->typdelim;
typioparam = my_extra->typioparam;
- /* Make a modifiable copy of the input */
- string_save = pstrdup(string);
+ /* Initialize dim, lBound for ReadArrayDimensions, ReadArrayStr */
+ for (int i = 0; i < MAXDIM; i++)
+ {
+ dim[i] = -1; /* indicates "not yet known" */
+ lBound[i] = 1; /* default lower bound */
+ }
/*
- * If the input string starts with dimension info, read and use that.
- * Otherwise, we require the input to be in curly-brace style, and we
- * prescan the input to determine dimensions.
+ * Start processing the input string.
*
- * Dimension info takes the form of one or more [n] or [m:n] items. The
- * outer loop iterates once per dimension item.
+ * If the input string starts with dimension info, read and use that.
+ * Otherwise, we'll determine the dimensions during ReadArrayStr.
*/
- p = string_save;
- ndim = 0;
- for (;;)
- {
- char *q;
- int ub;
-
- /*
- * Note: we currently allow whitespace between, but not within,
- * dimension items.
- */
- while (scanner_isspace(*p))
- p++;
- if (*p != '[')
- break; /* no more dimension items */
- p++;
- if (ndim >= MAXDIM)
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
- errmsg("number of array dimensions (%d) exceeds the maximum allowed (%d)",
- ndim + 1, MAXDIM)));
-
- for (q = p; isdigit((unsigned char) *q) || (*q == '-') || (*q == '+'); q++)
- /* skip */ ;
- if (q == p) /* no digits? */
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", string),
- errdetail("\"[\" must introduce explicitly-specified array dimensions.")));
-
- if (*q == ':')
- {
- /* [m:n] format */
- *q = '\0';
- lBound[ndim] = atoi(p);
- p = q + 1;
- for (q = p; isdigit((unsigned char) *q) || (*q == '-') || (*q == '+'); q++)
- /* skip */ ;
- if (q == p) /* no digits? */
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", string),
- errdetail("Missing array dimension value.")));
- }
- else
- {
- /* [n] format */
- lBound[ndim] = 1;
- }
- if (*q != ']')
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", string),
- errdetail("Missing \"%s\" after array dimensions.",
- "]")));
-
- *q = '\0';
- ub = atoi(p);
- p = q + 1;
- if (ub < lBound[ndim])
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
- errmsg("upper bound cannot be less than lower bound")));
-
- dim[ndim] = ub - lBound[ndim] + 1;
- ndim++;
- }
+ p = string;
+ if (!ReadArrayDimensions(&p, &ndim, dim, lBound, string, escontext))
+ return (Datum) 0;
if (ndim == 0)
{
- /* No array dimensions, so intuit dimensions from brace structure */
+ /* No array dimensions, so next character should be a left brace */
if (*p != '{')
ereturn(escontext, (Datum) 0,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("malformed array literal: \"%s\"", string),
errdetail("Array value must start with \"{\" or dimension information.")));
- ndim = ArrayCount(p, dim, typdelim, escontext);
- if (ndim < 0)
- PG_RETURN_NULL();
- for (i = 0; i < ndim; i++)
- lBound[i] = 1;
}
else
{
- int ndim_braces,
- dim_braces[MAXDIM];
-
/* If array dimensions are given, expect '=' operator */
if (strncmp(p, ASSGN, strlen(ASSGN)) != 0)
ereturn(escontext, (Datum) 0,
@@ -339,66 +271,68 @@ array_in(PG_FUNCTION_ARGS)
errdetail("Missing \"%s\" after array dimensions.",
ASSGN)));
p += strlen(ASSGN);
+ /* Allow whitespace after it */
while (scanner_isspace(*p))
p++;
- /*
- * intuit dimensions from brace structure -- it better match what we
- * were given
- */
if (*p != '{')
ereturn(escontext, (Datum) 0,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("malformed array literal: \"%s\"", string),
errdetail("Array contents must start with \"{\".")));
- ndim_braces = ArrayCount(p, dim_braces, typdelim, escontext);
- if (ndim_braces < 0)
- PG_RETURN_NULL();
- if (ndim_braces != ndim)
+ }
+
+ /* Parse the value part, in the curly braces: { ... } */
+ if (!ReadArrayStr(&p,
+ &my_extra->proc, typioparam, typmod,
+ typdelim,
+ typlen, typbyval, typalign,
+ &ndim,
+ dim,
+ &nitems,
+ &values, &nulls,
+ string,
+ escontext))
+ return (Datum) 0;
+
+ /* only whitespace is allowed after the closing brace */
+ while (*p)
+ {
+ if (!scanner_isspace(*p++))
ereturn(escontext, (Datum) 0,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
errmsg("malformed array literal: \"%s\"", string),
- errdetail("Specified array dimensions do not match array contents.")));
- for (i = 0; i < ndim; ++i)
- {
- if (dim[i] != dim_braces[i])
- ereturn(escontext, (Datum) 0,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", string),
- errdetail("Specified array dimensions do not match array contents.")));
- }
+ errdetail("Junk after closing right brace.")));
}
-#ifdef ARRAYDEBUG
- printf("array_in- ndim %d (", ndim);
- for (i = 0; i < ndim; i++)
- {
- printf(" %d", dim[i]);
- };
- printf(") for %s\n", string);
-#endif
-
- /* This checks for overflow of the array dimensions */
- nitems = ArrayGetNItemsSafe(ndim, dim, escontext);
- if (nitems < 0)
- PG_RETURN_NULL();
- if (!ArrayCheckBoundsSafe(ndim, dim, lBound, escontext))
- PG_RETURN_NULL();
-
/* Empty array? */
if (nitems == 0)
PG_RETURN_ARRAYTYPE_P(construct_empty_array(element_type));
- dataPtr = (Datum *) palloc(nitems * sizeof(Datum));
- nullsPtr = (bool *) palloc(nitems * sizeof(bool));
- if (!ReadArrayStr(p, string,
- nitems, ndim, dim,
- &my_extra->proc, typioparam, typmod,
