* json.c
* JSON data type support.
*
- * Portions Copyright (c) 1996-2012, PostgreSQL Global Development Group
+ * Portions Copyright (c) 1996-2019, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* IDENTIFICATION
*/
#include "postgres.h"
+#include "access/htup_details.h"
+#include "access/transam.h"
#include "catalog/pg_type.h"
#include "executor/spi.h"
+#include "funcapi.h"
#include "lib/stringinfo.h"
#include "libpq/pqformat.h"
#include "mb/pg_wchar.h"
+#include "miscadmin.h"
#include "parser/parse_coerce.h"
#include "utils/array.h"
#include "utils/builtins.h"
+#include "utils/date.h"
+#include "utils/datetime.h"
#include "utils/lsyscache.h"
#include "utils/json.h"
+#include "utils/jsonapi.h"
#include "utils/typcache.h"
+#include "utils/syscache.h"
-typedef enum
-{
- JSON_VALUE_INVALID,
- JSON_VALUE_STRING,
- JSON_VALUE_NUMBER,
- JSON_VALUE_OBJECT,
- JSON_VALUE_ARRAY,
- JSON_VALUE_TRUE,
- JSON_VALUE_FALSE,
- JSON_VALUE_NULL
-} JsonValueType;
-
-typedef struct
-{
- char *input;
- char *token_start;
- char *token_terminator;
- JsonValueType token_type;
- int line_number;
- char *line_start;
-} JsonLexContext;
-
-typedef enum
+/*
+ * The context of the parser is maintained by the recursive descent
+ * mechanism, but is passed explicitly to the error reporting routine
+ * for better diagnostics.
+ */
+typedef enum /* contexts of JSON parser */
{
JSON_PARSE_VALUE, /* expecting a value */
+ JSON_PARSE_STRING, /* expecting a string (for a field name) */
JSON_PARSE_ARRAY_START, /* saw '[', expecting value or ']' */
JSON_PARSE_ARRAY_NEXT, /* saw array element, expecting ',' or ']' */
JSON_PARSE_OBJECT_START, /* saw '{', expecting label or '}' */
JSON_PARSE_OBJECT_LABEL, /* saw object label, expecting ':' */
JSON_PARSE_OBJECT_NEXT, /* saw object value, expecting ',' or '}' */
- JSON_PARSE_OBJECT_COMMA /* saw object ',', expecting next label */
-} JsonParseState;
+ JSON_PARSE_OBJECT_COMMA, /* saw object ',', expecting next label */
+ JSON_PARSE_END /* saw the end of a document, expect nothing */
+} JsonParseContext;
-typedef struct JsonParseStack
+typedef enum /* type categories for datum_to_json */
{
- JsonParseState state;
-} JsonParseStack;
-
-typedef enum
+ JSONTYPE_NULL, /* null, so we didn't bother to identify */
+ JSONTYPE_BOOL, /* boolean (built-in types only) */
+ JSONTYPE_NUMERIC, /* numeric (ditto) */
+ JSONTYPE_DATE, /* we use special formatting for datetimes */
+ JSONTYPE_TIMESTAMP,
+ JSONTYPE_TIMESTAMPTZ,
+ JSONTYPE_JSON, /* JSON itself (and JSONB) */
+ JSONTYPE_ARRAY, /* array */
+ JSONTYPE_COMPOSITE, /* composite */
+ JSONTYPE_CAST, /* something with an explicit cast to JSON */
+ JSONTYPE_OTHER /* all else */
+} JsonTypeCategory;
+
+typedef struct JsonAggState
{
- JSON_STACKOP_NONE,
- JSON_STACKOP_PUSH,
- JSON_STACKOP_PUSH_WITH_PUSHBACK,
- JSON_STACKOP_POP
-} JsonStackOp;
-
-static void json_validate_cstring(char *input);
-static void json_lex(JsonLexContext *lex);
-static void json_lex_string(JsonLexContext *lex);
-static void json_lex_number(JsonLexContext *lex, char *s);
-static void report_parse_error(JsonParseStack *stack, JsonLexContext *lex);
-static void report_invalid_token(JsonLexContext *lex);
+ StringInfo str;
+ JsonTypeCategory key_category;
+ Oid key_output_func;
+ JsonTypeCategory val_category;
+ Oid val_output_func;
+} JsonAggState;
+
+static inline void json_lex(JsonLexContext *lex);
+static inline void json_lex_string(JsonLexContext *lex);
+static inline void json_lex_number(JsonLexContext *lex, char *s,
+ bool *num_err, int *total_len);
+static inline void parse_scalar(JsonLexContext *lex, JsonSemAction *sem);
+static void parse_object_field(JsonLexContext *lex, JsonSemAction *sem);
+static void parse_object(JsonLexContext *lex, JsonSemAction *sem);
+static void parse_array_element(JsonLexContext *lex, JsonSemAction *sem);
+static void parse_array(JsonLexContext *lex, JsonSemAction *sem);
+static void report_parse_error(JsonParseContext ctx, JsonLexContext *lex) pg_attribute_noreturn();
+static void report_invalid_token(JsonLexContext *lex) pg_attribute_noreturn();
+static int report_json_context(JsonLexContext *lex);
static char *extract_mb_char(char *s);
-static void composite_to_json(Datum composite, StringInfo result, bool use_line_feeds);
+static void composite_to_json(Datum composite, StringInfo result,
+ bool use_line_feeds);
static void array_dim_to_json(StringInfo result, int dim, int ndims, int *dims,
- Datum *vals, bool *nulls, int *valcount,
- TYPCATEGORY tcategory, Oid typoutputfunc,
- bool use_line_feeds);
-static void array_to_json_internal(Datum array, StringInfo result, bool use_line_feeds);
-
-/* fake type category for JSON so we can distinguish it in datum_to_json */
-#define TYPCATEGORY_JSON 'j'
-/* letters appearing in numeric output that aren't valid in a JSON number */
-#define NON_NUMERIC_LETTER "NnAaIiFfTtYy"
+ Datum *vals, bool *nulls, int *valcount,
+ JsonTypeCategory tcategory, Oid outfuncoid,
+ bool use_line_feeds);
+static void array_to_json_internal(Datum array, StringInfo result,
+ bool use_line_feeds);
+static void json_categorize_type(Oid typoid,
+ JsonTypeCategory *tcategory,
+ Oid *outfuncoid);
+static void datum_to_json(Datum val, bool is_null, StringInfo result,
+ JsonTypeCategory tcategory, Oid outfuncoid,
+ bool key_scalar);
+static void add_json(Datum val, bool is_null, StringInfo result,
+ Oid val_type, bool key_scalar);
+static text *catenate_stringinfo_string(StringInfo buffer, const char *addon);
+
+/* the null action object used for pure validation */
+static JsonSemAction nullSemAction =
+{
+ NULL, NULL, NULL, NULL, NULL,
+ NULL, NULL, NULL, NULL, NULL
+};
+
+/* Recursive Descent parser support routines */
+
+/*
+ * lex_peek
+ *
+ * what is the current look_ahead token?
+*/
+static inline JsonTokenType
+lex_peek(JsonLexContext *lex)
+{
+ return lex->token_type;
+}
+
+/*
+ * lex_accept
+ *
+ * accept the look_ahead token and move the lexer to the next token if the
+ * look_ahead token matches the token parameter. In that case, and if required,
+ * also hand back the de-escaped lexeme.
+ *
+ * returns true if the token matched, false otherwise.
+ */
+static inline bool
+lex_accept(JsonLexContext *lex, JsonTokenType token, char **lexeme)
+{
+ if (lex->token_type == token)
+ {
+ if (lexeme != NULL)
+ {
+ if (lex->token_type == JSON_TOKEN_STRING)
+ {
+ if (lex->strval != NULL)
+ *lexeme = pstrdup(lex->strval->data);
+ }
+ else
+ {
+ int len = (lex->token_terminator - lex->token_start);
+ char *tokstr = palloc(len + 1);
+
+ memcpy(tokstr, lex->token_start, len);
+ tokstr[len] = '\0';
+ *lexeme = tokstr;
+ }
+ }
+ json_lex(lex);
+ return true;
+ }
+ return false;
+}
+
+/*
+ * lex_accept
+ *
+ * move the lexer to the next token if the current look_ahead token matches
+ * the parameter token. Otherwise, report an error.
+ */
+static inline void
+lex_expect(JsonParseContext ctx, JsonLexContext *lex, JsonTokenType token)
+{
+ if (!lex_accept(lex, token, NULL))
+ report_parse_error(ctx, lex);
+}
+
+/* chars to consider as part of an alphanumeric token */
+#define JSON_ALPHANUMERIC_CHAR(c) \
+ (((c) >= 'a' && (c) <= 'z') || \
+ ((c) >= 'A' && (c) <= 'Z') || \
+ ((c) >= '0' && (c) <= '9') || \
+ (c) == '_' || \
+ IS_HIGHBIT_SET(c))
+
+/*
+ * Utility function to check if a string is a valid JSON number.
+ *
+ * str is of length len, and need not be null-terminated.
+ */
+bool
+IsValidJsonNumber(const char *str, int len)
+{
+ bool numeric_error;
+ int total_len;
+ JsonLexContext dummy_lex;
+
+ if (len <= 0)
+ return false;
+
+ /*
+ * json_lex_number expects a leading '-' to have been eaten already.
+ *
+ * having to cast away the constness of str is ugly, but there's not much
+ * easy alternative.
+ */
+ if (*str == '-')
+ {
+ dummy_lex.input = unconstify(char *, str) +1;
+ dummy_lex.input_length = len - 1;
+ }
+ else
+ {
+ dummy_lex.input = unconstify(char *, str);
+ dummy_lex.input_length = len;
+ }
+
+ json_lex_number(&dummy_lex, dummy_lex.input, &numeric_error, &total_len);
+
+ return (!numeric_error) && (total_len == dummy_lex.input_length);
+}
+
/*
* Input.
*/
Datum
json_in(PG_FUNCTION_ARGS)
{
- char *text = PG_GETARG_CSTRING(0);
+ char *json = PG_GETARG_CSTRING(0);
+ text *result = cstring_to_text(json);
+ JsonLexContext *lex;
- json_validate_cstring(text);
+ /* validate it */
+ lex = makeJsonLexContext(result, false);
+ pg_parse_json(lex, &nullSemAction);
- PG_RETURN_TEXT_P(cstring_to_text(text));
+ /* Internal representation is the same as text, for now */
+ PG_RETURN_TEXT_P(result);
}
/*
Datum
json_out(PG_FUNCTION_ARGS)
{
+ /* we needn't detoast because text_to_cstring will handle that */
Datum txt = PG_GETARG_DATUM(0);
PG_RETURN_CSTRING(TextDatumGetCString(txt));
Datum
json_send(PG_FUNCTION_ARGS)
{
- StringInfoData buf;
text *t = PG_GETARG_TEXT_PP(0);
+ StringInfoData buf;
pq_begintypsend(&buf);
pq_sendtext(&buf, VARDATA_ANY(t), VARSIZE_ANY_EXHDR(t));
json_recv(PG_FUNCTION_ARGS)
{
StringInfo buf = (StringInfo) PG_GETARG_POINTER(0);
- text *result;
char *str;
int nbytes;
+ JsonLexContext *lex;
str = pq_getmsgtext(buf, buf->len - buf->cursor, &nbytes);
- /*
- * We need a null-terminated string to pass to json_validate_cstring().
- * Rather than make a separate copy, make the temporary result one byte
- * bigger than it needs to be.
- */
- result = palloc(nbytes + 1 + VARHDRSZ);
- SET_VARSIZE(result, nbytes + VARHDRSZ);
- memcpy(VARDATA(result), str, nbytes);
- str = VARDATA(result);
- str[nbytes] = '\0';
-
/* Validate it. */
- json_validate_cstring(str);
+ lex = makeJsonLexContextCstringLen(str, nbytes, false);
+ pg_parse_json(lex, &nullSemAction);
- PG_RETURN_TEXT_P(result);
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(str, nbytes));
}
/*
- * Check whether supplied input is valid JSON.
+ * makeJsonLexContext
+ *
+ * lex constructor, with or without StringInfo object
+ * for de-escaped lexemes.
+ *
+ * Without is better as it makes the processing faster, so only make one
+ * if really required.
+ *
+ * If you already have the json as a text* value, use the first of these
+ * functions, otherwise use makeJsonLexContextCstringLen().
*/
-static void
-json_validate_cstring(char *input)
+JsonLexContext *
+makeJsonLexContext(text *json, bool need_escapes)
{
- JsonLexContext lex;
- JsonParseStack *stack,
- *stacktop;
- int stacksize;
-
- /* Set up lexing context. */
- lex.input = input;
- lex.token_terminator = lex.input;
- lex.line_number = 1;
- lex.line_start = input;
-
- /* Set up parse stack. */
- stacksize = 32;
- stacktop = palloc(sizeof(JsonParseStack) * stacksize);
- stack = stacktop;
- stack->state = JSON_PARSE_VALUE;
-
- /* Main parsing loop. */
- for (;;)
- {
- JsonStackOp op;
+ return makeJsonLexContextCstringLen(VARDATA_ANY(json),
+ VARSIZE_ANY_EXHDR(json),
+ need_escapes);
+}
- /* Fetch next token. */
- json_lex(&lex);
+JsonLexContext *
+makeJsonLexContextCstringLen(char *json, int len, bool need_escapes)
+{
+ JsonLexContext *lex = palloc0(sizeof(JsonLexContext));
+
+ lex->input = lex->token_terminator = lex->line_start = json;
+ lex->line_number = 1;
+ lex->input_length = len;
+ if (need_escapes)
+ lex->strval = makeStringInfo();
+ return lex;
+}
- /* Check for unexpected end of input. */
- if (lex.token_start == NULL)
- report_parse_error(stack, &lex);
+/*
+ * pg_parse_json
+ *
+ * Publicly visible entry point for the JSON parser.
+ *
+ * lex is a lexing context, set up for the json to be processed by calling
+ * makeJsonLexContext(). sem is a structure of function pointers to semantic
+ * action routines to be called at appropriate spots during parsing, and a
+ * pointer to a state object to be passed to those routines.
+ */
+void
+pg_parse_json(JsonLexContext *lex, JsonSemAction *sem)
+{
+ JsonTokenType tok;
-redo:
- /* Figure out what to do with this token. */
- op = JSON_STACKOP_NONE;
- switch (stack->state)
- {
- case JSON_PARSE_VALUE:
- if (lex.token_type != JSON_VALUE_INVALID)
- op = JSON_STACKOP_POP;
- else if (lex.token_start[0] == '[')
- stack->state = JSON_PARSE_ARRAY_START;
- else if (lex.token_start[0] == '{')
- stack->state = JSON_PARSE_OBJECT_START;
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_ARRAY_START:
- if (lex.token_type != JSON_VALUE_INVALID)
- stack->state = JSON_PARSE_ARRAY_NEXT;
- else if (lex.token_start[0] == ']')
- op = JSON_STACKOP_POP;
- else if (lex.token_start[0] == '['
- || lex.token_start[0] == '{')
- {
- stack->state = JSON_PARSE_ARRAY_NEXT;
- op = JSON_STACKOP_PUSH_WITH_PUSHBACK;
- }
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_ARRAY_NEXT:
- if (lex.token_type != JSON_VALUE_INVALID)
- report_parse_error(stack, &lex);
- else if (lex.token_start[0] == ']')
- op = JSON_STACKOP_POP;
- else if (lex.token_start[0] == ',')
- op = JSON_STACKOP_PUSH;
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_OBJECT_START:
- if (lex.token_type == JSON_VALUE_STRING)
- stack->state = JSON_PARSE_OBJECT_LABEL;
- else if (lex.token_type == JSON_VALUE_INVALID
- && lex.token_start[0] == '}')
- op = JSON_STACKOP_POP;
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_OBJECT_LABEL:
- if (lex.token_type == JSON_VALUE_INVALID
- && lex.token_start[0] == ':')
- {
- stack->state = JSON_PARSE_OBJECT_NEXT;
- op = JSON_STACKOP_PUSH;
- }
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_OBJECT_NEXT:
- if (lex.token_type != JSON_VALUE_INVALID)
- report_parse_error(stack, &lex);
- else if (lex.token_start[0] == '}')
- op = JSON_STACKOP_POP;
- else if (lex.token_start[0] == ',')
- stack->state = JSON_PARSE_OBJECT_COMMA;
- else
- report_parse_error(stack, &lex);
- break;
- case JSON_PARSE_OBJECT_COMMA:
- if (lex.token_type == JSON_VALUE_STRING)
- stack->state = JSON_PARSE_OBJECT_LABEL;
- else
- report_parse_error(stack, &lex);
- break;
- default:
- elog(ERROR, "unexpected json parse state: %d",
- (int) stack->state);
- }
+ /* get the initial token */
+ json_lex(lex);
- /* Push or pop the stack, if needed. */
- switch (op)
- {
- case JSON_STACKOP_PUSH:
- case JSON_STACKOP_PUSH_WITH_PUSHBACK:
- ++stack;
- if (stack >= &stacktop[stacksize])
- {
- int stackoffset = stack - stacktop;
+ tok = lex_peek(lex);
- stacksize = stacksize + 32;
- stacktop = repalloc(stacktop,
- sizeof(JsonParseStack) * stacksize);
- stack = stacktop + stackoffset;
- }
- stack->state = JSON_PARSE_VALUE;
- if (op == JSON_STACKOP_PUSH_WITH_PUSHBACK)
- goto redo;
- break;
- case JSON_STACKOP_POP:
- if (stack == stacktop)
- {
- /* Expect end of input. */
- json_lex(&lex);
- if (lex.token_start != NULL)
- report_parse_error(NULL, &lex);
- return;
- }
- --stack;
- break;
- case JSON_STACKOP_NONE:
- /* nothing to do */
- break;
- }
+ /* parse by recursive descent */
+ switch (tok)
+ {
+ case JSON_TOKEN_OBJECT_START:
+ parse_object(lex, sem);
+ break;
+ case JSON_TOKEN_ARRAY_START:
+ parse_array(lex, sem);
+ break;
+ default:
+ parse_scalar(lex, sem); /* json can be a bare scalar */
}
+
+ lex_expect(JSON_PARSE_END, lex, JSON_TOKEN_END);
+
}
/*
- * Lex one token from the input stream.
+ * json_count_array_elements
+ *
+ * Returns number of array elements in lex context at start of array token
+ * until end of array token at same nesting level.
+ *
+ * Designed to be called from array_start routines.
*/
-static void
-json_lex(JsonLexContext *lex)
+int
+json_count_array_elements(JsonLexContext *lex)
{
- char *s;
+ JsonLexContext copylex;
+ int count;
- /* Skip leading whitespace. */
- s = lex->token_terminator;
- while (*s == ' ' || *s == '\t' || *s == '\n' || *s == '\r')
- {
- if (*s == '\n')
- ++lex->line_number;
- ++s;
- }
- lex->token_start = s;
+ /*
+ * It's safe to do this with a shallow copy because the lexical routines
+ * don't scribble on the input. They do scribble on the other pointers
+ * etc, so doing this with a copy makes that safe.
+ */
+ memcpy(©lex, lex, sizeof(JsonLexContext));
+ copylex.strval = NULL; /* not interested in values here */
+ copylex.lex_level++;
- /* Determine token type. */
- if (strchr("{}[],:", s[0]))
+ count = 0;
+ lex_expect(JSON_PARSE_ARRAY_START, ©lex, JSON_TOKEN_ARRAY_START);
+ if (lex_peek(©lex) != JSON_TOKEN_ARRAY_END)
{
- /* strchr() doesn't return false on a NUL input. */
- if (s[0] == '\0')
- {
- /* End of string. */
- lex->token_start = NULL;
- lex->token_terminator = NULL;
- }
- else
+ do
{
- /* Single-character token, some kind of punctuation mark. */
- lex->token_terminator = s + 1;
+ count++;
+ parse_array_element(©lex, &nullSemAction);
}
- lex->token_type = JSON_VALUE_INVALID;
+ while (lex_accept(©lex, JSON_TOKEN_COMMA, NULL));
}
- else if (*s == '"')
+ lex_expect(JSON_PARSE_ARRAY_NEXT, ©lex, JSON_TOKEN_ARRAY_END);
+
+ return count;
+}
+
+/*
+ * Recursive Descent parse routines. There is one for each structural
+ * element in a json document:
+ * - scalar (string, number, true, false, null)
+ * - array ( [ ] )
+ * - array element
+ * - object ( { } )
+ * - object field
+ */
+static inline void
+parse_scalar(JsonLexContext *lex, JsonSemAction *sem)
+{
+ char *val = NULL;
+ json_scalar_action sfunc = sem->scalar;
+ char **valaddr;
+ JsonTokenType tok = lex_peek(lex);
+
+ valaddr = sfunc == NULL ? NULL : &val;
+
+ /* a scalar must be a string, a number, true, false, or null */
+ switch (tok)
{
- /* String. */
- json_lex_string(lex);
- lex->token_type = JSON_VALUE_STRING;
+ case JSON_TOKEN_TRUE:
+ lex_accept(lex, JSON_TOKEN_TRUE, valaddr);
+ break;
+ case JSON_TOKEN_FALSE:
+ lex_accept(lex, JSON_TOKEN_FALSE, valaddr);
+ break;
+ case JSON_TOKEN_NULL:
+ lex_accept(lex, JSON_TOKEN_NULL, valaddr);
+ break;
+ case JSON_TOKEN_NUMBER:
+ lex_accept(lex, JSON_TOKEN_NUMBER, valaddr);
+ break;
+ case JSON_TOKEN_STRING:
+ lex_accept(lex, JSON_TOKEN_STRING, valaddr);
+ break;
+ default:
+ report_parse_error(JSON_PARSE_VALUE, lex);
}
- else if (*s == '-')
+
+ if (sfunc != NULL)
+ (*sfunc) (sem->semstate, val, tok);
+}
+
+static void
+parse_object_field(JsonLexContext *lex, JsonSemAction *sem)
+{
+ /*
+ * An object field is "fieldname" : value where value can be a scalar,
+ * object or array. Note: in user-facing docs and error messages, we
+ * generally call a field name a "key".
+ */
+
+ char *fname = NULL; /* keep compiler quiet */
+ json_ofield_action ostart = sem->object_field_start;
+ json_ofield_action oend = sem->object_field_end;
+ bool isnull;
+ char **fnameaddr = NULL;
+ JsonTokenType tok;
+
+ if (ostart != NULL || oend != NULL)
+ fnameaddr = &fname;
+
+ if (!lex_accept(lex, JSON_TOKEN_STRING, fnameaddr))
+ report_parse_error(JSON_PARSE_STRING, lex);
+
+ lex_expect(JSON_PARSE_OBJECT_LABEL, lex, JSON_TOKEN_COLON);
+
+ tok = lex_peek(lex);
+ isnull = tok == JSON_TOKEN_NULL;
+
+ if (ostart != NULL)
+ (*ostart) (sem->semstate, fname, isnull);
+
+ switch (tok)
{
- /* Negative number. */
- json_lex_number(lex, s + 1);
- lex->token_type = JSON_VALUE_NUMBER;
+ case JSON_TOKEN_OBJECT_START:
+ parse_object(lex, sem);
+ break;
+ case JSON_TOKEN_ARRAY_START:
+ parse_array(lex, sem);
+ break;
+ default:
+ parse_scalar(lex, sem);
}
- else if (*s >= '0' && *s <= '9')
+
+ if (oend != NULL)
+ (*oend) (sem->semstate, fname, isnull);
+}
+
+static void
+parse_object(JsonLexContext *lex, JsonSemAction *sem)
+{
+ /*
+ * an object is a possibly empty sequence of object fields, separated by
+ * commas and surrounded by curly braces.