- typdelim,
- typlen, typbyval, typalign,
- dataPtr, nullsPtr,
- &hasnulls, &nbytes, escontext))
- PG_RETURN_NULL();
+ /*
+ * Check for nulls, compute total data space needed
+ */
+ hasnulls = false;
+ nbytes = 0;
+ for (int i = 0; i < nitems; i++)
+ {
+ if (nulls[i])
+ hasnulls = true;
+ else
+ {
+ /* let's just make sure data is not toasted */
+ if (typlen == -1)
+ values[i] = PointerGetDatum(PG_DETOAST_DATUM(values[i]));
+ nbytes = att_addlength_datum(nbytes, typlen, values[i]);
+ nbytes = att_align_nominal(nbytes, typalign);
+ /* check for overflow of total request */
+ if (!AllocSizeIsValid(nbytes))
+ ereturn(escontext, (Datum) 0,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("array size exceeds the maximum allowed (%d)",
+ (int) MaxAllocSize)));
+ }
+ }
if (hasnulls)
{
dataoffset = ARR_OVERHEAD_WITHNULLS(ndim, nitems);
@@ -409,6 +343,10 @@ array_in(PG_FUNCTION_ARGS)
dataoffset = 0; /* marker for no null bitmap */
nbytes += ARR_OVERHEAD_NONULLS(ndim);
}
+
+ /*
+ * Construct the final array datum
+ */
retval = (ArrayType *) palloc0(nbytes);
SET_VARSIZE(retval, nbytes);
retval->ndim = ndim;
@@ -424,302 +362,216 @@ array_in(PG_FUNCTION_ARGS)
memcpy(ARR_LBOUND(retval), lBound, ndim * sizeof(int));
CopyArrayEls(retval,
- dataPtr, nullsPtr, nitems,
+ values, nulls, nitems,
typlen, typbyval, typalign,
true);
- pfree(dataPtr);
- pfree(nullsPtr);
- pfree(string_save);
+ pfree(values);
+ pfree(nulls);
PG_RETURN_ARRAYTYPE_P(retval);
}
/*
- * ArrayCount
- * Determines the dimensions for an array string.
+ * ReadArrayDimensions
+ * parses the array dimensions part of the input and converts the values
+ * to internal format.
+ *
+ * On entry, *srcptr points to the string to parse. It is advanced to point
+ * after whitespace (if any) and dimension info (if any).
+ *
+ * *ndim_p, *dim, and *lBound are output variables. They are filled with the
+ * number of dimensions (<= MAXDIM), the length of each dimension, and the
+ * lower bounds of the slices, respectively. *ndim_p will be set to zero
+ * when no dimensions appear.
+ *
+ * 'origStr' is the original input string, used only in error messages.
+ * If *escontext points to an ErrorSaveContext, details of any error are
+ * reported there.
*
- * Returns number of dimensions as function result. The axis lengths are
- * returned in dim[], which must be of size MAXDIM.
+ * Result:
+ * true for success, false for failure (if escontext is provided).
*
- * If we detect an error, fill *escontext with error details and return -1
- * (unless escontext isn't provided, in which case errors will be thrown).
+ * Note that dim[] and lBound[] are allocated by the caller, and must have
+ * MAXDIM elements.
*/
-static int
-ArrayCount(const char *str, int *dim, char typdelim, Node *escontext)
+static bool
+ReadArrayDimensions(char **srcptr, int *ndim_p, int *dim, int *lBound,
+ const char *origStr, Node *escontext)
{
- int nest_level = 0,
- i;
- int ndim = 1,
- temp[MAXDIM],
- nelems[MAXDIM],
- nelems_last[MAXDIM];
- bool in_quotes = false;
- bool eoArray = false;
- bool empty_array = true;
- const char *ptr;
- ArrayParseState parse_state = ARRAY_NO_LEVEL;
+ char *p = *srcptr;
+ int ndim;
- for (i = 0; i < MAXDIM; ++i)
+ /*
+ * Dimension info takes the form of one or more [n] or [m:n] items. This
+ * loop iterates once per dimension item.
+ */
+ ndim = 0;
+ for (;;)
{
- temp[i] = dim[i] = nelems_last[i] = 0;
- nelems[i] = 1;
- }
+ char *q;
+ int ub;
+ int i;
- ptr = str;
- while (!eoArray)
- {
- bool itemdone = false;
+ /*
+ * Note: we currently allow whitespace between, but not within,
+ * dimension items.
+ */
+ while (scanner_isspace(*p))
+ p++;
+ if (*p != '[')
+ break; /* no more dimension items */
+ p++;
+ if (ndim >= MAXDIM)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("number of array dimensions (%d) exceeds the maximum allowed (%d)",
+ ndim + 1, MAXDIM)));
- while (!itemdone)
- {
- if (parse_state == ARRAY_ELEM_STARTED ||
- parse_state == ARRAY_QUOTED_ELEM_STARTED)
- empty_array = false;
+ q = p;
+ if (!ReadDimensionInt(&p, &i, origStr, escontext))
+ return false;
+ if (p == q) /* no digits? */
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("\"[\" must introduce explicitly-specified array dimensions.")));
- switch (*ptr)
- {
- case '\0':
- /* Signal a premature end of the string */
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected end of input.")));
- case '\\':
-
- /*
- * An escape must be after a level start, after an element
- * start, or after an element delimiter. In any case we
- * now must be past an element start.
- */
- if (parse_state != ARRAY_LEVEL_STARTED &&
- parse_state != ARRAY_ELEM_STARTED &&
- parse_state != ARRAY_QUOTED_ELEM_STARTED &&
- parse_state != ARRAY_ELEM_DELIMITED)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected \"%c\" character.",
- '\\')));
- if (parse_state != ARRAY_QUOTED_ELEM_STARTED)
- parse_state = ARRAY_ELEM_STARTED;
- /* skip the escaped character */
- if (*(ptr + 1))
- ptr++;
- else
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected end of input.")));
- break;
- case '"':
-
- /*
- * A quote must be after a level start, after a quoted
- * element start, or after an element delimiter. In any
- * case we now must be past an element start.
- */
- if (parse_state != ARRAY_LEVEL_STARTED &&
- parse_state != ARRAY_QUOTED_ELEM_STARTED &&
- parse_state != ARRAY_ELEM_DELIMITED)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected array element.")));
- in_quotes = !in_quotes;
- if (in_quotes)
- parse_state = ARRAY_QUOTED_ELEM_STARTED;
- else
- parse_state = ARRAY_QUOTED_ELEM_COMPLETED;
- break;
- case '{':
- if (!in_quotes)
- {
- /*
- * A left brace can occur if no nesting has occurred
- * yet, after a level start, or after a level
- * delimiter.
- */
- if (parse_state != ARRAY_NO_LEVEL &&
- parse_state != ARRAY_LEVEL_STARTED &&
- parse_state != ARRAY_LEVEL_DELIMITED)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected \"%c\" character.",
- '{')));
- parse_state = ARRAY_LEVEL_STARTED;
- if (nest_level >= MAXDIM)
- ereturn(escontext, -1,
- (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
- errmsg("number of array dimensions (%d) exceeds the maximum allowed (%d)",
- nest_level + 1, MAXDIM)));
- temp[nest_level] = 0;
- nest_level++;
- if (ndim < nest_level)
- ndim = nest_level;
- }
- break;
- case '}':
- if (!in_quotes)
- {
- /*
- * A right brace can occur after an element start, an
- * element completion, a quoted element completion, or
- * a level completion.
- */
- if (parse_state != ARRAY_ELEM_STARTED &&
- parse_state != ARRAY_ELEM_COMPLETED &&
- parse_state != ARRAY_QUOTED_ELEM_COMPLETED &&
- parse_state != ARRAY_LEVEL_COMPLETED &&
- !(nest_level == 1 && parse_state == ARRAY_LEVEL_STARTED))
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected \"%c\" character.",
- '}')));
- parse_state = ARRAY_LEVEL_COMPLETED;
- if (nest_level == 0)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unmatched \"%c\" character.", '}')));
- nest_level--;
-
- if (nelems_last[nest_level] != 0 &&
- nelems[nest_level] != nelems_last[nest_level])
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Multidimensional arrays must have "
- "sub-arrays with matching "
- "dimensions.")));
- nelems_last[nest_level] = nelems[nest_level];
- nelems[nest_level] = 1;
- if (nest_level == 0)
- eoArray = itemdone = true;
- else
- {
- /*
- * We don't set itemdone here; see comments in
- * ReadArrayStr
- */
- temp[nest_level - 1]++;
- }
- }
- break;
- default:
- if (!in_quotes)
- {
- if (*ptr == typdelim)
- {
- /*
- * Delimiters can occur after an element start, an
- * element completion, a quoted element
- * completion, or a level completion.
- */
- if (parse_state != ARRAY_ELEM_STARTED &&
- parse_state != ARRAY_ELEM_COMPLETED &&
- parse_state != ARRAY_QUOTED_ELEM_COMPLETED &&
- parse_state != ARRAY_LEVEL_COMPLETED)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected \"%c\" character.",
- typdelim)));
- if (parse_state == ARRAY_LEVEL_COMPLETED)
- parse_state = ARRAY_LEVEL_DELIMITED;
- else
- parse_state = ARRAY_ELEM_DELIMITED;
- itemdone = true;
- nelems[nest_level - 1]++;
- }
- else if (!scanner_isspace(*ptr))
- {
- /*
- * Other non-space characters must be after a
- * level start, after an element start, or after
- * an element delimiter. In any case we now must
- * be past an element start.
- */
- if (parse_state != ARRAY_LEVEL_STARTED &&
- parse_state != ARRAY_ELEM_STARTED &&
- parse_state != ARRAY_ELEM_DELIMITED)
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Unexpected array element.")));
- parse_state = ARRAY_ELEM_STARTED;
- }
- }
- break;
- }
- if (!itemdone)
- ptr++;
+ if (*p == ':')
+ {
+ /* [m:n] format */
+ lBound[ndim] = i;
+ p++;
+ q = p;
+ if (!ReadDimensionInt(&p, &ub, origStr, escontext))
+ return false;
+ if (p == q) /* no digits? */
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Missing array dimension value.")));
+ }
+ else
+ {
+ /* [n] format */
+ lBound[ndim] = 1;
+ ub = i;
}
- temp[ndim - 1]++;
- ptr++;
+ if (*p != ']')
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Missing \"%s\" after array dimensions.",
+ "]")));
+ p++;
+
+ /*
+ * Note: we could accept ub = lb-1 to represent a zero-length
+ * dimension. However, that would result in an empty array, for which
+ * we don't keep any dimension data, so that e.g. [1:0] and [101:100]
+ * would be equivalent. Given the lack of field demand, there seems
+ * little point in allowing such cases.
+ */
+ if (ub < lBound[ndim])
+ ereturn(escontext, false,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("upper bound cannot be less than lower bound")));
+
+ /* Upper bound of INT_MAX must be disallowed, cf ArrayCheckBounds() */
+ if (ub == INT_MAX)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("array upper bound is too large: %d", ub)));
+
+ /* Compute "ub - lBound[ndim] + 1", detecting overflow */
+ if (pg_sub_s32_overflow(ub, lBound[ndim], &ub) ||
+ pg_add_s32_overflow(ub, 1, &ub))
+ ereport(ERROR,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("array size exceeds the maximum allowed (%d)",
+ (int) MaxArraySize)));
+
+ dim[ndim] = ub;
+ ndim++;
}
- /* only whitespace is allowed after the closing brace */
- while (*ptr)
+ *srcptr = p;
+ *ndim_p = ndim;
+ return true;
+}
+
+/*
+ * ReadDimensionInt
+ * parse an integer, for the array dimensions
+ *
+ * On entry, *srcptr points to the string to parse. It is advanced past the
+ * digits of the integer. If there are no digits, returns true and leaves
+ * *srcptr unchanged.
+ *
+ * Result:
+ * true for success, false for failure (if escontext is provided).
+ * On success, the parsed integer is returned in *result.