+ */
+ json_struct_action ostart = sem->object_start;
+ json_struct_action oend = sem->object_end;
+ JsonTokenType tok;
+
+ check_stack_depth();
+
+ if (ostart != NULL)
+ (*ostart) (sem->semstate);
+
+ /*
+ * Data inside an object is at a higher nesting level than the object
+ * itself. Note that we increment this after we call the semantic routine
+ * for the object start and restore it before we call the routine for the
+ * object end.
+ */
+ lex->lex_level++;
+
+ /* we know this will succeed, just clearing the token */
+ lex_expect(JSON_PARSE_OBJECT_START, lex, JSON_TOKEN_OBJECT_START);
+
+ tok = lex_peek(lex);
+ switch (tok)
{
- /* Positive number. */
- json_lex_number(lex, s);
- lex->token_type = JSON_VALUE_NUMBER;
+ case JSON_TOKEN_STRING:
+ parse_object_field(lex, sem);
+ while (lex_accept(lex, JSON_TOKEN_COMMA, NULL))
+ parse_object_field(lex, sem);
+ break;
+ case JSON_TOKEN_OBJECT_END:
+ break;
+ default:
+ /* case of an invalid initial token inside the object */
+ report_parse_error(JSON_PARSE_OBJECT_START, lex);
}
- else
- {
- char *p;
- /*
- * We're not dealing with a string, number, legal punctuation mark, or
- * end of string. The only legal tokens we might find here are true,
- * false, and null, but for error reporting purposes we scan until we
- * see a non-alphanumeric character. That way, we can report the
- * whole word as an unexpected token, rather than just some
- * unintuitive prefix thereof.
- */
- for (p = s; (*p >= 'a' && *p <= 'z') || (*p >= 'A' && *p <= 'Z')
- || (*p >= '0' && *p <= '9') || *p == '_' || IS_HIGHBIT_SET(*p);
- ++p)
- ;
+ lex_expect(JSON_PARSE_OBJECT_NEXT, lex, JSON_TOKEN_OBJECT_END);
- /*
- * We got some sort of unexpected punctuation or an otherwise
- * unexpected character, so just complain about that one character.
- */
- if (p == s)
- {
- lex->token_terminator = s + 1;
- report_invalid_token(lex);
- }
+ lex->lex_level--;
- /*
- * We've got a real alphanumeric token here. If it happens to be
- * true, false, or null, all is well. If not, error out.
- */
- lex->token_terminator = p;
- if (p - s == 4)
- {
- if (memcmp(s, "true", 4) == 0)
- lex->token_type = JSON_VALUE_TRUE;
- else if (memcmp(s, "null", 4) == 0)
- lex->token_type = JSON_VALUE_NULL;
- else
- report_invalid_token(lex);
- }
- else if (p - s == 5 && memcmp(s, "false", 5) == 0)
- lex->token_type = JSON_VALUE_FALSE;
- else
- report_invalid_token(lex);
- }
+ if (oend != NULL)
+ (*oend) (sem->semstate);
}
-/*
- * The next token in the input stream is known to be a string; lex it.
- */
static void
-json_lex_string(JsonLexContext *lex)
+parse_array_element(JsonLexContext *lex, JsonSemAction *sem)
{
- char *s = lex->token_start + 1;
+ json_aelem_action astart = sem->array_element_start;
+ json_aelem_action aend = sem->array_element_end;
+ JsonTokenType tok = lex_peek(lex);
- for (s = lex->token_start + 1; *s != '"'; ++s)
- {
- /* Per RFC4627, these characters MUST be escaped. */
- if ((unsigned char) *s < 32)
- {
- /* A NUL byte marks the (premature) end of the string. */
- if (*s == '\0')
- {
- lex->token_terminator = s;
- report_invalid_token(lex);
- }
- ereport(ERROR,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json"),
- errdetail_internal("line %d: Character with value \"0x%02x\" must be escaped.",
- lex->line_number, (unsigned char) *s)));
- }
- else if (*s == '\\')
- {
- /* OK, we have an escape character. */
- ++s;
- if (*s == '\0')
- {
- lex->token_terminator = s;
- report_invalid_token(lex);
- }
- else if (*s == 'u')
- {
- int i;
- int ch = 0;
+ bool isnull;
- for (i = 1; i <= 4; ++i)
- {
- if (s[i] == '\0')
- {
- lex->token_terminator = s + i;
- report_invalid_token(lex);
- }
- else if (s[i] >= '0' && s[i] <= '9')
- ch = (ch * 16) + (s[i] - '0');
- else if (s[i] >= 'a' && s[i] <= 'f')
- ch = (ch * 16) + (s[i] - 'a') + 10;
- else if (s[i] >= 'A' && s[i] <= 'F')
- ch = (ch * 16) + (s[i] - 'A') + 10;
- else
- {
- ereport(ERROR,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json"),
- errdetail_internal("line %d: \"\\u\" must be followed by four hexadecimal digits.",
- lex->line_number)));
- }
- }
+ isnull = tok == JSON_TOKEN_NULL;
- /* Account for the four additional bytes we just parsed. */
- s += 4;
- }
- else if (!strchr("\"\\/bfnrt", *s))
- {
- /* Error out. */
- ereport(ERROR,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json"),
- errdetail_internal("line %d: Invalid escape \"\\%s\".",
- lex->line_number, extract_mb_char(s))));
- }
- }
+ if (astart != NULL)
+ (*astart) (sem->semstate, isnull);
+
+ /* an array element is any object, array or scalar */
+ switch (tok)
+ {
+ case JSON_TOKEN_OBJECT_START:
+ parse_object(lex, sem);
+ break;
+ case JSON_TOKEN_ARRAY_START:
+ parse_array(lex, sem);
+ break;
+ default:
+ parse_scalar(lex, sem);
}
+ if (aend != NULL)
+ (*aend) (sem->semstate, isnull);
+}
+
+static void
+parse_array(JsonLexContext *lex, JsonSemAction *sem)
+{
+ /*
+ * an array is a possibly empty sequence of array elements, separated by
+ * commas and surrounded by square brackets.
+ */
+ json_struct_action astart = sem->array_start;
+ json_struct_action aend = sem->array_end;
+
+ check_stack_depth();
+
+ if (astart != NULL)
+ (*astart) (sem->semstate);
+
+ /*
+ * Data inside an array is at a higher nesting level than the array
+ * itself. Note that we increment this after we call the semantic routine
+ * for the array start and restore it before we call the routine for the
+ * array end.
+ */
+ lex->lex_level++;
+
+ lex_expect(JSON_PARSE_ARRAY_START, lex, JSON_TOKEN_ARRAY_START);
+ if (lex_peek(lex) != JSON_TOKEN_ARRAY_END)
+ {
+
+ parse_array_element(lex, sem);
+
+ while (lex_accept(lex, JSON_TOKEN_COMMA, NULL))
+ parse_array_element(lex, sem);
+ }
+
+ lex_expect(JSON_PARSE_ARRAY_NEXT, lex, JSON_TOKEN_ARRAY_END);
+
+ lex->lex_level--;
+
+ if (aend != NULL)
+ (*aend) (sem->semstate);
+}
+
+/*
+ * Lex one token from the input stream.
+ */
+static inline void
+json_lex(JsonLexContext *lex)
+{
+ char *s;
+ int len;
+
+ /* Skip leading whitespace. */
+ s = lex->token_terminator;
+ len = s - lex->input;
+ while (len < lex->input_length &&
+ (*s == ' ' || *s == '\t' || *s == '\n' || *s == '\r'))
+ {
+ if (*s == '\n')
+ ++lex->line_number;
+ ++s;
+ ++len;
+ }
+ lex->token_start = s;
+
+ /* Determine token type. */
+ if (len >= lex->input_length)
+ {
+ lex->token_start = NULL;
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s;
+ lex->token_type = JSON_TOKEN_END;
+ }
+ else
+ switch (*s)
+ {
+ /* Single-character token, some kind of punctuation mark. */
+ case '{':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_OBJECT_START;
+ break;
+ case '}':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_OBJECT_END;
+ break;
+ case '[':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_ARRAY_START;
+ break;
+ case ']':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_ARRAY_END;
+ break;
+ case ',':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_COMMA;
+ break;
+ case ':':
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ lex->token_type = JSON_TOKEN_COLON;
+ break;
+ case '"':
+ /* string */
+ json_lex_string(lex);
+ lex->token_type = JSON_TOKEN_STRING;
+ break;
+ case '-':
+ /* Negative number. */
+ json_lex_number(lex, s + 1, NULL, NULL);
+ lex->token_type = JSON_TOKEN_NUMBER;
+ break;
+ case '0':
+ case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
+ case '8':
+ case '9':
+ /* Positive number. */
+ json_lex_number(lex, s, NULL, NULL);
+ lex->token_type = JSON_TOKEN_NUMBER;
+ break;
+ default:
+ {
+ char *p;
+
+ /*
+ * We're not dealing with a string, number, legal
+ * punctuation mark, or end of string. The only legal
+ * tokens we might find here are true, false, and null,
+ * but for error reporting purposes we scan until we see a
+ * non-alphanumeric character. That way, we can report
+ * the whole word as an unexpected token, rather than just
+ * some unintuitive prefix thereof.
+ */
+ for (p = s; p - s < lex->input_length - len && JSON_ALPHANUMERIC_CHAR(*p); p++)
+ /* skip */ ;
+
+ /*
+ * We got some sort of unexpected punctuation or an
+ * otherwise unexpected character, so just complain about
+ * that one character.
+ */
+ if (p == s)
+ {
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s + 1;
+ report_invalid_token(lex);
+ }
+
+ /*
+ * We've got a real alphanumeric token here. If it
+ * happens to be true, false, or null, all is well. If
+ * not, error out.
+ */
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = p;
+ if (p - s == 4)
+ {
+ if (memcmp(s, "true", 4) == 0)
+ lex->token_type = JSON_TOKEN_TRUE;
+ else if (memcmp(s, "null", 4) == 0)
+ lex->token_type = JSON_TOKEN_NULL;
+ else
+ report_invalid_token(lex);
+ }
+ else if (p - s == 5 && memcmp(s, "false", 5) == 0)
+ lex->token_type = JSON_TOKEN_FALSE;
+ else
+ report_invalid_token(lex);
+
+ }
+ } /* end of switch */
+}
+
+/*
+ * The next token in the input stream is known to be a string; lex it.
+ */
+static inline void
+json_lex_string(JsonLexContext *lex)
+{
+ char *s;
+ int len;
+ int hi_surrogate = -1;
+
+ if (lex->strval != NULL)
+ resetStringInfo(lex->strval);
+
+ Assert(lex->input_length > 0);
+ s = lex->token_start;
+ len = lex->token_start - lex->input;
+ for (;;)
+ {
+ s++;
+ len++;
+ /* Premature end of the string. */
+ if (len >= lex->input_length)
+ {
+ lex->token_terminator = s;
+ report_invalid_token(lex);
+ }
+ else if (*s == '"')
+ break;
+ else if ((unsigned char) *s < 32)
+ {
+ /* Per RFC4627, these characters MUST be escaped. */
+ /* Since *s isn't printable, exclude it from the context string */
+ lex->token_terminator = s;
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Character with value 0x%02x must be escaped.",
+ (unsigned char) *s),
+ report_json_context(lex)));
+ }
+ else if (*s == '\\')
+ {
+ /* OK, we have an escape character. */
+ s++;
+ len++;
+ if (len >= lex->input_length)
+ {
+ lex->token_terminator = s;
+ report_invalid_token(lex);
+ }
+ else if (*s == 'u')
+ {
+ int i;
+ int ch = 0;
+
+ for (i = 1; i <= 4; i++)
+ {
+ s++;
+ len++;
+ if (len >= lex->input_length)
+ {
+ lex->token_terminator = s;
+ report_invalid_token(lex);
+ }
+ else if (*s >= '0' && *s <= '9')
+ ch = (ch * 16) + (*s - '0');
+ else if (*s >= 'a' && *s <= 'f')
+ ch = (ch * 16) + (*s - 'a') + 10;
+ else if (*s >= 'A' && *s <= 'F')
+ ch = (ch * 16) + (*s - 'A') + 10;
+ else
+ {
+ lex->token_terminator = s + pg_mblen(s);
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s",
+ "json"),
+ errdetail("\"\\u\" must be followed by four hexadecimal digits."),
+ report_json_context(lex)));
+ }
+ }
+ if (lex->strval != NULL)
+ {
+ char utf8str[5];
+ int utf8len;
+
+ if (ch >= 0xd800 && ch <= 0xdbff)
+ {
+ if (hi_surrogate != -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s",
+ "json"),
+ errdetail("Unicode high surrogate must not follow a high surrogate."),
+ report_json_context(lex)));
+ hi_surrogate = (ch & 0x3ff) << 10;
+ continue;
+ }
+ else if (ch >= 0xdc00 && ch <= 0xdfff)
+ {
+ if (hi_surrogate == -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Unicode low surrogate must follow a high surrogate."),
+ report_json_context(lex)));
+ ch = 0x10000 + hi_surrogate + (ch & 0x3ff);
+ hi_surrogate = -1;
+ }
+
+ if (hi_surrogate != -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Unicode low surrogate must follow a high surrogate."),
+ report_json_context(lex)));
+
+ /*
+ * For UTF8, replace the escape sequence by the actual
+ * utf8 character in lex->strval. Do this also for other
+ * encodings if the escape designates an ASCII character,
+ * otherwise raise an error.
+ */
+
+ if (ch == 0)
+ {
+ /* We can't allow this, since our TEXT type doesn't */
+ ereport(ERROR,
+ (errcode(ERRCODE_UNTRANSLATABLE_CHARACTER),
+ errmsg("unsupported Unicode escape sequence"),
+ errdetail("\\u0000 cannot be converted to text."),
+ report_json_context(lex)));
+ }
+ else if (GetDatabaseEncoding() == PG_UTF8)
+ {
+ unicode_to_utf8(ch, (unsigned char *) utf8str);
+ utf8len = pg_utf_mblen((unsigned char *) utf8str);
+ appendBinaryStringInfo(lex->strval, utf8str, utf8len);
+ }
+ else if (ch <= 0x007f)
+ {
+ /*
+ * This is the only way to designate things like a
+ * form feed character in JSON, so it's useful in all
+ * encodings.
+ */
+ appendStringInfoChar(lex->strval, (char) ch);
+ }
+ else
+ {
+ ereport(ERROR,
+ (errcode(ERRCODE_UNTRANSLATABLE_CHARACTER),
+ errmsg("unsupported Unicode escape sequence"),
+ errdetail("Unicode escape values cannot be used for code point values above 007F when the server encoding is not UTF8."),
+ report_json_context(lex)));
+ }
+
+ }
+ }
+ else if (lex->strval != NULL)
+ {
+ if (hi_surrogate != -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s",
+ "json"),
+ errdetail("Unicode low surrogate must follow a high surrogate."),
+ report_json_context(lex)));
+
+ switch (*s)
+ {
+ case '"':
+ case '\\':
+ case '/':
+ appendStringInfoChar(lex->strval, *s);
+ break;
+ case 'b':
+ appendStringInfoChar(lex->strval, '\b');
+ break;
+ case 'f':
+ appendStringInfoChar(lex->strval, '\f');
+ break;
+ case 'n':
+ appendStringInfoChar(lex->strval, '\n');
+ break;
+ case 'r':
+ appendStringInfoChar(lex->strval, '\r');
+ break;
+ case 't':
+ appendStringInfoChar(lex->strval, '\t');
+ break;
+ default:
+ /* Not a valid string escape, so error out. */
+ lex->token_terminator = s + pg_mblen(s);
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s",
+ "json"),
+ errdetail("Escape sequence \"\\%s\" is invalid.",
+ extract_mb_char(s)),
+ report_json_context(lex)));
+ }
+ }
+ else if (strchr("\"\\/bfnrt", *s) == NULL)
+ {
+ /*
+ * Simpler processing if we're not bothered about de-escaping
+ *
+ * It's very tempting to remove the strchr() call here and
+ * replace it with a switch statement, but testing so far has
+ * shown it's not a performance win.
+ */
+ lex->token_terminator = s + pg_mblen(s);
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Escape sequence \"\\%s\" is invalid.",
+ extract_mb_char(s)),
+ report_json_context(lex)));
+ }
+
+ }
+ else if (lex->strval != NULL)
+ {
+ if (hi_surrogate != -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Unicode low surrogate must follow a high surrogate."),
+ report_json_context(lex)));
+
+ appendStringInfoChar(lex->strval, *s);
+ }
+
+ }
+
+ if (hi_surrogate != -1)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Unicode low surrogate must follow a high surrogate."),
+ report_json_context(lex)));
+
/* Hooray, we found the end of the string! */
+ lex->prev_token_terminator = lex->token_terminator;
lex->token_terminator = s + 1;
}
-/*-------------------------------------------------------------------------
+/*
* The next token in the input stream is known to be a number; lex it.
*
* In JSON, a number consists of four parts:
* begin with a '0'.
*
* (3) An optional decimal part, consisting of a period ('.') followed by
- * one or more digits. (Note: While this part can be omitted
+ * one or more digits. (Note: While this part can be omitted
* completely, it's not OK to have only the decimal point without
* any digits afterwards.)
*
* (4) An optional exponent part, consisting of 'e' or 'E', optionally
- * followed by '+' or '-', followed by one or more digits. (Note:
+ * followed by '+' or '-', followed by one or more digits. (Note:
* As with the decimal part, if 'e' or 'E' is present, it must be
* followed by at least one digit.)
*
* The 's' argument to this function points to the ostensible beginning
- * of part 2 - i.e. the character after any optional minus sign, and the
+ * of part 2 - i.e. the character after any optional minus sign, or the
* first character of the string if there is none.
*
- *-------------------------------------------------------------------------
+ * If num_err is not NULL, we return an error flag to *num_err rather than
+ * raising an error for a badly-formed number. Also, if total_len is not NULL
+ * the distance from lex->input to the token end+1 is returned to *total_len.
*/
-static void
-json_lex_number(JsonLexContext *lex, char *s)
+static inline void
+json_lex_number(JsonLexContext *lex, char *s,
+ bool *num_err, int *total_len)
{
bool error = false;
- char *p;
+ int len = s - lex->input;
/* Part (1): leading sign indicator. */
/* Caller already did this for us; so do nothing. */
/* Part (2): parse main digit string. */
- if (*s == '0')
- ++s;
- else if (*s >= '1' && *s <= '9')
+ if (len < lex->input_length && *s == '0')
+ {
+ s++;
+ len++;
+ }
+ else if (len < lex->input_length && *s >= '1' && *s <= '9')
{
do
{
- ++s;
- } while (*s >= '0' && *s <= '9');
+ s++;
+ len++;
+ } while (len < lex->input_length && *s >= '0' && *s <= '9');
}
else
error = true;
/* Part (3): parse optional decimal portion. */
- if (*s == '.')
+ if (len < lex->input_length && *s == '.')
{
- ++s;
- if (*s < '0' || *s > '9')
+ s++;
+ len++;
+ if (len == lex->input_length || *s < '0' || *s > '9')
error = true;
else
{
do
{
- ++s;
- } while (*s >= '0' && *s <= '9');
+ s++;
+ len++;
+ } while (len < lex->input_length && *s >= '0' && *s <= '9');
}
}
/* Part (4): parse optional exponent. */
- if (*s == 'e' || *s == 'E')
+ if (len < lex->input_length && (*s == 'e' || *s == 'E'))
{
- ++s;
- if (*s == '+' || *s == '-')
- ++s;
- if (*s < '0' || *s > '9')
+ s++;
+ len++;
+ if (len < lex->input_length && (*s == '+' || *s == '-'))
+ {
+ s++;
+ len++;
+ }
+ if (len == lex->input_length || *s < '0' || *s > '9')
error = true;
else
{
do
{
- ++s;
- } while (*s >= '0' && *s <= '9');
+ s++;
+ len++;
+ } while (len < lex->input_length && *s >= '0' && *s <= '9');
}
}
- /* Check for trailing garbage. */
- for (p = s; (*p >= 'a' && *p <= 'z') || (*p >= 'A' && *p <= 'Z')
- || (*p >= '0' && *p <= '9') || *p == '_' || IS_HIGHBIT_SET(*p); ++p)
- ;
- lex->token_terminator = p;
- if (p > s || error)
- report_invalid_token(lex);
+ /*
+ * Check for trailing garbage. As in json_lex(), any alphanumeric stuff
+ * here should be considered part of the token for error-reporting
+ * purposes.
+ */
+ for (; len < lex->input_length && JSON_ALPHANUMERIC_CHAR(*s); s++, len++)
+ error = true;
+
+ if (total_len != NULL)
+ *total_len = len;
+
+ if (num_err != NULL)
+ {
+ /* let the caller handle any error */
+ *num_err = error;
+ }
+ else
+ {
+ /* return token endpoint */
+ lex->prev_token_terminator = lex->token_terminator;
+ lex->token_terminator = s;
+ /* handle error if any */
+ if (error)
+ report_invalid_token(lex);
+ }
}
/*
* Report a parse error.
+ *
+ * lex->token_start and lex->token_terminator must identify the current token.
*/
static void
-report_parse_error(JsonParseStack *stack, JsonLexContext *lex)
+report_parse_error(JsonParseContext ctx, JsonLexContext *lex)
{
- char *detail = NULL;
- char *token = NULL;
+ char *token;
int toklen;
/* Handle case where the input ended prematurely. */
- if (lex->token_start == NULL)
+ if (lex->token_start == NULL || lex->token_type == JSON_TOKEN_END)
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json: \"%s\"",
- lex->input),
- errdetail_internal("The input string ended unexpectedly.")));
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("The input string ended unexpectedly."),
+ report_json_context(lex)));
- /* Work out the offending token. */
+ /* Separate out the current token. */
toklen = lex->token_terminator - lex->token_start;
token = palloc(toklen + 1);
memcpy(token, lex->token_start, toklen);
token[toklen] = '\0';
- /* Select correct detail message. */
- if (stack == NULL)
- detail = "line %d: Expected end of input, but found \"%s\".";
+ /* Complain, with the appropriate detail message. */
+ if (ctx == JSON_PARSE_END)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected end of input, but found \"%s\".",
+ token),
+ report_json_context(lex)));
else
{
- switch (stack->state)
+ switch (ctx)
{
case JSON_PARSE_VALUE:
- detail = "line %d: Expected string, number, object, array, true, false, or null, but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected JSON value, but found \"%s\".",
+ token),
+ report_json_context(lex)));
+ break;
+ case JSON_PARSE_STRING:
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected string, but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_ARRAY_START:
- detail = "line %d: Expected array element or \"]\", but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected array element or \"]\", but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_ARRAY_NEXT:
- detail = "line %d: Expected \",\" or \"]\", but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected \",\" or \"]\", but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_OBJECT_START:
- detail = "line %d: Expected string or \"}\", but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected string or \"}\", but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_OBJECT_LABEL:
- detail = "line %d: Expected \":\", but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected \":\", but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_OBJECT_NEXT:
- detail = "line %d: Expected \",\" or \"}\", but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected \",\" or \"}\", but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
case JSON_PARSE_OBJECT_COMMA:
- detail = "line %d: Expected string, but found \"%s\".";
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Expected string, but found \"%s\".",
+ token),
+ report_json_context(lex)));
break;
+ default:
+ elog(ERROR, "unexpected json parse state: %d", ctx);
}
}
-
- ereport(ERROR,
- (errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json: \"%s\"",
- lex->input),
- detail ? errdetail_internal(detail, lex->line_number, token) : 0));
}
/*
* Report an invalid input token.