+ */
+static bool
+ReadDimensionInt(char **srcptr, int *result,
+ const char *origStr, Node *escontext)
+{
+ char *p = *srcptr;
+ long l;
+
+ /* don't accept leading whitespace */
+ if (!isdigit((unsigned char) *p) && *p != '-' && *p != '+')
{
- if (!scanner_isspace(*ptr++))
- ereturn(escontext, -1,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"", str),
- errdetail("Junk after closing right brace.")));
+ *result = 0;
+ return true;
}
- /* special case for an empty array */
- if (empty_array)
- return 0;
+ errno = 0;
+ l = strtol(p, srcptr, 10);
- for (i = 0; i < ndim; ++i)
- dim[i] = temp[i];
+ if (errno == ERANGE || l > PG_INT32_MAX || l < PG_INT32_MIN)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("array bound is out of integer range")));
- return ndim;
+ *result = (int) l;
+ return true;
}
/*
* ReadArrayStr :
- * parses the array string pointed to by "arrayStr" and converts the values
- * to internal format. Unspecified elements are initialized to nulls.
- * The array dimensions must already have been determined.
+ * parses the array string pointed to by *srcptr and converts the values
+ * to internal format. Determines the array dimensions as it goes.
*
- * Inputs:
- * arrayStr: the string to parse.
- * CAUTION: the contents of "arrayStr" will be modified!
- * origStr: the unmodified input string, used only in error messages.
- * nitems: total number of array elements, as already determined.
- * ndim: number of array dimensions
- * dim[]: array axis lengths
+ * On entry, *srcptr points to the string to parse (it must point to a '{').
+ * On successful return, it is advanced to point past the closing '}'.
+ *
+ * If dimensions were specified explicitly, they are passed in *ndim_p and
+ * dim. This function will check that the array values match the specified
+ * dimensions. If *ndim_p == 0, this function deduces the dimensions from
+ * the structure of the input, and stores them in *ndim_p and the dim array.
+ *
+ * Element type information:
* inputproc: type-specific input procedure for element datatype.
* typioparam, typmod: auxiliary values to pass to inputproc.
* typdelim: the value delimiter (type-specific).
* typlen, typbyval, typalign: storage parameters of element datatype.
*
* Outputs:
- * values[]: filled with converted data values.
- * nulls[]: filled with is-null markers.
- * *hasnulls: set true iff there are any null elements.
- * *nbytes: set to total size of data area needed (including alignment
- * padding but not including array header overhead).
- * *escontext: if this points to an ErrorSaveContext, details of
- * any error are reported there.
+ * *ndim_p, dim: dimensions deduced from the input structure.
+ * *nitems_p: total number of elements.
+ * *values_p[]: palloc'd array, filled with converted data values.
+ * *nulls_p[]: palloc'd array, filled with is-null markers.
+ *
+ * 'origStr' is the original input string, used only in error messages.
+ * If *escontext points to an ErrorSaveContext, details of any error are
+ * reported there.
*
* Result:
* true for success, false for failure (if escontext is provided).
- *
- * Note that values[] and nulls[] are allocated by the caller, and must have
- * nitems elements.
*/
static bool
-ReadArrayStr(char *arrayStr,
- const char *origStr,
- int nitems,
- int ndim,
- int *dim,
+ReadArrayStr(char **srcptr,
FmgrInfo *inputproc,
Oid typioparam,
int32 typmod,
@@ -727,224 +579,351 @@ ReadArrayStr(char *arrayStr,
int typlen,
bool typbyval,
char typalign,
- Datum *values,
- bool *nulls,
- bool *hasnulls,
- int32 *nbytes,
+ int *ndim_p,
+ int *dim,
+ int *nitems_p,
+ Datum **values_p,
+ bool **nulls_p,
+ const char *origStr,
Node *escontext)
{
- int i,
- nest_level = 0;
- char *srcptr;
- bool in_quotes = false;
- bool eoArray = false;
- bool hasnull;
- int32 totbytes;
- int indx[MAXDIM] = {0},
- prod[MAXDIM];
+ int ndim = *ndim_p;
+ bool dimensions_specified = (ndim != 0);
+ int maxitems;
+ Datum *values;
+ bool *nulls;
+ StringInfoData elembuf;
+ int nest_level;
+ int nitems;
+ bool ndim_frozen;
+ bool expect_delim;
+ int nelems[MAXDIM];
+
+ /* Allocate some starting output workspace; we'll enlarge as needed */
+ maxitems = 16;
+ values = palloc_array(Datum, maxitems);
+ nulls = palloc_array(bool, maxitems);
+
+ /* Allocate workspace to hold (string representation of) one element */
+ initStringInfo(&elembuf);
+
+ /* Loop below assumes first token is ATOK_LEVEL_START */
+ Assert(**srcptr == '{');
+
+ /* Parse tokens until we reach the matching right brace */
+ nest_level = 0;
+ nitems = 0;
+ ndim_frozen = dimensions_specified;
+ expect_delim = false;
+ do
+ {
+ ArrayToken tok;
- mda_get_prod(ndim, dim, prod);
+ tok = ReadArrayToken(srcptr, &elembuf, typdelim, origStr, escontext);
- /* Initialize is-null markers to true */
- memset(nulls, true, nitems * sizeof(bool));
+ switch (tok)
+ {
+ case ATOK_LEVEL_START:
+ /* Can't write left brace where delim is expected */
+ if (expect_delim)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected \"%c\" character.", '{')));
- /*
- * We have to remove " and \ characters to create a clean item value to
- * pass to the datatype input routine. We overwrite each item value
- * in-place within arrayStr to do this. srcptr is the current scan point,
- * and dstptr is where we are copying to.
- *
- * We also want to suppress leading and trailing unquoted whitespace. We
- * use the leadingspace flag to suppress leading space. Trailing space is
- * tracked by using dstendptr to point to the last significant output
- * character.
- *
- * The error checking in this routine is mostly pro-forma, since we expect
- * that ArrayCount() already validated the string. So we don't bother
- * with errdetail messages.
- */
- srcptr = arrayStr;
- while (!eoArray)
- {
- bool itemdone = false;
- bool leadingspace = true;
- bool hasquoting = false;
- char *itemstart;
- char *dstptr;
- char *dstendptr;
+ /* Initialize element counting in the new level */
+ if (nest_level >= MAXDIM)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("number of array dimensions (%d) exceeds the maximum allowed (%d)",
+ nest_level + 1, MAXDIM)));
- i = -1;
- itemstart = dstptr = dstendptr = srcptr;
+ nelems[nest_level] = 0;
+ nest_level++;
+ if (nest_level > ndim)
+ {
+ /* Can't increase ndim once it's frozen */
+ if (ndim_frozen)
+ goto dimension_error;
+ ndim = nest_level;
+ }
+ break;
- while (!itemdone)
- {
- switch (*srcptr)
- {
- case '\0':
- /* Signal a premature end of the string */
+ case ATOK_LEVEL_END:
+ /* Can't get here with nest_level == 0 */
+ Assert(nest_level > 0);
+
+ /*
+ * We allow a right brace to terminate an empty sub-array,
+ * otherwise it must occur where we expect a delimiter.
+ */
+ if (nelems[nest_level - 1] > 0 && !expect_delim)
ereturn(escontext, false,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"",
- origStr)));
- break;
- case '\\':
- /* Skip backslash, copy next character as-is. */
- srcptr++;
- if (*srcptr == '\0')
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected \"%c\" character.",
+ '}')));
+ nest_level--;
+ /* Nested sub-arrays count as elements of outer level */
+ if (nest_level > 0)
+ nelems[nest_level - 1]++;
+
+ /*
+ * Note: if we had dimensionality info, then dim[nest_level]
+ * is initially non-negative, and we'll check each sub-array's
+ * length against that.
+ */
+ if (dim[nest_level] < 0)
+ {
+ /* Save length of first sub-array of this level */
+ dim[nest_level] = nelems[nest_level];
+ }
+ else if (nelems[nest_level] != dim[nest_level])
+ {
+ /* Subsequent sub-arrays must have same length */
+ goto dimension_error;
+ }
+
+ /*
+ * Must have a delim or another right brace following, unless
+ * we have reached nest_level 0, where this won't matter.
+ */
+ expect_delim = true;
+ break;
+
+ case ATOK_DELIM:
+ if (!expect_delim)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected \"%c\" character.",
+ typdelim)));
+ expect_delim = false;
+ break;
+
+ case ATOK_ELEM:
+ case ATOK_ELEM_NULL:
+ /* Can't get here with nest_level == 0 */
+ Assert(nest_level > 0);
+
+ /* Disallow consecutive ELEM tokens */
+ if (expect_delim)
+ ereturn(escontext, false,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected array element.")));
+
+ /* Enlarge the values/nulls arrays if needed */
+ if (nitems >= maxitems)
+ {
+ if (maxitems >= MaxArraySize)
ereturn(escontext, false,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"",
- origStr)));
- *dstptr++ = *srcptr++;
- /* Treat the escaped character as non-whitespace */
- leadingspace = false;
- dstendptr = dstptr;
- hasquoting = true; /* can't be a NULL marker */
- break;
- case '"':
- in_quotes = !in_quotes;
- if (in_quotes)
- leadingspace = false;
- else
- {
- /*
- * Advance dstendptr when we exit in_quotes; this
- * saves having to do it in all the other in_quotes
- * cases.
- */
- dstendptr = dstptr;
- }
- hasquoting = true; /* can't be a NULL marker */
- srcptr++;
- break;
- case '{':
- if (!in_quotes)
- {
- if (nest_level >= ndim)
- ereturn(escontext, false,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"",
- origStr)));
- nest_level++;
- indx[nest_level - 1] = 0;
- srcptr++;
- }
- else
- *dstptr++ = *srcptr++;
- break;
- case '}':
- if (!in_quotes)
- {
- if (nest_level == 0)
- ereturn(escontext, false,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"",
- origStr)));
- if (i == -1)
- i = ArrayGetOffset0(ndim, indx, prod);
- indx[nest_level - 1] = 0;
- nest_level--;
- if (nest_level == 0)
- eoArray = itemdone = true;
- else
- indx[nest_level - 1]++;
- srcptr++;
- }
- else
- *dstptr++ = *srcptr++;
- break;
- default:
- if (in_quotes)
- *dstptr++ = *srcptr++;
- else if (*srcptr == typdelim)
- {
- if (i == -1)
- i = ArrayGetOffset0(ndim, indx, prod);
- itemdone = true;
- indx[ndim - 1]++;
- srcptr++;
- }
- else if (scanner_isspace(*srcptr))
- {
- /*
- * If leading space, drop it immediately. Else, copy
- * but don't advance dstendptr.
- */
- if (leadingspace)
- srcptr++;
- else
- *dstptr++ = *srcptr++;
- }
- else
- {
- *dstptr++ = *srcptr++;
- leadingspace = false;
- dstendptr = dstptr;
- }
- break;
- }
+ (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
+ errmsg("array size exceeds the maximum allowed (%d)",
+ (int) MaxArraySize)));
+ maxitems = Min(maxitems * 2, MaxArraySize);
+ values = repalloc_array(values, Datum, maxitems);
+ nulls = repalloc_array(nulls, bool, maxitems);
+ }
+
+ /* Read the element's value, or check that NULL is allowed */
+ if (!InputFunctionCallSafe(inputproc,
+ (tok == ATOK_ELEM_NULL) ? NULL : elembuf.data,
+ typioparam, typmod,
+ escontext,
+ &values[nitems]))
+ return false;
+ nulls[nitems] = (tok == ATOK_ELEM_NULL);
+ nitems++;
+
+ /*
+ * Once we have found an element, the number of dimensions can
+ * no longer increase, and subsequent elements must all be at
+ * the same nesting depth.