+ *
+ * lex->token_start and lex->token_terminator must identify the token.
*/
static void
report_invalid_token(JsonLexContext *lex)
char *token;
int toklen;
+ /* Separate out the offending token. */
toklen = lex->token_terminator - lex->token_start;
token = palloc(toklen + 1);
memcpy(token, lex->token_start, toklen);
ereport(ERROR,
(errcode(ERRCODE_INVALID_TEXT_REPRESENTATION),
- errmsg("invalid input syntax for type json"),
- errdetail_internal("line %d: Token \"%s\" is invalid.",
- lex->line_number, token)));
+ errmsg("invalid input syntax for type %s", "json"),
+ errdetail("Token \"%s\" is invalid.", token),
+ report_json_context(lex)));
+}
+
+/*
+ * Report a CONTEXT line for bogus JSON input.
+ *
+ * lex->token_terminator must be set to identify the spot where we detected
+ * the error. Note that lex->token_start might be NULL, in case we recognized
+ * error at EOF.
+ *
+ * The return value isn't meaningful, but we make it non-void so that this
+ * can be invoked inside ereport().
+ */
+static int
+report_json_context(JsonLexContext *lex)
+{
+ const char *context_start;
+ const char *context_end;
+ const char *line_start;
+ int line_number;
+ char *ctxt;
+ int ctxtlen;
+ const char *prefix;
+ const char *suffix;
+
+ /* Choose boundaries for the part of the input we will display */
+ context_start = lex->input;
+ context_end = lex->token_terminator;
+ line_start = context_start;
+ line_number = 1;
+ for (;;)
+ {
+ /* Always advance over newlines */
+ if (context_start < context_end && *context_start == '\n')
+ {
+ context_start++;
+ line_start = context_start;
+ line_number++;
+ continue;
+ }
+ /* Otherwise, done as soon as we are close enough to context_end */
+ if (context_end - context_start < 50)
+ break;
+ /* Advance to next multibyte character */
+ if (IS_HIGHBIT_SET(*context_start))
+ context_start += pg_mblen(context_start);
+ else
+ context_start++;
+ }
+
+ /*
+ * We add "..." to indicate that the excerpt doesn't start at the
+ * beginning of the line ... but if we're within 3 characters of the
+ * beginning of the line, we might as well just show the whole line.
+ */
+ if (context_start - line_start <= 3)
+ context_start = line_start;
+
+ /* Get a null-terminated copy of the data to present */
+ ctxtlen = context_end - context_start;
+ ctxt = palloc(ctxtlen + 1);
+ memcpy(ctxt, context_start, ctxtlen);
+ ctxt[ctxtlen] = '\0';
+
+ /*
+ * Show the context, prefixing "..." if not starting at start of line, and
+ * suffixing "..." if not ending at end of line.
+ */
+ prefix = (context_start > line_start) ? "..." : "";
+ suffix = (lex->token_type != JSON_TOKEN_END && context_end - lex->input < lex->input_length && *context_end != '\n' && *context_end != '\r') ? "..." : "";
+
+ return errcontext("JSON data, line %d: %s%s%s",
+ line_number, prefix, ctxt, suffix);
}
/*
memcpy(res, s, len);
res[len] = '\0';
- return res;
+ return res;
+}
+
+/*
+ * Determine how we want to print values of a given type in datum_to_json.
+ *
+ * Given the datatype OID, return its JsonTypeCategory, as well as the type's
+ * output function OID. If the returned category is JSONTYPE_CAST, we
+ * return the OID of the type->JSON cast function instead.
+ */
+static void
+json_categorize_type(Oid typoid,
+ JsonTypeCategory *tcategory,
+ Oid *outfuncoid)
+{
+ bool typisvarlena;
+
+ /* Look through any domain */
+ typoid = getBaseType(typoid);
+
+ *outfuncoid = InvalidOid;
+
+ /*
+ * We need to get the output function for everything except date and
+ * timestamp types, array and composite types, booleans, and non-builtin
+ * types where there's a cast to json.
+ */
+
+ switch (typoid)
+ {
+ case BOOLOID:
+ *tcategory = JSONTYPE_BOOL;
+ break;
+
+ case INT2OID:
+ case INT4OID:
+ case INT8OID:
+ case FLOAT4OID:
+ case FLOAT8OID:
+ case NUMERICOID:
+ getTypeOutputInfo(typoid, outfuncoid, &typisvarlena);
+ *tcategory = JSONTYPE_NUMERIC;
+ break;
+
+ case DATEOID:
+ *tcategory = JSONTYPE_DATE;
+ break;
+
+ case TIMESTAMPOID:
+ *tcategory = JSONTYPE_TIMESTAMP;
+ break;
+
+ case TIMESTAMPTZOID:
+ *tcategory = JSONTYPE_TIMESTAMPTZ;
+ break;
+
+ case JSONOID:
+ case JSONBOID:
+ getTypeOutputInfo(typoid, outfuncoid, &typisvarlena);
+ *tcategory = JSONTYPE_JSON;
+ break;
+
+ default:
+ /* Check for arrays and composites */
+ if (OidIsValid(get_element_type(typoid)) || typoid == ANYARRAYOID
+ || typoid == RECORDARRAYOID)
+ *tcategory = JSONTYPE_ARRAY;
+ else if (type_is_rowtype(typoid)) /* includes RECORDOID */
+ *tcategory = JSONTYPE_COMPOSITE;
+ else
+ {
+ /* It's probably the general case ... */
+ *tcategory = JSONTYPE_OTHER;
+ /* but let's look for a cast to json, if it's not built-in */
+ if (typoid >= FirstNormalObjectId)
+ {
+ Oid castfunc;
+ CoercionPathType ctype;
+
+ ctype = find_coercion_pathway(JSONOID, typoid,
+ COERCION_EXPLICIT,
+ &castfunc);
+ if (ctype == COERCION_PATH_FUNC && OidIsValid(castfunc))
+ {
+ *tcategory = JSONTYPE_CAST;
+ *outfuncoid = castfunc;
+ }
+ else
+ {
+ /* non builtin type with no cast */
+ getTypeOutputInfo(typoid, outfuncoid, &typisvarlena);
+ }
+ }
+ else
+ {
+ /* any other builtin type */
+ getTypeOutputInfo(typoid, outfuncoid, &typisvarlena);
+ }
+ }
+ break;
+ }
+}
+
+/*
+ * Turn a Datum into JSON text, appending the string to "result".
+ *
+ * tcategory and outfuncoid are from a previous call to json_categorize_type,
+ * except that if is_null is true then they can be invalid.
+ *
+ * If key_scalar is true, the value is being printed as a key, so insist
+ * it's of an acceptable type, and force it to be quoted.
+ */
+static void
+datum_to_json(Datum val, bool is_null, StringInfo result,
+ JsonTypeCategory tcategory, Oid outfuncoid,
+ bool key_scalar)
+{
+ char *outputstr;
+ text *jsontext;
+
+ check_stack_depth();
+
+ /* callers are expected to ensure that null keys are not passed in */
+ Assert(!(key_scalar && is_null));
+
+ if (is_null)
+ {
+ appendStringInfoString(result, "null");
+ return;
+ }
+
+ if (key_scalar &&
+ (tcategory == JSONTYPE_ARRAY ||
+ tcategory == JSONTYPE_COMPOSITE ||
+ tcategory == JSONTYPE_JSON ||
+ tcategory == JSONTYPE_CAST))
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("key value must be scalar, not array, composite, or json")));
+
+ switch (tcategory)
+ {
+ case JSONTYPE_ARRAY:
+ array_to_json_internal(val, result, false);
+ break;
+ case JSONTYPE_COMPOSITE:
+ composite_to_json(val, result, false);
+ break;
+ case JSONTYPE_BOOL:
+ outputstr = DatumGetBool(val) ? "true" : "false";
+ if (key_scalar)
+ escape_json(result, outputstr);
+ else
+ appendStringInfoString(result, outputstr);
+ break;
+ case JSONTYPE_NUMERIC:
+ outputstr = OidOutputFunctionCall(outfuncoid, val);
+
+ /*
+ * Don't call escape_json for a non-key if it's a valid JSON
+ * number.
+ */
+ if (!key_scalar && IsValidJsonNumber(outputstr, strlen(outputstr)))
+ appendStringInfoString(result, outputstr);
+ else
+ escape_json(result, outputstr);
+ pfree(outputstr);
+ break;
+ case JSONTYPE_DATE:
+ {
+ char buf[MAXDATELEN + 1];
+
+ JsonEncodeDateTime(buf, val, DATEOID, NULL);
+ appendStringInfo(result, "\"%s\"", buf);
+ }
+ break;
+ case JSONTYPE_TIMESTAMP:
+ {
+ char buf[MAXDATELEN + 1];
+
+ JsonEncodeDateTime(buf, val, TIMESTAMPOID, NULL);
+ appendStringInfo(result, "\"%s\"", buf);
+ }
+ break;
+ case JSONTYPE_TIMESTAMPTZ:
+ {
+ char buf[MAXDATELEN + 1];
+
+ JsonEncodeDateTime(buf, val, TIMESTAMPTZOID, NULL);
+ appendStringInfo(result, "\"%s\"", buf);
+ }
+ break;
+ case JSONTYPE_JSON:
+ /* JSON and JSONB output will already be escaped */
+ outputstr = OidOutputFunctionCall(outfuncoid, val);
+ appendStringInfoString(result, outputstr);
+ pfree(outputstr);
+ break;
+ case JSONTYPE_CAST:
+ /* outfuncoid refers to a cast function, not an output function */
+ jsontext = DatumGetTextPP(OidFunctionCall1(outfuncoid, val));
+ outputstr = text_to_cstring(jsontext);
+ appendStringInfoString(result, outputstr);
+ pfree(outputstr);
+ pfree(jsontext);
+ break;
+ default:
+ outputstr = OidOutputFunctionCall(outfuncoid, val);
+ escape_json(result, outputstr);
+ pfree(outputstr);
+ break;
+ }
+}
+
+/*
+ * Encode 'value' of datetime type 'typid' into JSON string in ISO format using
+ * optionally preallocated buffer 'buf'. Optional 'tzp' determines time-zone
+ * offset (in seconds) in which we want to show timestamptz.
+ */
+char *
+JsonEncodeDateTime(char *buf, Datum value, Oid typid, const int *tzp)
+{
+ if (!buf)
+ buf = palloc(MAXDATELEN + 1);
+
+ switch (typid)
+ {
+ case DATEOID:
+ {
+ DateADT date;
+ struct pg_tm tm;
+
+ date = DatumGetDateADT(value);
+
+ /* Same as date_out(), but forcing DateStyle */
+ if (DATE_NOT_FINITE(date))
+ EncodeSpecialDate(date, buf);
+ else
+ {
+ j2date(date + POSTGRES_EPOCH_JDATE,
+ &(tm.tm_year), &(tm.tm_mon), &(tm.tm_mday));
+ EncodeDateOnly(&tm, USE_XSD_DATES, buf);
+ }
+ }
+ break;
+ case TIMEOID:
+ {
+ TimeADT time = DatumGetTimeADT(value);
+ struct pg_tm tt,
+ *tm = &tt;
+ fsec_t fsec;
+
+ /* Same as time_out(), but forcing DateStyle */
+ time2tm(time, tm, &fsec);
+ EncodeTimeOnly(tm, fsec, false, 0, USE_XSD_DATES, buf);
+ }
+ break;
+ case TIMETZOID:
+ {
+ TimeTzADT *time = DatumGetTimeTzADTP(value);
+ struct pg_tm tt,
+ *tm = &tt;
+ fsec_t fsec;
+ int tz;
+
+ /* Same as timetz_out(), but forcing DateStyle */
+ timetz2tm(time, tm, &fsec, &tz);
+ EncodeTimeOnly(tm, fsec, true, tz, USE_XSD_DATES, buf);
+ }
+ break;
+ case TIMESTAMPOID:
+ {
+ Timestamp timestamp;
+ struct pg_tm tm;
+ fsec_t fsec;
+
+ timestamp = DatumGetTimestamp(value);
+ /* Same as timestamp_out(), but forcing DateStyle */
+ if (TIMESTAMP_NOT_FINITE(timestamp))
+ EncodeSpecialTimestamp(timestamp, buf);
+ else if (timestamp2tm(timestamp, NULL, &tm, &fsec, NULL, NULL) == 0)
+ EncodeDateTime(&tm, fsec, false, 0, NULL, USE_XSD_DATES, buf);
+ else
+ ereport(ERROR,
+ (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
+ errmsg("timestamp out of range")));
+ }
+ break;
+ case TIMESTAMPTZOID:
+ {
+ TimestampTz timestamp;
+ struct pg_tm tm;
+ int tz;
+ fsec_t fsec;
+ const char *tzn = NULL;
+
+ timestamp = DatumGetTimestampTz(value);
+
+ /*
+ * If a time zone is specified, we apply the time-zone shift,
+ * convert timestamptz to pg_tm as if it were without a time
+ * zone, and then use the specified time zone for converting
+ * the timestamp into a string.