+ */
+ ndim_frozen = true;
+ if (nest_level != ndim)
+ goto dimension_error;
+ /* Count the new element */
+ nelems[nest_level - 1]++;
+
+ /* Must have a delim or a right brace following */
+ expect_delim = true;
+ break;
+
+ case ATOK_ERROR:
+ return false;
}
+ } while (nest_level > 0);
- Assert(dstptr < srcptr);
- *dstendptr = '\0';
+ /* Clean up and return results */
+ pfree(elembuf.data);
- if (i < 0 || i >= nitems)
- ereturn(escontext, false,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("malformed array literal: \"%s\"",
- origStr)));
+ *ndim_p = ndim;
+ *nitems_p = nitems;
+ *values_p = values;
+ *nulls_p = nulls;
+ return true;
+
+dimension_error:
+ if (dimensions_specified)
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Specified array dimensions do not match array contents.")));
+ else
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Multidimensional arrays must have sub-arrays with matching dimensions.")));
+ return false;
+}
+
+/*
+ * ReadArrayToken
+ * read one token from an array value string
+ *
+ * Starts scanning from *srcptr. On non-error return, *srcptr is
+ * advanced past the token.
+ *
+ * If the token is ATOK_ELEM, the de-escaped string is returned in elembuf.
+ */
+static ArrayToken
+ReadArrayToken(char **srcptr, StringInfo elembuf, char typdelim,
+ const char *origStr, Node *escontext)
+{
+ char *p = *srcptr;
+ int dstlen;
+ bool has_escapes;
+
+ resetStringInfo(elembuf);
- if (Array_nulls && !hasquoting &&
- pg_strcasecmp(itemstart, "NULL") == 0)
+ /* Identify token type. Loop advances over leading whitespace. */
+ for (;;)
+ {
+ switch (*p)
{
- /* it's a NULL item */
- if (!InputFunctionCallSafe(inputproc, NULL,
- typioparam, typmod,
- escontext,
- &values[i]))
- return false;
- nulls[i] = true;
+ case '\0':
+ goto ending_error;
+ case '{':
+ *srcptr = p + 1;
+ return ATOK_LEVEL_START;
+ case '}':
+ *srcptr = p + 1;
+ return ATOK_LEVEL_END;
+ case '"':
+ p++;
+ goto quoted_element;
+ default:
+ if (*p == typdelim)
+ {
+ *srcptr = p + 1;
+ return ATOK_DELIM;
+ }
+ if (scanner_isspace(*p))
+ {
+ p++;
+ continue;
+ }
+ goto unquoted_element;
}
- else
+ }
+
+quoted_element:
+ for (;;)
+ {
+ switch (*p)
{
- if (!InputFunctionCallSafe(inputproc, itemstart,
- typioparam, typmod,
- escontext,
- &values[i]))
- return false;
- nulls[i] = false;
+ case '\0':
+ goto ending_error;
+ case '\\':
+ /* Skip backslash, copy next character as-is. */
+ p++;
+ if (*p == '\0')
+ goto ending_error;
+ appendStringInfoChar(elembuf, *p++);
+ break;
+ case '"':
+ *srcptr = p + 1;
+ return ATOK_ELEM;
+ default:
+ appendStringInfoChar(elembuf, *p++);
+ break;
}
}
+unquoted_element:
+
/*
- * Check for nulls, compute total data space needed
+ * We don't include trailing whitespace in the result. dstlen tracks how
+ * much of the output string is known to not be trailing whitespace.
*/
- hasnull = false;
- totbytes = 0;
- for (i = 0; i < nitems; i++)
+ dstlen = 0;
+ has_escapes = false;
+ for (;;)
{
- if (nulls[i])
- hasnull = true;
- else
+ switch (*p)
{
- /* let's just make sure data is not toasted */
- if (typlen == -1)
- values[i] = PointerGetDatum(PG_DETOAST_DATUM(values[i]));
- totbytes = att_addlength_datum(totbytes, typlen, values[i]);
- totbytes = att_align_nominal(totbytes, typalign);
- /* check for overflow of total request */
- if (!AllocSizeIsValid(totbytes))
- ereturn(escontext, false,
- (errcode(ERRCODE_PROGRAM_LIMIT_EXCEEDED),
- errmsg("array size exceeds the maximum allowed (%d)",
- (int) MaxAllocSize)));
+ case '\0':
+ goto ending_error;
+ case '{':
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected \"%c\" character.",
+ '{')));
+ return ATOK_ERROR;
+ case '"':
+
+ /*
+ * XXX "Unexpected %c character" would be more apropos, but
+ * this message is what the pre-v17 implementation produced,
+ * so we'll keep it for now.
+ */
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected array element.")));
+ return ATOK_ERROR;
+ case '\\':
+ /* Skip backslash, copy next character as-is. */
+ p++;
+ if (*p == '\0')
+ goto ending_error;
+ appendStringInfoChar(elembuf, *p++);
+ dstlen = elembuf->len; /* treat it as non-whitespace */
+ has_escapes = true;
+ break;
+ default:
+ /* End of elem? */
+ if (*p == typdelim || *p == '}')
+ {
+ /* hack: truncate the output string to dstlen */
+ elembuf->data[dstlen] = '\0';
+ elembuf->len = dstlen;
+ *srcptr = p;
+ /* Check if it's unquoted "NULL" */
+ if (Array_nulls && !has_escapes &&
+ pg_strcasecmp(elembuf->data, "NULL") == 0)
+ return ATOK_ELEM_NULL;
+ else
+ return ATOK_ELEM;
+ }
+ appendStringInfoChar(elembuf, *p);
+ if (!scanner_isspace(*p))
+ dstlen = elembuf->len;
+ p++;
+ break;
}
}
- *hasnulls = hasnull;
- *nbytes = totbytes;
- return true;
-}
+ending_error:
+ errsave(escontext,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("malformed array literal: \"%s\"", origStr),
+ errdetail("Unexpected end of input.")));
+ return ATOK_ERROR;
+}
/*
* Copy data into an array object from a temporary array of Datums.