+ */
+ if (tzp)
+ {
+ tz = *tzp;
+ timestamp -= (TimestampTz) tz * USECS_PER_SEC;
+ }
+
+ /* Same as timestamptz_out(), but forcing DateStyle */
+ if (TIMESTAMP_NOT_FINITE(timestamp))
+ EncodeSpecialTimestamp(timestamp, buf);
+ else if (timestamp2tm(timestamp, tzp ? NULL : &tz, &tm, &fsec,
+ tzp ? NULL : &tzn, NULL) == 0)
+ {
+ if (tzp)
+ tm.tm_isdst = 1; /* set time-zone presence flag */
+
+ EncodeDateTime(&tm, fsec, true, tz, tzn, USE_XSD_DATES, buf);
+ }
+ else
+ ereport(ERROR,
+ (errcode(ERRCODE_DATETIME_VALUE_OUT_OF_RANGE),
+ errmsg("timestamp out of range")));
+ }
+ break;
+ default:
+ elog(ERROR, "unknown jsonb value datetime type oid %d", typid);
+ return NULL;
+ }
+
+ return buf;
+}
+
+/*
+ * Process a single dimension of an array.
+ * If it's the innermost dimension, output the values, otherwise call
+ * ourselves recursively to process the next dimension.
+ */
+static void
+array_dim_to_json(StringInfo result, int dim, int ndims, int *dims, Datum *vals,
+ bool *nulls, int *valcount, JsonTypeCategory tcategory,
+ Oid outfuncoid, bool use_line_feeds)
+{
+ int i;
+ const char *sep;
+
+ Assert(dim < ndims);
+
+ sep = use_line_feeds ? ",\n " : ",";
+
+ appendStringInfoChar(result, '[');
+
+ for (i = 1; i <= dims[dim]; i++)
+ {
+ if (i > 1)
+ appendStringInfoString(result, sep);
+
+ if (dim + 1 == ndims)
+ {
+ datum_to_json(vals[*valcount], nulls[*valcount], result, tcategory,
+ outfuncoid, false);
+ (*valcount)++;
+ }
+ else
+ {
+ /*
+ * Do we want line feeds on inner dimensions of arrays? For now
+ * we'll say no.
+ */
+ array_dim_to_json(result, dim + 1, ndims, dims, vals, nulls,
+ valcount, tcategory, outfuncoid, false);
+ }
+ }
+
+ appendStringInfoChar(result, ']');
+}
+
+/*
+ * Turn an array into JSON.
+ */
+static void
+array_to_json_internal(Datum array, StringInfo result, bool use_line_feeds)
+{
+ ArrayType *v = DatumGetArrayTypeP(array);
+ Oid element_type = ARR_ELEMTYPE(v);
+ int *dim;
+ int ndim;
+ int nitems;
+ int count = 0;
+ Datum *elements;
+ bool *nulls;
+ int16 typlen;
+ bool typbyval;
+ char typalign;
+ JsonTypeCategory tcategory;
+ Oid outfuncoid;
+
+ ndim = ARR_NDIM(v);
+ dim = ARR_DIMS(v);
+ nitems = ArrayGetNItems(ndim, dim);
+
+ if (nitems <= 0)
+ {
+ appendStringInfoString(result, "[]");
+ return;
+ }
+
+ get_typlenbyvalalign(element_type,
+ &typlen, &typbyval, &typalign);
+
+ json_categorize_type(element_type,
+ &tcategory, &outfuncoid);
+
+ deconstruct_array(v, element_type, typlen, typbyval,
+ typalign, &elements, &nulls,
+ &nitems);
+
+ array_dim_to_json(result, 0, ndim, dim, elements, nulls, &count, tcategory,
+ outfuncoid, use_line_feeds);
+
+ pfree(elements);
+ pfree(nulls);
+}
+
+/*
+ * Turn a composite / record into JSON.
+ */
+static void
+composite_to_json(Datum composite, StringInfo result, bool use_line_feeds)
+{
+ HeapTupleHeader td;
+ Oid tupType;
+ int32 tupTypmod;
+ TupleDesc tupdesc;
+ HeapTupleData tmptup,
+ *tuple;
+ int i;
+ bool needsep = false;
+ const char *sep;
+
+ sep = use_line_feeds ? ",\n " : ",";
+
+ td = DatumGetHeapTupleHeader(composite);
+
+ /* Extract rowtype info and find a tupdesc */
+ tupType = HeapTupleHeaderGetTypeId(td);
+ tupTypmod = HeapTupleHeaderGetTypMod(td);
+ tupdesc = lookup_rowtype_tupdesc(tupType, tupTypmod);
+
+ /* Build a temporary HeapTuple control structure */
+ tmptup.t_len = HeapTupleHeaderGetDatumLength(td);
+ tmptup.t_data = td;
+ tuple = &tmptup;
+
+ appendStringInfoChar(result, '{');
+
+ for (i = 0; i < tupdesc->natts; i++)
+ {
+ Datum val;
+ bool isnull;
+ char *attname;
+ JsonTypeCategory tcategory;
+ Oid outfuncoid;
+ Form_pg_attribute att = TupleDescAttr(tupdesc, i);
+
+ if (att->attisdropped)
+ continue;
+
+ if (needsep)
+ appendStringInfoString(result, sep);
+ needsep = true;
+
+ attname = NameStr(att->attname);
+ escape_json(result, attname);
+ appendStringInfoChar(result, ':');
+
+ val = heap_getattr(tuple, i + 1, tupdesc, &isnull);
+
+ if (isnull)
+ {
+ tcategory = JSONTYPE_NULL;
+ outfuncoid = InvalidOid;
+ }
+ else
+ json_categorize_type(att->atttypid, &tcategory, &outfuncoid);
+
+ datum_to_json(val, isnull, result, tcategory, outfuncoid, false);
+ }
+
+ appendStringInfoChar(result, '}');
+ ReleaseTupleDesc(tupdesc);
+}
+
+/*
+ * Append JSON text for "val" to "result".
+ *
+ * This is just a thin wrapper around datum_to_json. If the same type will be
+ * printed many times, avoid using this; better to do the json_categorize_type
+ * lookups only once.
+ */
+static void
+add_json(Datum val, bool is_null, StringInfo result,
+ Oid val_type, bool key_scalar)
+{
+ JsonTypeCategory tcategory;
+ Oid outfuncoid;
+
+ if (val_type == InvalidOid)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("could not determine input data type")));
+
+ if (is_null)
+ {
+ tcategory = JSONTYPE_NULL;
+ outfuncoid = InvalidOid;
+ }
+ else
+ json_categorize_type(val_type,
+ &tcategory, &outfuncoid);
+
+ datum_to_json(val, is_null, result, tcategory, outfuncoid, key_scalar);
+}
+
+/*
+ * SQL function array_to_json(row)
+ */
+Datum
+array_to_json(PG_FUNCTION_ARGS)
+{
+ Datum array = PG_GETARG_DATUM(0);
+ StringInfo result;
+
+ result = makeStringInfo();
+
+ array_to_json_internal(array, result, false);
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
+}
+
+/*
+ * SQL function array_to_json(row, prettybool)
+ */
+Datum
+array_to_json_pretty(PG_FUNCTION_ARGS)
+{
+ Datum array = PG_GETARG_DATUM(0);
+ bool use_line_feeds = PG_GETARG_BOOL(1);
+ StringInfo result;
+
+ result = makeStringInfo();
+
+ array_to_json_internal(array, result, use_line_feeds);
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
+}
+
+/*
+ * SQL function row_to_json(row)
+ */
+Datum
+row_to_json(PG_FUNCTION_ARGS)
+{
+ Datum array = PG_GETARG_DATUM(0);
+ StringInfo result;
+
+ result = makeStringInfo();
+
+ composite_to_json(array, result, false);
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
+}
+
+/*
+ * SQL function row_to_json(row, prettybool)
+ */
+Datum
+row_to_json_pretty(PG_FUNCTION_ARGS)
+{
+ Datum array = PG_GETARG_DATUM(0);
+ bool use_line_feeds = PG_GETARG_BOOL(1);
+ StringInfo result;
+
+ result = makeStringInfo();
+
+ composite_to_json(array, result, use_line_feeds);
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
+}
+
+/*
+ * SQL function to_json(anyvalue)
+ */
+Datum
+to_json(PG_FUNCTION_ARGS)
+{
+ Datum val = PG_GETARG_DATUM(0);
+ Oid val_type = get_fn_expr_argtype(fcinfo->flinfo, 0);
+ StringInfo result;
+ JsonTypeCategory tcategory;
+ Oid outfuncoid;
+
+ if (val_type == InvalidOid)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("could not determine input data type")));
+
+ json_categorize_type(val_type,
+ &tcategory, &outfuncoid);
+
+ result = makeStringInfo();
+
+ datum_to_json(val, false, result, tcategory, outfuncoid, false);
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
+}
+
+/*
+ * json_agg transition function
+ *
+ * aggregate input column as a json array value.
+ */
+Datum
+json_agg_transfn(PG_FUNCTION_ARGS)
+{
+ MemoryContext aggcontext,
+ oldcontext;
+ JsonAggState *state;
+ Datum val;
+
+ if (!AggCheckCallContext(fcinfo, &aggcontext))
+ {
+ /* cannot be called directly because of internal-type argument */
+ elog(ERROR, "json_agg_transfn called in non-aggregate context");
+ }
+
+ if (PG_ARGISNULL(0))
+ {
+ Oid arg_type = get_fn_expr_argtype(fcinfo->flinfo, 1);
+
+ if (arg_type == InvalidOid)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("could not determine input data type")));
+
+ /*
+ * Make this state object in a context where it will persist for the
+ * duration of the aggregate call. MemoryContextSwitchTo is only
+ * needed the first time, as the StringInfo routines make sure they
+ * use the right context to enlarge the object if necessary.
+ */
+ oldcontext = MemoryContextSwitchTo(aggcontext);
+ state = (JsonAggState *) palloc(sizeof(JsonAggState));
+ state->str = makeStringInfo();
+ MemoryContextSwitchTo(oldcontext);
+
+ appendStringInfoChar(state->str, '[');
+ json_categorize_type(arg_type, &state->val_category,
+ &state->val_output_func);
+ }
+ else
+ {
+ state = (JsonAggState *) PG_GETARG_POINTER(0);
+ appendStringInfoString(state->str, ", ");
+ }
+
+ /* fast path for NULLs */
+ if (PG_ARGISNULL(1))
+ {
+ datum_to_json((Datum) 0, true, state->str, JSONTYPE_NULL,
+ InvalidOid, false);
+ PG_RETURN_POINTER(state);
+ }
+
+ val = PG_GETARG_DATUM(1);
+
+ /* add some whitespace if structured type and not first item */
+ if (!PG_ARGISNULL(0) &&
+ (state->val_category == JSONTYPE_ARRAY ||
+ state->val_category == JSONTYPE_COMPOSITE))
+ {
+ appendStringInfoString(state->str, "\n ");
+ }
+
+ datum_to_json(val, false, state->str, state->val_category,
+ state->val_output_func, false);
+
+ /*
+ * The transition type for json_agg() is declared to be "internal", which
+ * is a pass-by-value type the same size as a pointer. So we can safely
+ * pass the JsonAggState pointer through nodeAgg.c's machinations.
+ */
+ PG_RETURN_POINTER(state);
+}
+
+/*
+ * json_agg final function
+ */
+Datum
+json_agg_finalfn(PG_FUNCTION_ARGS)
+{
+ JsonAggState *state;
+
+ /* cannot be called directly because of internal-type argument */
+ Assert(AggCheckCallContext(fcinfo, NULL));
+
+ state = PG_ARGISNULL(0) ?
+ NULL :
+ (JsonAggState *) PG_GETARG_POINTER(0);
+
+ /* NULL result for no rows in, as is standard with aggregates */
+ if (state == NULL)
+ PG_RETURN_NULL();
+
+ /* Else return state with appropriate array terminator added */
+ PG_RETURN_TEXT_P(catenate_stringinfo_string(state->str, "]"));
+}
+
+/*
+ * json_object_agg transition function.
+ *
+ * aggregate two input columns as a single json object value.
+ */
+Datum
+json_object_agg_transfn(PG_FUNCTION_ARGS)
+{
+ MemoryContext aggcontext,
+ oldcontext;
+ JsonAggState *state;
+ Datum arg;
+
+ if (!AggCheckCallContext(fcinfo, &aggcontext))
+ {
+ /* cannot be called directly because of internal-type argument */
+ elog(ERROR, "json_object_agg_transfn called in non-aggregate context");
+ }
+
+ if (PG_ARGISNULL(0))
+ {
+ Oid arg_type;
+
+ /*
+ * Make the StringInfo in a context where it will persist for the
+ * duration of the aggregate call. Switching context is only needed
+ * for this initial step, as the StringInfo routines make sure they
+ * use the right context to enlarge the object if necessary.
+ */
+ oldcontext = MemoryContextSwitchTo(aggcontext);
+ state = (JsonAggState *) palloc(sizeof(JsonAggState));
+ state->str = makeStringInfo();
+ MemoryContextSwitchTo(oldcontext);
+
+ arg_type = get_fn_expr_argtype(fcinfo->flinfo, 1);
+
+ if (arg_type == InvalidOid)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("could not determine data type for argument %d", 1)));
+
+ json_categorize_type(arg_type, &state->key_category,
+ &state->key_output_func);
+
+ arg_type = get_fn_expr_argtype(fcinfo->flinfo, 2);
+
+ if (arg_type == InvalidOid)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("could not determine data type for argument %d", 2)));
+
+ json_categorize_type(arg_type, &state->val_category,
+ &state->val_output_func);
+
+ appendStringInfoString(state->str, "{ ");
+ }
+ else
+ {
+ state = (JsonAggState *) PG_GETARG_POINTER(0);
+ appendStringInfoString(state->str, ", ");
+ }
+
+ /*
+ * Note: since json_object_agg() is declared as taking type "any", the
+ * parser will not do any type conversion on unknown-type literals (that
+ * is, undecorated strings or NULLs). Such values will arrive here as
+ * type UNKNOWN, which fortunately does not matter to us, since
+ * unknownout() works fine.
+ */
+
+ if (PG_ARGISNULL(1))
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("field name must not be null")));
+
+ arg = PG_GETARG_DATUM(1);
+
+ datum_to_json(arg, false, state->str, state->key_category,
+ state->key_output_func, true);
+
+ appendStringInfoString(state->str, " : ");
+
+ if (PG_ARGISNULL(2))
+ arg = (Datum) 0;
+ else
+ arg = PG_GETARG_DATUM(2);
+
+ datum_to_json(arg, PG_ARGISNULL(2), state->str, state->val_category,
+ state->val_output_func, false);
+
+ PG_RETURN_POINTER(state);
}
/*
- * Turn a scalar Datum into JSON. Hand off a non-scalar datum to
- * composite_to_json or array_to_json_internal as appropriate.
+ * json_object_agg final function.
*/
-static inline void
-datum_to_json(Datum val, bool is_null, StringInfo result, TYPCATEGORY tcategory,
- Oid typoutputfunc)
+Datum
+json_object_agg_finalfn(PG_FUNCTION_ARGS)
{
+ JsonAggState *state;
- char *outputstr;
+ /* cannot be called directly because of internal-type argument */
+ Assert(AggCheckCallContext(fcinfo, NULL));
- if (is_null)
- {
- appendStringInfoString(result, "null");
- return;
- }
+ state = PG_ARGISNULL(0) ? NULL : (JsonAggState *) PG_GETARG_POINTER(0);
- switch (tcategory)
- {
- case TYPCATEGORY_ARRAY:
- array_to_json_internal(val, result, false);
- break;
- case TYPCATEGORY_COMPOSITE:
- composite_to_json(val, result, false);
- break;
- case TYPCATEGORY_BOOLEAN:
- if (DatumGetBool(val))
- appendStringInfoString(result, "true");
- else
- appendStringInfoString(result, "false");
- break;
- case TYPCATEGORY_NUMERIC:
- outputstr = OidOutputFunctionCall(typoutputfunc, val);
+ /* NULL result for no rows in, as is standard with aggregates */
+ if (state == NULL)
+ PG_RETURN_NULL();
- /*
- * Don't call escape_json here if it's a valid JSON number.
- * Numeric output should usually be a valid JSON number and JSON
- * numbers shouldn't be quoted. Quote cases like "Nan" and
- * "Infinity", however.
- */
- if (strpbrk(outputstr, NON_NUMERIC_LETTER) == NULL)
- appendStringInfoString(result, outputstr);
- else
- escape_json(result, outputstr);
- pfree(outputstr);
- break;
- case TYPCATEGORY_JSON:
- /* JSON will already be escaped */
- outputstr = OidOutputFunctionCall(typoutputfunc, val);
- appendStringInfoString(result, outputstr);
- pfree(outputstr);
- break;
- default:
- outputstr = OidOutputFunctionCall(typoutputfunc, val);
- escape_json(result, outputstr);
- pfree(outputstr);
- }
+ /* Else return state with appropriate object terminator added */
+ PG_RETURN_TEXT_P(catenate_stringinfo_string(state->str, " }"));
}
/*
- * Process a single dimension of an array.
- * If it's the innermost dimension, output the values, otherwise call
- * ourselves recursively to process the next dimension.
+ * Helper function for aggregates: return given StringInfo's contents plus
+ * specified trailing string, as a text datum. We need this because aggregate
+ * final functions are not allowed to modify the aggregate state.
*/
-static void
-array_dim_to_json(StringInfo result, int dim, int ndims, int *dims, Datum *vals,
- bool *nulls, int *valcount, TYPCATEGORY tcategory,
- Oid typoutputfunc, bool use_line_feeds)
+static text *
+catenate_stringinfo_string(StringInfo buffer, const char *addon)
{
+ /* custom version of cstring_to_text_with_len */
+ int buflen = buffer->len;
+ int addlen = strlen(addon);
+ text *result = (text *) palloc(buflen + addlen + VARHDRSZ);
+
+ SET_VARSIZE(result, buflen + addlen + VARHDRSZ);
+ memcpy(VARDATA(result), buffer->data, buflen);
+ memcpy(VARDATA(result) + buflen, addon, addlen);
+ return result;
+}
+
+/*
+ * SQL function json_build_object(variadic "any")
+ */
+Datum
+json_build_object(PG_FUNCTION_ARGS)
+{
+ int nargs = PG_NARGS();
int i;
- char *sep;
+ const char *sep = "";
+ StringInfo result;
+ Datum *args;
+ bool *nulls;
+ Oid *types;
- Assert(dim < ndims);
+ /* fetch argument values to build the object */
+ nargs = extract_variadic_args(fcinfo, 0, false, &args, &types, &nulls);
- sep = use_line_feeds ? ",\n " : ",";
+ if (nargs < 0)
+ PG_RETURN_NULL();
- appendStringInfoChar(result, '[');
+ if (nargs % 2 != 0)
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("argument list must have even number of elements"),
+ /* translator: %s is a SQL function name */
+ errhint("The arguments of %s must consist of alternating keys and values.",
+ "json_build_object()")));
- for (i = 1; i <= dims[dim]; i++)
+ result = makeStringInfo();
+
+ appendStringInfoChar(result, '{');
+
+ for (i = 0; i < nargs; i += 2)
{
- if (i > 1)
- appendStringInfoString(result, sep);
+ appendStringInfoString(result, sep);
+ sep = ", ";
- if (dim + 1 == ndims)
- {
- datum_to_json(vals[*valcount], nulls[*valcount], result, tcategory,
- typoutputfunc);
- (*valcount)++;
- }
- else
- {
- /*
- * Do we want line feeds on inner dimensions of arrays? For now
- * we'll say no.
- */
- array_dim_to_json(result, dim + 1, ndims, dims, vals, nulls,
- valcount, tcategory, typoutputfunc, false);
- }
+ /* process key */
+ if (nulls[i])
+ ereport(ERROR,
+ (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
+ errmsg("argument %d cannot be null", i + 1),
+ errhint("Object keys should be text.")));
+
+ add_json(args[i], false, result, types[i], true);
+
+ appendStringInfoString(result, " : ");
+
+ /* process value */
+ add_json(args[i + 1], nulls[i + 1], result, types[i + 1], false);
}
- appendStringInfoChar(result, ']');
+ appendStringInfoChar(result, '}');
+
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
}
/*
- * Turn an array into JSON.
+ * degenerate case of json_build_object where it gets 0 arguments.
*/
-static void
-array_to_json_internal(Datum array, StringInfo result, bool use_line_feeds)
+Datum
+json_build_object_noargs(PG_FUNCTION_ARGS)
{
- ArrayType *v = DatumGetArrayTypeP(array);
- Oid element_type = ARR_ELEMTYPE(v);
- int *dim;
- int ndim;
- int nitems;
- int count = 0;
- Datum *elements;
- bool *nulls;
+ PG_RETURN_TEXT_P(cstring_to_text_with_len("{}", 2));
+}
- int16 typlen;
- bool typbyval;
- char typalign,
- typdelim;
- Oid typioparam;
- Oid typoutputfunc;
- TYPCATEGORY tcategory;
+/*
+ * SQL function json_build_array(variadic "any")
+ */
+Datum
+json_build_array(PG_FUNCTION_ARGS)
+{
+ int nargs;
+ int i;
+ const char *sep = "";
+ StringInfo result;
+ Datum *args;
+ bool *nulls;
+ Oid *types;
- ndim = ARR_NDIM(v);
- dim = ARR_DIMS(v);
- nitems = ArrayGetNItems(ndim, dim);
+ /* fetch argument values to build the array */
+ nargs = extract_variadic_args(fcinfo, 0, false, &args, &types, &nulls);
- if (nitems <= 0)
- {
- appendStringInfoString(result, "[]");
- return;
- }
+ if (nargs < 0)
+ PG_RETURN_NULL();
- get_type_io_data(element_type, IOFunc_output,
- &typlen, &typbyval, &typalign,
- &typdelim, &typioparam, &typoutputfunc);
+ result = makeStringInfo();
- deconstruct_array(v, element_type, typlen, typbyval,
- typalign, &elements, &nulls,
- &nitems);
+ appendStringInfoChar(result, '[');
- if (element_type == RECORDOID)
- tcategory = TYPCATEGORY_COMPOSITE;
- else if (element_type == JSONOID)
- tcategory = TYPCATEGORY_JSON;
- else
- tcategory = TypeCategory(element_type);
+ for (i = 0; i < nargs; i++)
+ {
+ appendStringInfoString(result, sep);
+ sep = ", ";
+ add_json(args[i], nulls[i], result, types[i], false);
+ }
- array_dim_to_json(result, 0, ndim, dim, elements, nulls, &count, tcategory,
- typoutputfunc, use_line_feeds);
+ appendStringInfoChar(result, ']');
- pfree(elements);
- pfree(nulls);
+ PG_RETURN_TEXT_P(cstring_to_text_with_len(result->data, result->len));
}
/*
- * Turn a composite / record into JSON.