diff --git a/src/backend/utils/adt/arrayutils.c b/src/backend/utils/adt/arrayutils.c
index aed799234c..6243971543 100644
--- a/src/backend/utils/adt/arrayutils.c
+++ b/src/backend/utils/adt/arrayutils.c
@@ -43,21 +43,6 @@ ArrayGetOffset(int n, const int *dim, const int *lb, const int *indx)
return offset;
}
-/*
- * Same, but subscripts are assumed 0-based, and use a scale array
- * instead of raw dimension data (see mda_get_prod to create scale array)
- */
-int
-ArrayGetOffset0(int n, const int *tup, const int *scale)
-{
- int i,
- lin = 0;
-
- for (i = 0; i < n; i++)
- lin += tup[i] * scale[i];
- return lin;
-}
-
/*
* Convert array dimensions into number of elements
*
diff --git a/src/include/utils/array.h b/src/include/utils/array.h
index e6c8d88d9f..7308007185 100644
--- a/src/include/utils/array.h
+++ b/src/include/utils/array.h
@@ -455,7 +455,6 @@ extern void array_free_iterator(ArrayIterator iterator);
*/
extern int ArrayGetOffset(int n, const int *dim, const int *lb, const int *indx);
-extern int ArrayGetOffset0(int n, const int *tup, const int *scale);
extern int ArrayGetNItems(int ndim, const int *dims);
extern int ArrayGetNItemsSafe(int ndim, const int *dims,
struct Node *escontext);
diff --git a/src/test/regress/expected/arrays.out b/src/test/regress/expected/arrays.out
index 957498432d..2212851538 100644
--- a/src/test/regress/expected/arrays.out
+++ b/src/test/regress/expected/arrays.out
@@ -1492,17 +1492,22 @@ select '{{1,{2}},{2,3}}'::text[];
ERROR: malformed array literal: "{{1,{2}},{2,3}}"
LINE 1: select '{{1,{2}},{2,3}}'::text[];
^
-DETAIL: Unexpected "{" character.
-select '{{},{}}'::text[];
-ERROR: malformed array literal: "{{},{}}"
-LINE 1: select '{{},{}}'::text[];
- ^
-DETAIL: Unexpected "}" character.
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
select E'{{1,2},\\{2,3}}'::text[];
ERROR: malformed array literal: "{{1,2},\{2,3}}"
LINE 1: select E'{{1,2},\\{2,3}}'::text[];
^
-DETAIL: Unexpected "\" character.
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select E'{"a"a}'::text[];
+ERROR: malformed array literal: "{"a"a}"
+LINE 1: select E'{"a"a}'::text[];
+ ^
+DETAIL: Unexpected array element.
+select E'{a"a"}'::text[];
+ERROR: malformed array literal: "{a"a"}"
+LINE 1: select E'{a"a"}'::text[];
+ ^
+DETAIL: Unexpected array element.
select '{{"1 2" x},{3}}'::text[];
ERROR: malformed array literal: "{{"1 2" x},{3}}"
LINE 1: select '{{"1 2" x},{3}}'::text[];
@@ -1518,11 +1523,116 @@ ERROR: malformed array literal: "{ }}"
LINE 1: select '{ }}'::text[];
^
DETAIL: Junk after closing right brace.
+select '}{'::text[];
+ERROR: malformed array literal: "}{"
+LINE 1: select '}{'::text[];
+ ^
+DETAIL: Array value must start with "{" or dimension information.
+select '{foo{}}'::text[];
+ERROR: malformed array literal: "{foo{}}"
+LINE 1: select '{foo{}}'::text[];
+ ^
+DETAIL: Unexpected "{" character.
+select '{"foo"{}}'::text[];
+ERROR: malformed array literal: "{"foo"{}}"
+LINE 1: select '{"foo"{}}'::text[];
+ ^
+DETAIL: Unexpected "{" character.
+select '{foo,,bar}'::text[];
+ERROR: malformed array literal: "{foo,,bar}"
+LINE 1: select '{foo,,bar}'::text[];
+ ^
+DETAIL: Unexpected "," character.
+select '{{1},{{2}}}'::text[];
+ERROR: malformed array literal: "{{1},{{2}}}"
+LINE 1: select '{{1},{{2}}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '{{{1}},{2}}'::text[];
+ERROR: malformed array literal: "{{{1}},{2}}"
+LINE 1: select '{{{1}},{2}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '{{},{{}}}'::text[];
+ERROR: malformed array literal: "{{},{{}}}"
+LINE 1: select '{{},{{}}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '{{{}},{}}'::text[];
+ERROR: malformed array literal: "{{{}},{}}"
+LINE 1: select '{{{}},{}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '{{1},{}}'::text[];
+ERROR: malformed array literal: "{{1},{}}"
+LINE 1: select '{{1},{}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '{{},{1}}'::text[];
+ERROR: malformed array literal: "{{},{1}}"
+LINE 1: select '{{},{1}}'::text[];
+ ^
+DETAIL: Multidimensional arrays must have sub-arrays with matching dimensions.
+select '[1:0]={}'::int[];
+ERROR: upper bound cannot be less than lower bound
+LINE 1: select '[1:0]={}'::int[];
+ ^
+select '[2147483646:2147483647]={1,2}'::int[];
+ERROR: array upper bound is too large: 2147483647
+LINE 1: select '[2147483646:2147483647]={1,2}'::int[];
+ ^
+select '[1:-1]={}'::int[];
+ERROR: upper bound cannot be less than lower bound
+LINE 1: select '[1:-1]={}'::int[];
+ ^
+select '[2]={1}'::int[];
+ERROR: malformed array literal: "[2]={1}"
+LINE 1: select '[2]={1}'::int[];
+ ^
+DETAIL: Specified array dimensions do not match array contents.
+select '[1:]={1}'::int[];
+ERROR: malformed array literal: "[1:]={1}"
+LINE 1: select '[1:]={1}'::int[];
+ ^
+DETAIL: Missing array dimension value.
+select '[:1]={1}'::int[];
+ERROR: malformed array literal: "[:1]={1}"
+LINE 1: select '[:1]={1}'::int[];
+ ^
+DETAIL: "[" must introduce explicitly-specified array dimensions.
select array[];
ERROR: cannot determine type of empty array
LINE 1: select array[];
^
HINT: Explicitly cast to the desired type, for example ARRAY[]::integer[].