+ * degenerate case of json_build_array where it gets 0 arguments.
*/
-static void
-composite_to_json(Datum composite, StringInfo result, bool use_line_feeds)
+Datum
+json_build_array_noargs(PG_FUNCTION_ARGS)
{
- HeapTupleHeader td;
- Oid tupType;
- int32 tupTypmod;
- TupleDesc tupdesc;
- HeapTupleData tmptup,
- *tuple;
- int i;
- bool needsep = false;
- char *sep;
-
- sep = use_line_feeds ? ",\n " : ",";
-
- td = DatumGetHeapTupleHeader(composite);
+ PG_RETURN_TEXT_P(cstring_to_text_with_len("[]", 2));
+}
- /* Extract rowtype info and find a tupdesc */
- tupType = HeapTupleHeaderGetTypeId(td);
- tupTypmod = HeapTupleHeaderGetTypMod(td);
- tupdesc = lookup_rowtype_tupdesc(tupType, tupTypmod);
+/*
+ * SQL function json_object(text[])
+ *
+ * take a one or two dimensional array of text as key/value pairs
+ * for a json object.
+ */
+Datum
+json_object(PG_FUNCTION_ARGS)
+{
+ ArrayType *in_array = PG_GETARG_ARRAYTYPE_P(0);
+ int ndims = ARR_NDIM(in_array);
+ StringInfoData result;
+ Datum *in_datums;
+ bool *in_nulls;
+ int in_count,
+ count,
+ i;
+ text *rval;
+ char *v;
+
+ switch (ndims)
+ {
+ case 0:
+ PG_RETURN_DATUM(CStringGetTextDatum("{}"));
+ break;
- /* Build a temporary HeapTuple control structure */
- tmptup.t_len = HeapTupleHeaderGetDatumLength(td);
- tmptup.t_data = td;
- tuple = &tmptup;
+ case 1:
+ if ((ARR_DIMS(in_array)[0]) % 2)
+ ereport(ERROR,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("array must have even number of elements")));
+ break;
- appendStringInfoChar(result, '{');
+ case 2:
+ if ((ARR_DIMS(in_array)[1]) != 2)
+ ereport(ERROR,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("array must have two columns")));
+ break;
- for (i = 0; i < tupdesc->natts; i++)
- {
- Datum val,
- origval;
- bool isnull;
- char *attname;
- TYPCATEGORY tcategory;
- Oid typoutput;
- bool typisvarlena;
+ default:
+ ereport(ERROR,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("wrong number of array subscripts")));
+ }
- if (tupdesc->attrs[i]->attisdropped)
- continue;
+ deconstruct_array(in_array,
+ TEXTOID, -1, false, 'i',
+ &in_datums, &in_nulls, &in_count);
- if (needsep)
- appendStringInfoString(result, sep);
- needsep = true;
+ count = in_count / 2;
- attname = NameStr(tupdesc->attrs[i]->attname);
- escape_json(result, attname);
- appendStringInfoChar(result, ':');
+ initStringInfo(&result);
- origval = heap_getattr(tuple, i + 1, tupdesc, &isnull);
+ appendStringInfoChar(&result, '{');
- if (tupdesc->attrs[i]->atttypid == RECORDARRAYOID)
- tcategory = TYPCATEGORY_ARRAY;
- else if (tupdesc->attrs[i]->atttypid == RECORDOID)
- tcategory = TYPCATEGORY_COMPOSITE;
- else if (tupdesc->attrs[i]->atttypid == JSONOID)
- tcategory = TYPCATEGORY_JSON;
+ for (i = 0; i < count; ++i)
+ {
+ if (in_nulls[i * 2])
+ ereport(ERROR,
+ (errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
+ errmsg("null value not allowed for object key")));
+
+ v = TextDatumGetCString(in_datums[i * 2]);
+ if (i > 0)
+ appendStringInfoString(&result, ", ");
+ escape_json(&result, v);
+ appendStringInfoString(&result, " : ");
+ pfree(v);
+ if (in_nulls[i * 2 + 1])
+ appendStringInfoString(&result, "null");
else
- tcategory = TypeCategory(tupdesc->attrs[i]->atttypid);
+ {
+ v = TextDatumGetCString(in_datums[i * 2 + 1]);
+ escape_json(&result, v);
+ pfree(v);
+ }
+ }
- getTypeOutputInfo(tupdesc->attrs[i]->atttypid,
- &typoutput, &typisvarlena);
+ appendStringInfoChar(&result, '}');
- /*
- * If we have a toasted datum, forcibly detoast it here to avoid
- * memory leakage inside the type's output routine.
- */
- if (typisvarlena && !isnull)
- val = PointerGetDatum(PG_DETOAST_DATUM(origval));
- else
- val = origval;
+ pfree(in_datums);
+ pfree(in_nulls);
- datum_to_json(val, isnull, result, tcategory, typoutput);
+ rval = cstring_to_text_with_len(result.data, result.len);
+ pfree(result.data);
- /* Clean up detoasted copy, if any */
- if (val != origval)
- pfree(DatumGetPointer(val));
- }
+ PG_RETURN_TEXT_P(rval);
- appendStringInfoChar(result, '}');
- ReleaseTupleDesc(tupdesc);
}
/*
- * SQL function array_to_json(row)
+ * SQL function json_object(text[], text[])
+ *
+ * take separate key and value arrays of text to construct a json object
+ * pairwise.
*/
-extern Datum
-array_to_json(PG_FUNCTION_ARGS)
+Datum
+json_object_two_arg(PG_FUNCTION_ARGS)
{
- Datum array = PG_GETARG_DATUM(0);
- StringInfo result;
-
- result = makeStringInfo();
-
- array_to_json_internal(array, result, false);
-
- PG_RETURN_TEXT_P(cstring_to_text(result->data));
-};
+ ArrayType *key_array = PG_GETARG_ARRAYTYPE_P(0);
+ ArrayType *val_array = PG_GETARG_ARRAYTYPE_P(1);
+ int nkdims = ARR_NDIM(key_array);
+ int nvdims = ARR_NDIM(val_array);
+ StringInfoData result;
+ Datum *key_datums,
+ *val_datums;
+ bool *key_nulls,
+ *val_nulls;
+ int key_count,
+ val_count,
+ i;
+ text *rval;
+ char *v;
+
+ if (nkdims > 1 || nkdims != nvdims)
+ ereport(ERROR,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("wrong number of array subscripts")));
-/*
- * SQL function array_to_json(row, prettybool)
- */
-extern Datum
-array_to_json_pretty(PG_FUNCTION_ARGS)
-{
- Datum array = PG_GETARG_DATUM(0);
- bool use_line_feeds = PG_GETARG_BOOL(1);
- StringInfo result;
+ if (nkdims == 0)
+ PG_RETURN_DATUM(CStringGetTextDatum("{}"));
- result = makeStringInfo();
+ deconstruct_array(key_array,
+ TEXTOID, -1, false, 'i',
+ &key_datums, &key_nulls, &key_count);
- array_to_json_internal(array, result, use_line_feeds);
+ deconstruct_array(val_array,
+ TEXTOID, -1, false, 'i',
+ &val_datums, &val_nulls, &val_count);
- PG_RETURN_TEXT_P(cstring_to_text(result->data));
-};
+ if (key_count != val_count)
+ ereport(ERROR,
+ (errcode(ERRCODE_ARRAY_SUBSCRIPT_ERROR),
+ errmsg("mismatched array dimensions")));
-/*
- * SQL function row_to_json(row)
- */
-extern Datum
-row_to_json(PG_FUNCTION_ARGS)
-{
- Datum array = PG_GETARG_DATUM(0);
- StringInfo result;
+ initStringInfo(&result);
- result = makeStringInfo();
+ appendStringInfoChar(&result, '{');
- composite_to_json(array, result, false);
+ for (i = 0; i < key_count; ++i)
+ {
+ if (key_nulls[i])
+ ereport(ERROR,
+ (errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
+ errmsg("null value not allowed for object key")));
+
+ v = TextDatumGetCString(key_datums[i]);
+ if (i > 0)
+ appendStringInfoString(&result, ", ");
+ escape_json(&result, v);
+ appendStringInfoString(&result, " : ");
+ pfree(v);
+ if (val_nulls[i])
+ appendStringInfoString(&result, "null");
+ else
+ {
+ v = TextDatumGetCString(val_datums[i]);
+ escape_json(&result, v);
+ pfree(v);
+ }
+ }
- PG_RETURN_TEXT_P(cstring_to_text(result->data));
-};
+ appendStringInfoChar(&result, '}');
-/*
- * SQL function row_to_json(row, prettybool)
- */
-extern Datum
-row_to_json_pretty(PG_FUNCTION_ARGS)
-{
- Datum array = PG_GETARG_DATUM(0);
- bool use_line_feeds = PG_GETARG_BOOL(1);
- StringInfo result;
+ pfree(key_datums);
+ pfree(key_nulls);
+ pfree(val_datums);
+ pfree(val_nulls);
- result = makeStringInfo();
+ rval = cstring_to_text_with_len(result.data, result.len);
+ pfree(result.data);
- composite_to_json(array, result, use_line_feeds);
+ PG_RETURN_TEXT_P(rval);
+}
- PG_RETURN_TEXT_P(cstring_to_text(result->data));
-};
/*
* Produce a JSON string literal, properly escaping characters in the text.
{
const char *p;
- appendStringInfoCharMacro(buf, '\"');
+ appendStringInfoCharMacro(buf, '"');
for (p = str; *p; p++)
{
switch (*p)
break;
}
}
- appendStringInfoCharMacro(buf, '\"');
+ appendStringInfoCharMacro(buf, '"');
+}
+
+/*
+ * SQL function json_typeof(json) -> text
+ *
+ * Returns the type of the outermost JSON value as TEXT. Possible types are
+ * "object", "array", "string", "number", "boolean", and "null".
+ *
+ * Performs a single call to json_lex() to get the first token of the supplied
+ * value. This initial token uniquely determines the value's type. As our
+ * input must already have been validated by json_in() or json_recv(), the
+ * initial token should never be JSON_TOKEN_OBJECT_END, JSON_TOKEN_ARRAY_END,
+ * JSON_TOKEN_COLON, JSON_TOKEN_COMMA, or JSON_TOKEN_END.
+ */
+Datum
+json_typeof(PG_FUNCTION_ARGS)
+{
+ text *json;
+
+ JsonLexContext *lex;
+ JsonTokenType tok;
+ char *type;
+
+ json = PG_GETARG_TEXT_PP(0);
+ lex = makeJsonLexContext(json, false);
+
+ /* Lex exactly one token from the input and check its type. */
+ json_lex(lex);
+ tok = lex_peek(lex);
+ switch (tok)
+ {
+ case JSON_TOKEN_OBJECT_START:
+ type = "object";
+ break;
+ case JSON_TOKEN_ARRAY_START:
+ type = "array";
+ break;
+ case JSON_TOKEN_STRING:
+ type = "string";
+ break;
+ case JSON_TOKEN_NUMBER:
+ type = "number";
+ break;
+ case JSON_TOKEN_TRUE:
+ case JSON_TOKEN_FALSE:
+ type = "boolean";
+ break;
+ case JSON_TOKEN_NULL:
+ type = "null";
+ break;
+ default:
+ elog(ERROR, "unexpected json token: %d", tok);
+ }
+
+ PG_RETURN_TEXT_P(cstring_to_text(type));
}