+select '{{1,},{1},}'::text[];
+ERROR: malformed array literal: "{{1,},{1},}"
+LINE 1: select '{{1,},{1},}'::text[];
+ ^
+DETAIL: Unexpected "}" character.
+select '{{1,},{1}}'::text[];
+ERROR: malformed array literal: "{{1,},{1}}"
+LINE 1: select '{{1,},{1}}'::text[];
+ ^
+DETAIL: Unexpected "}" character.
+select '{{1,}}'::text[];
+ERROR: malformed array literal: "{{1,}}"
+LINE 1: select '{{1,}}'::text[];
+ ^
+DETAIL: Unexpected "}" character.
+select '{1,}'::text[];
+ERROR: malformed array literal: "{1,}"
+LINE 1: select '{1,}'::text[];
+ ^
+DETAIL: Unexpected "}" character.
+select '[21474836488:21474836489]={1,2}'::int[];
+ERROR: array bound is out of integer range
+LINE 1: select '[21474836488:21474836489]={1,2}'::int[];
+ ^
+select '[-2147483649:-2147483648]={1,2}'::int[];
+ERROR: array bound is out of integer range
+LINE 1: select '[-2147483649:-2147483648]={1,2}'::int[];
+ ^
-- none of the above should be accepted
-- all of the following should be accepted
select '{}'::text[];
@@ -1531,12 +1641,30 @@ select '{}'::text[];
{}
(1 row)
+select '{{},{}}'::text[];
+ text
+------
+ {}
+(1 row)
+
select '{{{1,2,3,4},{2,3,4,5}},{{3,4,5,6},{4,5,6,7}}}'::text[];
text
-----------------------------------------------
{{{1,2,3,4},{2,3,4,5}},{{3,4,5,6},{4,5,6,7}}}
(1 row)
+select '{null,n\ull,"null"}'::text[];
+ text
+----------------------
+ {NULL,"null","null"}
+(1 row)
+
+select '{ ab\c , "ab\"c" }'::text[];
+ text
+---------------
+ {abc,"ab\"c"}
+(1 row)
+
select '{0 second ,0 second}'::interval[];
interval
---------------
@@ -1570,12 +1698,30 @@ select array[]::text[];
{}
(1 row)
+select '[2]={1,7}'::int[];
+ int4
+-------
+ {1,7}
+(1 row)
+
select '[0:1]={1.1,2.2}'::float8[];
float8
-----------------
[0:1]={1.1,2.2}
(1 row)
+select '[2147483646:2147483646]={1}'::int[];
+ int4
+-----------------------------
+ [2147483646:2147483646]={1}
+(1 row)
+
+select '[-2147483648:-2147483647]={1,2}'::int[];
+ int4
+---------------------------------
+ [-2147483648:-2147483647]={1,2}
+(1 row)
+
-- all of the above should be accepted
-- tests for array aggregates
CREATE TEMP TABLE arraggtest ( f1 INT[], f2 TEXT[][], f3 FLOAT[]);
diff --git a/src/test/regress/sql/arrays.sql b/src/test/regress/sql/arrays.sql
index daf805c382..8181e172a9 100644
--- a/src/test/regress/sql/arrays.sql
+++ b/src/test/regress/sql/arrays.sql
@@ -473,17 +473,43 @@ select 'foo' ilike all (array['F%', '%O']); -- t
-- none of the following should be accepted
select '{{1,{2}},{2,3}}'::text[];
-select '{{},{}}'::text[];
select E'{{1,2},\\{2,3}}'::text[];
+select E'{"a"a}'::text[];
+select E'{a"a"}'::text[];
select '{{"1 2" x},{3}}'::text[];
select '{}}'::text[];
select '{ }}'::text[];
+select '}{'::text[];
+select '{foo{}}'::text[];
+select '{"foo"{}}'::text[];
+select '{foo,,bar}'::text[];
+select '{{1},{{2}}}'::text[];
+select '{{{1}},{2}}'::text[];
+select '{{},{{}}}'::text[];
+select '{{{}},{}}'::text[];
+select '{{1},{}}'::text[];
+select '{{},{1}}'::text[];
+select '[1:0]={}'::int[];
+select '[2147483646:2147483647]={1,2}'::int[];
+select '[1:-1]={}'::int[];
+select '[2]={1}'::int[];
+select '[1:]={1}'::int[];
+select '[:1]={1}'::int[];
select array[];
+select '{{1,},{1},}'::text[];
+select '{{1,},{1}}'::text[];
+select '{{1,}}'::text[];
+select '{1,}'::text[];
+select '[21474836488:21474836489]={1,2}'::int[];
+select '[-2147483649:-2147483648]={1,2}'::int[];
-- none of the above should be accepted
-- all of the following should be accepted
select '{}'::text[];
+select '{{},{}}'::text[];
select '{{{1,2,3,4},{2,3,4,5}},{{3,4,5,6},{4,5,6,7}}}'::text[];
+select '{null,n\ull,"null"}'::text[];
+select '{ ab\c , "ab\"c" }'::text[];
select '{0 second ,0 second}'::interval[];
select '{ { "," } , { 3 } }'::text[];
select ' { { " 0 second " , 0 second } }'::text[];
@@ -492,7 +518,10 @@ select '{
@ 1 hour @ 42 minutes @ 20 seconds
}'::interval[];
select array[]::text[];
+select '[2]={1,7}'::int[];
select '[0:1]={1.1,2.2}'::float8[];
+select '[2147483646:2147483646]={1}'::int[];
+select '[-2147483648:-2147483647]={1,2}'::int[];
-- all of the above should be accepted
-- tests for array aggregates
diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list
index 87c1aee379..5c4c1e1600 100644
--- a/src/tools/pgindent/typedefs.list
+++ b/src/tools/pgindent/typedefs.list
@@ -148,8 +148,8 @@ ArrayIOData
ArrayIterator
ArrayMapState
ArrayMetaState
-ArrayParseState
ArraySubWorkspace
+ArrayToken
ArrayType
AsyncQueueControl
AsyncQueueEntry
pgsql-hackers by date: