jsonapi.c File Reference

#include "postgres.h"
#include "common/jsonapi.h"
#include "mb/pg_wchar.h"
#include "port/pg_lfind.h"
#include "miscadmin.h"

Include dependency graph for jsonapi.c:

Go to the source code of this file.

Data Structures
struct	JsonParserStack
struct	JsonIncrementalState
struct	td_entry

Macros
#define	JSON_NUM_TERMINALS   13
#define	JSON_NUM_NONTERMINALS   5
#define	JSON_NT_OFFSET   JSON_NT_JSON
#define	OFS(NT)   (NT) - JSON_NT_OFFSET
#define	IS_SEM(x)   ((x) & 0x40)
#define	IS_NT(x)   ((x) & 0x20)
#define	TD_ENTRY(PROD)   { sizeof(PROD) - 1, (PROD) }
#define	JSON_ALPHANUMERIC_CHAR(c)
#define	JS_STACK_CHUNK_SIZE   64
#define	JS_MAX_PROD_LEN   10 /* more than we need */
#define	JSON_TD_MAX_STACK
#define	FAIL_OR_INCOMPLETE_AT_CHAR_START(code)
#define	FAIL_AT_CHAR_END(code)
#define	json_token_error(lex, format)

Enumerations
enum	JsonParseContext {   JSON_PARSE_VALUE , JSON_PARSE_STRING , JSON_PARSE_ARRAY_START , JSON_PARSE_ARRAY_NEXT ,   JSON_PARSE_OBJECT_START , JSON_PARSE_OBJECT_LABEL , JSON_PARSE_OBJECT_NEXT , JSON_PARSE_OBJECT_COMMA ,   JSON_PARSE_END }
enum	JsonNonTerminal {   JSON_NT_JSON = 32 , JSON_NT_ARRAY_ELEMENTS , JSON_NT_MORE_ARRAY_ELEMENTS , JSON_NT_KEY_PAIRS ,   JSON_NT_MORE_KEY_PAIRS }
enum	JsonParserSem {   JSON_SEM_OSTART = 64 , JSON_SEM_OEND , JSON_SEM_ASTART , JSON_SEM_AEND ,   JSON_SEM_OFIELD_INIT , JSON_SEM_OFIELD_START , JSON_SEM_OFIELD_END , JSON_SEM_AELEM_START ,   JSON_SEM_AELEM_END , JSON_SEM_SCALAR_INIT , JSON_SEM_SCALAR_CALL }

Functions
static JsonParseErrorType	json_lex_string (JsonLexContext *lex)
static JsonParseErrorType	json_lex_number (JsonLexContext *lex, const char *s, bool *num_err, size_t *total_len)
static JsonParseErrorType	parse_scalar (JsonLexContext *lex, const JsonSemAction *sem)
static JsonParseErrorType	parse_object_field (JsonLexContext *lex, const JsonSemAction *sem)
static JsonParseErrorType	parse_object (JsonLexContext *lex, const JsonSemAction *sem)
static JsonParseErrorType	parse_array_element (JsonLexContext *lex, const JsonSemAction *sem)
static JsonParseErrorType	parse_array (JsonLexContext *lex, const JsonSemAction *sem)
static JsonParseErrorType	report_parse_error (JsonParseContext ctx, JsonLexContext *lex)
static JsonTokenType	lex_peek (JsonLexContext *lex)
static JsonParseErrorType	lex_expect (JsonParseContext ctx, JsonLexContext *lex, JsonTokenType token)
bool	IsValidJsonNumber (const char *str, size_t len)
JsonLexContext *	makeJsonLexContextCstringLen (JsonLexContext *lex, const char *json, size_t len, int encoding, bool need_escapes)
JsonLexContext *	makeJsonLexContextIncremental (JsonLexContext *lex, int encoding, bool need_escapes)
static void	inc_lex_level (JsonLexContext *lex)
static void	dec_lex_level (JsonLexContext *lex)
static void	push_prediction (JsonParserStack *pstack, td_entry entry)
static char	pop_prediction (JsonParserStack *pstack)
static char	next_prediction (JsonParserStack *pstack)
static bool	have_prediction (JsonParserStack *pstack)
static void	set_fname (JsonLexContext *lex, char *fname)
static char *	get_fname (JsonLexContext *lex)
static void	set_fnull (JsonLexContext *lex, bool fnull)
static bool	get_fnull (JsonLexContext *lex)
void	freeJsonLexContext (JsonLexContext *lex)
JsonParseErrorType	pg_parse_json (JsonLexContext *lex, const JsonSemAction *sem)
JsonParseErrorType	json_count_array_elements (JsonLexContext *lex, int *elements)
JsonParseErrorType	pg_parse_json_incremental (JsonLexContext *lex, const JsonSemAction *sem, const char *json, size_t len, bool is_last)
JsonParseErrorType	json_lex (JsonLexContext *lex)
char *	json_errdetail (JsonParseErrorType error, JsonLexContext *lex)

Variables
static char	JSON_PROD_EPSILON [] = {0}
static char	JSON_PROD_SCALAR_STRING [] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_STRING, JSON_SEM_SCALAR_INIT, 0}
static char	JSON_PROD_SCALAR_NUMBER [] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_NUMBER, JSON_SEM_SCALAR_INIT, 0}
static char	JSON_PROD_SCALAR_TRUE [] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_TRUE, JSON_SEM_SCALAR_INIT, 0}
static char	JSON_PROD_SCALAR_FALSE [] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_FALSE, JSON_SEM_SCALAR_INIT, 0}
static char	JSON_PROD_SCALAR_NULL [] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_NULL, JSON_SEM_SCALAR_INIT, 0}
static char	JSON_PROD_OBJECT [] = {JSON_SEM_OEND, JSON_TOKEN_OBJECT_END, JSON_NT_KEY_PAIRS, JSON_TOKEN_OBJECT_START, JSON_SEM_OSTART, 0}
static char	JSON_PROD_ARRAY [] = {JSON_SEM_AEND, JSON_TOKEN_ARRAY_END, JSON_NT_ARRAY_ELEMENTS, JSON_TOKEN_ARRAY_START, JSON_SEM_ASTART, 0}
static char	JSON_PROD_ARRAY_ELEMENTS [] = {JSON_NT_MORE_ARRAY_ELEMENTS, JSON_SEM_AELEM_END, JSON_NT_JSON, JSON_SEM_AELEM_START, 0}
static char	JSON_PROD_MORE_ARRAY_ELEMENTS [] = {JSON_NT_MORE_ARRAY_ELEMENTS, JSON_SEM_AELEM_END, JSON_NT_JSON, JSON_SEM_AELEM_START, JSON_TOKEN_COMMA, 0}
static char	JSON_PROD_KEY_PAIRS [] = {JSON_NT_MORE_KEY_PAIRS, JSON_SEM_OFIELD_END, JSON_NT_JSON, JSON_SEM_OFIELD_START, JSON_TOKEN_COLON, JSON_TOKEN_STRING, JSON_SEM_OFIELD_INIT, 0}
static char	JSON_PROD_MORE_KEY_PAIRS [] = {JSON_NT_MORE_KEY_PAIRS, JSON_SEM_OFIELD_END, JSON_NT_JSON, JSON_SEM_OFIELD_START, JSON_TOKEN_COLON, JSON_TOKEN_STRING, JSON_SEM_OFIELD_INIT, JSON_TOKEN_COMMA, 0}
static td_entry	td_parser_table [JSON_NUM_NONTERMINALS][JSON_NUM_TERMINALS]
static char	JSON_PROD_GOAL [] = {JSON_TOKEN_END, JSON_NT_JSON, 0}
const JsonSemAction	nullSemAction

Macro Definition Documentation

FAIL_AT_CHAR_END

#define FAIL_AT_CHAR_END

(

code

)

Value:

    do { \
        const char     *term = s + pg_encoding_mblen(lex->input_encoding, s); \
        lex->token_terminator = (term <= end) ? term : end; \
        return code; \
    } while (0)

FAIL_OR_INCOMPLETE_AT_CHAR_START

#define FAIL_OR_INCOMPLETE_AT_CHAR_START

(

code

)

Value:

    do { \
        if (lex->incremental && !lex->inc_state->is_last_chunk) \
        { \
            appendBinaryStringInfo(&lex->inc_state->partial_token, \
                                   lex->token_start, end - lex->token_start); \
            return JSON_INCOMPLETE; \
        } \
        lex->token_terminator = s; \
        return code; \
    } while (0)

IS_NT

#define IS_NT

(

x

)

((x) & 0x20)

Definition at line 119 of file jsonapi.c.

IS_SEM

#define IS_SEM

(

x

)

((x) & 0x40)

Definition at line 118 of file jsonapi.c.

JS_MAX_PROD_LEN

#define JS_MAX_PROD_LEN   10 /* more than we need */

Definition at line 362 of file jsonapi.c.

JS_STACK_CHUNK_SIZE

#define JS_STACK_CHUNK_SIZE   64

Definition at line 361 of file jsonapi.c.

JSON_ALPHANUMERIC_CHAR

#define JSON_ALPHANUMERIC_CHAR

(

c

)

Value:

    (((c) >= 'a' && (c) <= 'z') || \
     ((c) >= 'A' && (c) <= 'Z') || \
     ((c) >= '0' && (c) <= '9') || \
     (c) == '_' || \
     IS_HIGHBIT_SET(c))

Definition at line 259 of file jsonapi.c.

JSON_NT_OFFSET

#define JSON_NT_OFFSET   JSON_NT_JSON

Definition at line 114 of file jsonapi.c.

JSON_NUM_NONTERMINALS

#define JSON_NUM_NONTERMINALS   5

Definition at line 113 of file jsonapi.c.

JSON_NUM_TERMINALS

#define JSON_NUM_TERMINALS   13

Definition at line 112 of file jsonapi.c.

JSON_TD_MAX_STACK

#define JSON_TD_MAX_STACK

Value:

                                 6400   /* hard coded for now - this is a REALLY high
                                 * number */

Definition at line 363 of file jsonapi.c.

json_token_error

#define json_token_error	(		lex,
			format
	)

Value:

    appendStringInfo((lex)->errormsg, _(format), \
                     (int) ((lex)->token_terminator - (lex)->token_start), \
                     (lex)->token_start);

OFS

#define OFS

(

NT

)

(NT) - JSON_NT_OFFSET

Definition at line 116 of file jsonapi.c.

TD_ENTRY

#define TD_ENTRY

(

PROD

)

{ sizeof(PROD) - 1, (PROD) }

Definition at line 178 of file jsonapi.c.

Enumeration Type Documentation

JsonNonTerminal

enum JsonNonTerminal

Enumerator
JSON_NT_JSON
JSON_NT_ARRAY_ELEMENTS
JSON_NT_MORE_ARRAY_ELEMENTS
JSON_NT_KEY_PAIRS
JSON_NT_MORE_KEY_PAIRS

Definition at line 53 of file jsonapi.c.

{
    JSON_NT_JSON = 32,
    JSON_NT_ARRAY_ELEMENTS,
    JSON_NT_MORE_ARRAY_ELEMENTS,
    JSON_NT_KEY_PAIRS,
    JSON_NT_MORE_KEY_PAIRS,
};

JsonParseContext

enum JsonParseContext

Enumerator
JSON_PARSE_VALUE
JSON_PARSE_STRING
JSON_PARSE_ARRAY_START
JSON_PARSE_ARRAY_NEXT
JSON_PARSE_OBJECT_START
JSON_PARSE_OBJECT_LABEL
JSON_PARSE_OBJECT_NEXT
JSON_PARSE_OBJECT_COMMA
JSON_PARSE_END

Definition at line 33 of file jsonapi.c.

{
    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 */
    JSON_PARSE_END,             /* saw the end of a document, expect nothing */
} JsonParseContext;

JsonParserSem

enum JsonParserSem

Enumerator
JSON_SEM_OSTART
JSON_SEM_OEND
JSON_SEM_ASTART
JSON_SEM_AEND
JSON_SEM_OFIELD_INIT
JSON_SEM_OFIELD_START
JSON_SEM_OFIELD_END
JSON_SEM_AELEM_START
JSON_SEM_AELEM_END
JSON_SEM_SCALAR_INIT
JSON_SEM_SCALAR_CALL

Definition at line 62 of file jsonapi.c.

{
    JSON_SEM_OSTART = 64,
    JSON_SEM_OEND,
    JSON_SEM_ASTART,
    JSON_SEM_AEND,
    JSON_SEM_OFIELD_INIT,
    JSON_SEM_OFIELD_START,
    JSON_SEM_OFIELD_END,
    JSON_SEM_AELEM_START,
    JSON_SEM_AELEM_END,
    JSON_SEM_SCALAR_INIT,
    JSON_SEM_SCALAR_CALL,
};

Function Documentation

dec_lex_level()

static void dec_lex_level ( JsonLexContext *  lex )

inlinestatic

Definition at line 418 of file jsonapi.c.

{
    lex->lex_level -= 1;

References JsonLexContext::lex_level.

Referenced by pg_parse_json_incremental().

freeJsonLexContext()

void freeJsonLexContext

(

JsonLexContext *

lex

)

Definition at line 482 of file jsonapi.c.

{
    if (lex->flags & JSONLEX_FREE_STRVAL)
        destroyStringInfo(lex->strval);
 
    if (lex->errormsg)
        destroyStringInfo(lex->errormsg);
 
    if (lex->incremental)
    {
        pfree(lex->inc_state->partial_token.data);
        pfree(lex->inc_state);
        pfree(lex->pstack->prediction);
        pfree(lex->pstack->fnames);
        pfree(lex->pstack->fnull);
        pfree(lex->pstack);
    }
 
    if (lex->flags & JSONLEX_FREE_STRUCT)
        pfree(lex);

References StringInfoData::data, destroyStringInfo(), JsonLexContext::errormsg, JsonLexContext::flags, JsonParserStack::fnames, JsonParserStack::fnull, JsonLexContext::inc_state, JsonLexContext::incremental, JSONLEX_FREE_STRUCT, JSONLEX_FREE_STRVAL, JsonIncrementalState::partial_token, pfree(), JsonParserStack::prediction, JsonLexContext::pstack, and JsonLexContext::strval.

Referenced by datum_to_jsonb_internal(), each_worker(), elements_worker(), get_json_object_as_hash(), get_worker(), iterate_json_values(), json_object_keys(), json_parse_manifest(), json_parse_manifest_incremental_shutdown(), json_validate(), main(), populate_array_json(), populate_recordset_worker(), and transform_json_string_values().

get_fname()

static char* get_fname ( JsonLexContext *  lex )

inlinestatic

Definition at line 457 of file jsonapi.c.

{
    return lex->pstack->fnames[lex->lex_level];

References JsonParserStack::fnames, JsonLexContext::lex_level, and JsonLexContext::pstack.

Referenced by pg_parse_json_incremental().

get_fnull()

static bool get_fnull ( JsonLexContext *  lex )

inlinestatic

Definition at line 469 of file jsonapi.c.

{
    return lex->pstack->fnull[lex->lex_level];

References JsonParserStack::fnull, JsonLexContext::lex_level, and JsonLexContext::pstack.

Referenced by pg_parse_json_incremental().

have_prediction()

static bool have_prediction ( JsonParserStack *  pstack )

inlinestatic

Definition at line 445 of file jsonapi.c.

{
    return pstack->pred_index > 0;

References JsonParserStack::pred_index.

Referenced by pg_parse_json_incremental().

inc_lex_level()

static void inc_lex_level ( JsonLexContext *  lex )

inlinestatic

Definition at line 397 of file jsonapi.c.

{
    lex->lex_level += 1;
 
    if (lex->incremental && lex->lex_level >= lex->pstack->stack_size)
    {
        lex->pstack->stack_size += JS_STACK_CHUNK_SIZE;
        lex->pstack->prediction =
            repalloc(lex->pstack->prediction,
                     lex->pstack->stack_size * JS_MAX_PROD_LEN);
        if (lex->pstack->fnames)
            lex->pstack->fnames =
                repalloc(lex->pstack->fnames,
                         lex->pstack->stack_size * sizeof(char *));
        if (lex->pstack->fnull)
            lex->pstack->fnull =
                repalloc(lex->pstack->fnull, lex->pstack->stack_size * sizeof(bool));
    }

References JsonParserStack::fnames, JsonParserStack::fnull, JsonLexContext::incremental, JS_MAX_PROD_LEN, JS_STACK_CHUNK_SIZE, JsonLexContext::lex_level, JsonParserStack::prediction, JsonLexContext::pstack, repalloc(), and JsonParserStack::stack_size.

Referenced by pg_parse_json_incremental().

IsValidJsonNumber()

bool IsValidJsonNumber	(	const char *	str,
		size_t	len
	)

Definition at line 272 of file jsonapi.c.

{
    bool        numeric_error;
    size_t      total_len;
    JsonLexContext dummy_lex;
 
    if (len <= 0)
        return false;
 
    dummy_lex.incremental = false;
    dummy_lex.inc_state = NULL;
    dummy_lex.pstack = NULL;
 
    /*
     * 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 = str + 1;
        dummy_lex.input_length = len - 1;
    }
    else
    {
        dummy_lex.input = str;
        dummy_lex.input_length = len;
    }
 
    dummy_lex.token_start = dummy_lex.input;
 
    json_lex_number(&dummy_lex, dummy_lex.input, &numeric_error, &total_len);
 
    return (!numeric_error) && (total_len == dummy_lex.input_length);
}

References JsonLexContext::inc_state, JsonLexContext::incremental, JsonLexContext::input, JsonLexContext::input_length, json_lex_number(), len, JsonLexContext::pstack, str, and JsonLexContext::token_start.

Referenced by hstore_to_json_loose(), and hstore_to_jsonb_loose().

json_count_array_elements()

JsonParseErrorType json_count_array_elements	(	JsonLexContext *	lex,
		int *	elements
	)

Definition at line 587 of file jsonapi.c.

{
    JsonLexContext copylex;
    int         count;
    JsonParseErrorType result;
 
    /*
     * 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(&copylex, lex, sizeof(JsonLexContext));
    copylex.strval = NULL;      /* not interested in values here */
    copylex.lex_level++;
 
    count = 0;
    result = lex_expect(JSON_PARSE_ARRAY_START, &copylex,
                        JSON_TOKEN_ARRAY_START);
    if (result != JSON_SUCCESS)
        return result;
    if (lex_peek(&copylex) != JSON_TOKEN_ARRAY_END)
    {
        while (1)
        {
            count++;
            result = parse_array_element(&copylex, &nullSemAction);
            if (result != JSON_SUCCESS)
                return result;
            if (copylex.token_type != JSON_TOKEN_COMMA)
                break;
            result = json_lex(&copylex);
            if (result != JSON_SUCCESS)
                return result;
        }
    }
    result = lex_expect(JSON_PARSE_ARRAY_NEXT, &copylex,
                        JSON_TOKEN_ARRAY_END);
    if (result != JSON_SUCCESS)
        return result;
 
    *elements = count;
    return JSON_SUCCESS;

References json_lex(), JSON_PARSE_ARRAY_NEXT, JSON_PARSE_ARRAY_START, JSON_SUCCESS, JSON_TOKEN_ARRAY_END, JSON_TOKEN_ARRAY_START, JSON_TOKEN_COMMA, lex_expect(), JsonLexContext::lex_level, lex_peek(), nullSemAction, parse_array_element(), JsonLexContext::strval, and JsonLexContext::token_type.

Referenced by get_array_start().

json_errdetail()

char* json_errdetail	(	JsonParseErrorType	error,
		JsonLexContext *	lex
	)

Definition at line 2096 of file jsonapi.c.

{
    if (lex->errormsg)
        resetStringInfo(lex->errormsg);
    else
        lex->errormsg = makeStringInfo();
 
    /*
     * A helper for error messages that should print the current token. The
     * format must contain exactly one %.*s specifier.
     */
#define json_token_error(lex, format) \
    appendStringInfo((lex)->errormsg, _(format), \
                     (int) ((lex)->token_terminator - (lex)->token_start), \
                     (lex)->token_start);
 
    switch (error)
    {
        case JSON_INCOMPLETE:
        case JSON_SUCCESS:
            /* fall through to the error code after switch */
            break;
        case JSON_INVALID_LEXER_TYPE:
            if (lex->incremental)
                return _("Recursive descent parser cannot use incremental lexer.");
            else
                return _("Incremental parser requires incremental lexer.");
        case JSON_NESTING_TOO_DEEP:
            return (_("JSON nested too deep, maximum permitted depth is 6400."));
        case JSON_ESCAPING_INVALID:
            json_token_error(lex, "Escape sequence \"\\%.*s\" is invalid.");
            break;
        case JSON_ESCAPING_REQUIRED:
            appendStringInfo(lex->errormsg,
                             _("Character with value 0x%02x must be escaped."),
                             (unsigned char) *(lex->token_terminator));
            break;
        case JSON_EXPECTED_END:
            json_token_error(lex, "Expected end of input, but found \"%.*s\".");
            break;
        case JSON_EXPECTED_ARRAY_FIRST:
            json_token_error(lex, "Expected array element or \"]\", but found \"%.*s\".");
            break;
        case JSON_EXPECTED_ARRAY_NEXT:
            json_token_error(lex, "Expected \",\" or \"]\", but found \"%.*s\".");
            break;
        case JSON_EXPECTED_COLON:
            json_token_error(lex, "Expected \":\", but found \"%.*s\".");
            break;
        case JSON_EXPECTED_JSON:
            json_token_error(lex, "Expected JSON value, but found \"%.*s\".");
            break;
        case JSON_EXPECTED_MORE:
            return _("The input string ended unexpectedly.");
        case JSON_EXPECTED_OBJECT_FIRST:
            json_token_error(lex, "Expected string or \"}\", but found \"%.*s\".");
            break;
        case JSON_EXPECTED_OBJECT_NEXT:
            json_token_error(lex, "Expected \",\" or \"}\", but found \"%.*s\".");
            break;
        case JSON_EXPECTED_STRING:
            json_token_error(lex, "Expected string, but found \"%.*s\".");
            break;
        case JSON_INVALID_TOKEN:
            json_token_error(lex, "Token \"%.*s\" is invalid.");
            break;
        case JSON_UNICODE_CODE_POINT_ZERO:
            return _("\\u0000 cannot be converted to text.");
        case JSON_UNICODE_ESCAPE_FORMAT:
            return _("\"\\u\" must be followed by four hexadecimal digits.");
        case JSON_UNICODE_HIGH_ESCAPE:
            /* note: this case is only reachable in frontend not backend */
            return _("Unicode escape values cannot be used for code point values above 007F when the encoding is not UTF8.");
        case JSON_UNICODE_UNTRANSLATABLE:
 
            /*
             * Note: this case is only reachable in backend and not frontend.
             * #ifdef it away so the frontend doesn't try to link against
             * backend functionality.
             */
#ifndef FRONTEND
            return psprintf(_("Unicode escape value could not be translated to the server's encoding %s."),
                            GetDatabaseEncodingName());
#else
            Assert(false);
            break;
#endif
        case JSON_UNICODE_HIGH_SURROGATE:
            return _("Unicode high surrogate must not follow a high surrogate.");
        case JSON_UNICODE_LOW_SURROGATE:
            return _("Unicode low surrogate must follow a high surrogate.");
        case JSON_SEM_ACTION_FAILED:
            /* fall through to the error code after switch */
            break;
    }
#undef json_token_error
 
    /*
     * We don't use a default: case, so that the compiler will warn about
     * unhandled enum values.  But this needs to be here anyway to cover the
     * possibility of an incorrect input.
     */
    if (lex->errormsg->len == 0)
        appendStringInfo(lex->errormsg,
                         "unexpected json parse error type: %d",
                         (int) error);
 
    return lex->errormsg->data;

References _, appendStringInfo(), Assert, StringInfoData::data, error(), JsonLexContext::errormsg, GetDatabaseEncodingName(), JsonLexContext::incremental, JSON_ESCAPING_INVALID, JSON_ESCAPING_REQUIRED, JSON_EXPECTED_ARRAY_FIRST, JSON_EXPECTED_ARRAY_NEXT, JSON_EXPECTED_COLON, JSON_EXPECTED_END, JSON_EXPECTED_JSON, JSON_EXPECTED_MORE, JSON_EXPECTED_OBJECT_FIRST, JSON_EXPECTED_OBJECT_NEXT, JSON_EXPECTED_STRING, JSON_INCOMPLETE, JSON_INVALID_LEXER_TYPE, JSON_INVALID_TOKEN, JSON_NESTING_TOO_DEEP, JSON_SEM_ACTION_FAILED, JSON_SUCCESS, json_token_error, JSON_UNICODE_CODE_POINT_ZERO, JSON_UNICODE_ESCAPE_FORMAT, JSON_UNICODE_HIGH_ESCAPE, JSON_UNICODE_HIGH_SURROGATE, JSON_UNICODE_LOW_SURROGATE, JSON_UNICODE_UNTRANSLATABLE, StringInfoData::len, makeStringInfo(), psprintf(), resetStringInfo(), and JsonLexContext::token_terminator.

Referenced by json_errsave_error(), json_parse_manifest(), json_parse_manifest_incremental_chunk(), and main().

json_lex()

JsonParseErrorType json_lex

(

JsonLexContext *

lex

)

Definition at line 1308 of file jsonapi.c.

{
    const char *s;
    const char *const end = lex->input + lex->input_length;
    JsonParseErrorType result;
 
    if (lex->incremental && lex->inc_state->partial_completed)
    {
        /*
         * We just lexed a completed partial token on the last call, so reset
         * everything
         */
        resetStringInfo(&(lex->inc_state->partial_token));
        lex->token_terminator = lex->input;
        lex->inc_state->partial_completed = false;
    }
 
    s = lex->token_terminator;
 
    if (lex->incremental && lex->inc_state->partial_token.len)
    {
        /*
         * We have a partial token. Extend it and if completed lex it by a
         * recursive call
         */
        StringInfo  ptok = &(lex->inc_state->partial_token);
        size_t      added = 0;
        bool        tok_done = false;
        JsonLexContext dummy_lex;
        JsonParseErrorType partial_result;
 
        if (ptok->data[0] == '"')
        {
            /*
             * It's a string. Accumulate characters until we reach an
             * unescaped '"'.
             */
            int         escapes = 0;
 
            for (int i = ptok->len - 1; i > 0; i--)
            {
                /* count the trailing backslashes on the partial token */
                if (ptok->data[i] == '\\')
                    escapes++;
                else
                    break;
            }
 
            for (size_t i = 0; i < lex->input_length; i++)
            {
                char        c = lex->input[i];
 
                appendStringInfoCharMacro(ptok, c);
                added++;
                if (c == '"' && escapes % 2 == 0)
                {
                    tok_done = true;
                    break;
                }
                if (c == '\\')
                    escapes++;
                else
                    escapes = 0;
            }
        }
        else
        {
            /* not a string */
            char        c = ptok->data[0];
 
            if (c == '-' || (c >= '0' && c <= '9'))
            {
                /* for numbers look for possible numeric continuations */
 
                bool        numend = false;
 
                for (size_t i = 0; i < lex->input_length && !numend; i++)
                {
                    char        cc = lex->input[i];
 
                    switch (cc)
                    {
                        case '+':
                        case '-':
                        case 'e':
                        case 'E':
                        case '0':
                        case '1':
                        case '2':
                        case '3':
                        case '4':
                        case '5':
                        case '6':
                        case '7':
                        case '8':
                        case '9':
                            {
                                appendStringInfoCharMacro(ptok, cc);
                                added++;
                            }
                            break;
                        default:
                            numend = true;
                    }
                }
            }
 
            /*
             * Add any remaining alphanumeric chars. This takes care of the
             * {null, false, true} literals as well as any trailing
             * alphanumeric junk on non-string tokens.
             */
            for (size_t i = added; i < lex->input_length; i++)
            {
                char        cc = lex->input[i];
 
                if (JSON_ALPHANUMERIC_CHAR(cc))
                {
                    appendStringInfoCharMacro(ptok, cc);
                    added++;
                }
                else
                {
                    tok_done = true;
                    break;
                }
            }
            if (added == lex->input_length &&
                lex->inc_state->is_last_chunk)
            {
                tok_done = true;
            }
        }
 
        if (!tok_done)
        {
            /* We should have consumed the whole chunk in this case. */
            Assert(added == lex->input_length);
 
            if (!lex->inc_state->is_last_chunk)
                return JSON_INCOMPLETE;
 
            /* json_errdetail() needs access to the accumulated token. */
            lex->token_start = ptok->data;
            lex->token_terminator = ptok->data + ptok->len;
            return JSON_INVALID_TOKEN;
        }
 
        /*
         * Everything up to lex->input[added] has been added to the partial
         * token, so move the input past it.
         */
        lex->input += added;
        lex->input_length -= added;
 
        dummy_lex.input = dummy_lex.token_terminator =
            dummy_lex.line_start = ptok->data;
        dummy_lex.line_number = lex->line_number;
        dummy_lex.input_length = ptok->len;
        dummy_lex.input_encoding = lex->input_encoding;
        dummy_lex.incremental = false;
        dummy_lex.strval = lex->strval;
 
        partial_result = json_lex(&dummy_lex);
 
        /*
         * We either have a complete token or an error. In either case we need
         * to point to the partial token data for the semantic or error
         * routines. If it's not an error we'll readjust on the next call to
         * json_lex.
         */
        lex->token_type = dummy_lex.token_type;
        lex->line_number = dummy_lex.line_number;
 
        /*
         * We know the prev_token_terminator must be back in some previous
         * piece of input, so we just make it NULL.
         */
        lex->prev_token_terminator = NULL;
 
        /*
         * Normally token_start would be ptok->data, but it could be later,
         * see json_lex_string's handling of invalid escapes.
         */
        lex->token_start = dummy_lex.token_start;
        lex->token_terminator = dummy_lex.token_terminator;
        if (partial_result == JSON_SUCCESS)
        {
            /* make sure we've used all the input */
            if (lex->token_terminator - lex->token_start != ptok->len)
            {
                Assert(false);
                return JSON_INVALID_TOKEN;
            }
 
            lex->inc_state->partial_completed = true;
        }
        return partial_result;
        /* end of partial token processing */
    }
 
    /* Skip leading whitespace. */
    while (s < end && (*s == ' ' || *s == '\t' || *s == '\n' || *s == '\r'))
    {
        if (*s++ == '\n')
        {
            ++lex->line_number;
            lex->line_start = s;
        }
    }
    lex->token_start = s;
 
    /* Determine token type. */
    if (s >= end)
    {
        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 */
                result = json_lex_string(lex);
                if (result != JSON_SUCCESS)
                    return result;
                lex->token_type = JSON_TOKEN_STRING;
                break;
            case '-':
                /* Negative number. */
                result = json_lex_number(lex, s + 1, NULL, NULL);
                if (result != JSON_SUCCESS)
                    return result;
                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. */
                result = json_lex_number(lex, s, NULL, NULL);
                if (result != JSON_SUCCESS)
                    return result;
                lex->token_type = JSON_TOKEN_NUMBER;
                break;
            default:
                {
                    const 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 < end && 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;
                        return JSON_INVALID_TOKEN;
                    }
 
                    if (lex->incremental && !lex->inc_state->is_last_chunk &&
                        p == lex->input + lex->input_length)
                    {
                        appendBinaryStringInfo(
                                               &(lex->inc_state->partial_token), s, end - s);
                        return JSON_INCOMPLETE;
                    }
 
                    /*
                     * 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
                            return JSON_INVALID_TOKEN;
                    }
                    else if (p - s == 5 && memcmp(s, "false", 5) == 0)
                        lex->token_type = JSON_TOKEN_FALSE;
                    else
                        return JSON_INVALID_TOKEN;
                }
        }                       /* end of switch */
    }
 
    if (lex->incremental && lex->token_type == JSON_TOKEN_END && !lex->inc_state->is_last_chunk)
        return JSON_INCOMPLETE;
    else
        return JSON_SUCCESS;

References appendBinaryStringInfo(), appendStringInfoCharMacro, Assert, StringInfoData::data, i, JsonLexContext::inc_state, JsonLexContext::incremental, JsonLexContext::input, JsonLexContext::input_encoding, JsonLexContext::input_length, JsonIncrementalState::is_last_chunk, JSON_ALPHANUMERIC_CHAR, JSON_INCOMPLETE, JSON_INVALID_TOKEN, json_lex_number(), json_lex_string(), JSON_SUCCESS, JSON_TOKEN_ARRAY_END, JSON_TOKEN_ARRAY_START, JSON_TOKEN_COLON, JSON_TOKEN_COMMA, JSON_TOKEN_END, JSON_TOKEN_FALSE, JSON_TOKEN_NULL, JSON_TOKEN_NUMBER, JSON_TOKEN_OBJECT_END, JSON_TOKEN_OBJECT_START, JSON_TOKEN_STRING, JSON_TOKEN_TRUE, StringInfoData::len, JsonLexContext::line_number, JsonLexContext::line_start, JsonIncrementalState::partial_completed, JsonIncrementalState::partial_token, JsonLexContext::prev_token_terminator, resetStringInfo(), JsonLexContext::strval, JsonLexContext::token_start, JsonLexContext::token_terminator, and JsonLexContext::token_type.

Referenced by json_count_array_elements(), json_get_first_token(), json_typeof(), lex_expect(), parse_array(), parse_object(), parse_object_field(), parse_scalar(), pg_parse_json(), and pg_parse_json_incremental().

json_lex_number()

static JsonParseErrorType json_lex_number ( JsonLexContext *  lex, const char *  s, bool *  num_err, size_t *  total_len  )

inlinestatic

Definition at line 1942 of file jsonapi.c.

{
    bool        error = false;
    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 (len < lex->input_length && *s == '0')
    {
        s++;
        len++;
    }
    else if (len < lex->input_length && *s >= '1' && *s <= '9')
    {
        do
        {
            s++;
            len++;
        } while (len < lex->input_length && *s >= '0' && *s <= '9');
    }
    else
        error = true;
 
    /* Part (3): parse optional decimal portion. */
    if (len < lex->input_length && *s == '.')
    {
        s++;
        len++;
        if (len == lex->input_length || *s < '0' || *s > '9')
            error = true;
        else
        {
            do
            {
                s++;
                len++;
            } while (len < lex->input_length && *s >= '0' && *s <= '9');
        }
    }
 
    /* Part (4): parse optional exponent. */
    if (len < lex->input_length && (*s == 'e' || *s == 'E'))
    {
        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++;
                len++;
            } while (len < lex->input_length && *s >= '0' && *s <= '9');
        }
    }
 
    /*
     * 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 (lex->incremental && !lex->inc_state->is_last_chunk &&
        len >= lex->input_length)
    {
        appendBinaryStringInfo(&lex->inc_state->partial_token,
                               lex->token_start, s - lex->token_start);
        if (num_err != NULL)
            *num_err = error;
 
        return JSON_INCOMPLETE;
    }
    else 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)
            return JSON_INVALID_TOKEN;
    }
 
    return JSON_SUCCESS;

References appendBinaryStringInfo(), error(), JsonLexContext::inc_state, JsonLexContext::incremental, JsonLexContext::input, JsonLexContext::input_length, JsonIncrementalState::is_last_chunk, JSON_ALPHANUMERIC_CHAR, JSON_INCOMPLETE, JSON_INVALID_TOKEN, JSON_SUCCESS, len, JsonIncrementalState::partial_token, JsonLexContext::prev_token_terminator, JsonLexContext::token_start, and JsonLexContext::token_terminator.

Referenced by IsValidJsonNumber(), and json_lex().

json_lex_string()

static JsonParseErrorType json_lex_string ( JsonLexContext *  lex )

inlinestatic

Definition at line 1671 of file jsonapi.c.

{
    const char *s;
    const char *const end = lex->input + lex->input_length;
    int         hi_surrogate = -1;
 
    /* Convenience macros for error exits */
#define FAIL_OR_INCOMPLETE_AT_CHAR_START(code) \
    do { \
        if (lex->incremental && !lex->inc_state->is_last_chunk) \
        { \
            appendBinaryStringInfo(&lex->inc_state->partial_token, \
                                   lex->token_start, end - lex->token_start); \
            return JSON_INCOMPLETE; \
        } \
        lex->token_terminator = s; \
        return code; \
    } while (0)
#define FAIL_AT_CHAR_END(code) \
    do { \
        const char     *term = s + pg_encoding_mblen(lex->input_encoding, s); \
        lex->token_terminator = (term <= end) ? term : end; \
        return code; \
    } while (0)
 
    if (lex->strval != NULL)
        resetStringInfo(lex->strval);
 
    Assert(lex->input_length > 0);
    s = lex->token_start;
    for (;;)
    {
        s++;
        /* Premature end of the string. */
        if (s >= end)
            FAIL_OR_INCOMPLETE_AT_CHAR_START(JSON_INVALID_TOKEN);
        else if (*s == '"')
            break;
        else if (*s == '\\')
        {
            /* OK, we have an escape character. */
            s++;
            if (s >= end)
                FAIL_OR_INCOMPLETE_AT_CHAR_START(JSON_INVALID_TOKEN);
            else if (*s == 'u')
            {
                int         i;
                int         ch = 0;
 
                for (i = 1; i <= 4; i++)
                {
                    s++;
                    if (s >= end)
                        FAIL_OR_INCOMPLETE_AT_CHAR_START(JSON_INVALID_TOKEN);
                    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
                        FAIL_AT_CHAR_END(JSON_UNICODE_ESCAPE_FORMAT);
                }
                if (lex->strval != NULL)
                {
                    /*
                     * Combine surrogate pairs.
                     */
                    if (is_utf16_surrogate_first(ch))
                    {
                        if (hi_surrogate != -1)
                            FAIL_AT_CHAR_END(JSON_UNICODE_HIGH_SURROGATE);
                        hi_surrogate = ch;
                        continue;
                    }
                    else if (is_utf16_surrogate_second(ch))
                    {
                        if (hi_surrogate == -1)
                            FAIL_AT_CHAR_END(JSON_UNICODE_LOW_SURROGATE);
                        ch = surrogate_pair_to_codepoint(hi_surrogate, ch);
                        hi_surrogate = -1;
                    }
 
                    if (hi_surrogate != -1)
                        FAIL_AT_CHAR_END(JSON_UNICODE_LOW_SURROGATE);
 
                    /*
                     * Reject invalid cases.  We can't have a value above
                     * 0xFFFF here (since we only accepted 4 hex digits
                     * above), so no need to test for out-of-range chars.
                     */
                    if (ch == 0)
                    {
                        /* We can't allow this, since our TEXT type doesn't */
                        FAIL_AT_CHAR_END(JSON_UNICODE_CODE_POINT_ZERO);
                    }
 
                    /*
                     * Add the represented character to lex->strval.  In the
                     * backend, we can let pg_unicode_to_server_noerror()
                     * handle any required character set conversion; in
                     * frontend, we can only deal with trivial conversions.
                     */
#ifndef FRONTEND
                    {
                        char        cbuf[MAX_UNICODE_EQUIVALENT_STRING + 1];
 
                        if (!pg_unicode_to_server_noerror(ch, (unsigned char *) cbuf))
                            FAIL_AT_CHAR_END(JSON_UNICODE_UNTRANSLATABLE);
                        appendStringInfoString(lex->strval, cbuf);
                    }
#else
                    if (lex->input_encoding == PG_UTF8)
                    {
                        /* OK, we can map the code point to UTF8 easily */
                        char        utf8str[5];
                        int         utf8len;
 
                        unicode_to_utf8(ch, (unsigned char *) utf8str);
                        utf8len = pg_utf_mblen((unsigned char *) utf8str);
                        appendBinaryStringInfo(lex->strval, utf8str, utf8len);
                    }
                    else if (ch <= 0x007f)
                    {
                        /* The ASCII range is the same in all encodings */
                        appendStringInfoChar(lex->strval, (char) ch);
                    }
                    else
                        FAIL_AT_CHAR_END(JSON_UNICODE_HIGH_ESCAPE);
#endif                          /* FRONTEND */
                }
            }
            else if (lex->strval != NULL)
            {
                if (hi_surrogate != -1)
                    FAIL_AT_CHAR_END(JSON_UNICODE_LOW_SURROGATE);
 
                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 signal error.  We
                         * adjust token_start so that just the escape sequence
                         * is reported, not the whole string.
                         */
                        lex->token_start = s;
                        FAIL_AT_CHAR_END(JSON_ESCAPING_INVALID);
                }
            }
            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_start = s;
                FAIL_AT_CHAR_END(JSON_ESCAPING_INVALID);
            }
        }
        else
        {
            const char *p = s;
 
            if (hi_surrogate != -1)
                FAIL_AT_CHAR_END(JSON_UNICODE_LOW_SURROGATE);
 
            /*
             * Skip to the first byte that requires special handling, so we
             * can batch calls to appendBinaryStringInfo.
             */
            while (p < end - sizeof(Vector8) &&
                   !pg_lfind8('\\', (uint8 *) p, sizeof(Vector8)) &&
                   !pg_lfind8('"', (uint8 *) p, sizeof(Vector8)) &&
                   !pg_lfind8_le(31, (uint8 *) p, sizeof(Vector8)))
                p += sizeof(Vector8);
 
            for (; p < end; p++)
            {
                if (*p == '\\' || *p == '"')
                    break;
                else if ((unsigned char) *p <= 31)
                {
                    /* Per RFC4627, these characters MUST be escaped. */
                    /*
                     * Since *p isn't printable, exclude it from the context
                     * string
                     */
                    lex->token_terminator = p;
                    return JSON_ESCAPING_REQUIRED;
                }
            }
 
            if (lex->strval != NULL)
                appendBinaryStringInfo(lex->strval, s, p - s);
 
            /*
             * s will be incremented at the top of the loop, so set it to just
             * behind our lookahead position
             */
            s = p - 1;
        }
    }
 
    if (hi_surrogate != -1)
    {
        lex->token_terminator = s + 1;
        return JSON_UNICODE_LOW_SURROGATE;
    }
 
    /* Hooray, we found the end of the string! */
    lex->prev_token_terminator = lex->token_terminator;
    lex->token_terminator = s + 1;
    return JSON_SUCCESS;
 
#undef FAIL_OR_INCOMPLETE_AT_CHAR_START
#undef FAIL_AT_CHAR_END

References appendBinaryStringInfo(), appendStringInfoChar(), appendStringInfoString(), Assert, FAIL_AT_CHAR_END, FAIL_OR_INCOMPLETE_AT_CHAR_START, i, JsonLexContext::input, JsonLexContext::input_encoding, JsonLexContext::input_length, is_utf16_surrogate_first(), is_utf16_surrogate_second(), JSON_ESCAPING_INVALID, JSON_ESCAPING_REQUIRED, JSON_INVALID_TOKEN, JSON_SUCCESS, JSON_UNICODE_CODE_POINT_ZERO, JSON_UNICODE_ESCAPE_FORMAT, JSON_UNICODE_HIGH_ESCAPE, JSON_UNICODE_HIGH_SURROGATE, JSON_UNICODE_LOW_SURROGATE, JSON_UNICODE_UNTRANSLATABLE, MAX_UNICODE_EQUIVALENT_STRING, pg_lfind8(), pg_lfind8_le(), pg_unicode_to_server_noerror(), PG_UTF8, pg_utf_mblen, JsonLexContext::prev_token_terminator, resetStringInfo(), JsonLexContext::strval, surrogate_pair_to_codepoint(), JsonLexContext::token_start, JsonLexContext::token_terminator, and unicode_to_utf8().

Referenced by json_lex().

lex_expect()

static JsonParseErrorType lex_expect ( JsonParseContext  ctx, JsonLexContext *  lex, JsonTokenType  token  )

inlinestatic

Definition at line 250 of file jsonapi.c.

{
    if (lex_peek(lex) == token)
        return json_lex(lex);
    else
        return report_parse_error(ctx, lex);
}

References json_lex(), lex_peek(), report_parse_error(), and token.

Referenced by json_count_array_elements(), parse_array(), parse_object(), parse_object_field(), and pg_parse_json().

lex_peek()

static JsonTokenType lex_peek ( JsonLexContext *  lex )

inlinestatic

Definition at line 238 of file jsonapi.c.

{
    return lex->token_type;
}

References JsonLexContext::token_type.

Referenced by json_count_array_elements(), lex_expect(), parse_array(), parse_array_element(), parse_object(), parse_object_field(), parse_scalar(), pg_parse_json(), and pg_parse_json_incremental().

makeJsonLexContextCstringLen()

JsonLexContext* makeJsonLexContextCstringLen	(	JsonLexContext *	lex,
		const char *	json,
		size_t	len,
		int	encoding,
		bool	need_escapes
	)

Definition at line 326 of file jsonapi.c.

{
    if (lex == NULL)
    {
        lex = palloc0(sizeof(JsonLexContext));
        lex->flags |= JSONLEX_FREE_STRUCT;
    }
    else
        memset(lex, 0, sizeof(JsonLexContext));
 
    lex->errormsg = NULL;
    lex->input = lex->token_terminator = lex->line_start = json;
    lex->line_number = 1;
    lex->input_length = len;
    lex->input_encoding = encoding;
    if (need_escapes)
    {
        lex->strval = makeStringInfo();
        lex->flags |= JSONLEX_FREE_STRVAL;
    }
 
    return lex;
}

References encoding, JsonLexContext::errormsg, JsonLexContext::flags, JsonLexContext::input, JsonLexContext::input_encoding, JsonLexContext::input_length, JSONLEX_FREE_STRUCT, JSONLEX_FREE_STRVAL, len, JsonLexContext::line_number, JsonLexContext::line_start, makeStringInfo(), palloc0(), JsonLexContext::strval, and JsonLexContext::token_terminator.

Referenced by get_json_object_as_hash(), json_parse_manifest(), json_recv(), jsonb_from_cstring(), main(), makeJsonLexContext(), and populate_array_json().

makeJsonLexContextIncremental()

JsonLexContext* makeJsonLexContextIncremental	(	JsonLexContext *	lex,
		int	encoding,
		bool	need_escapes
	)

Definition at line 366 of file jsonapi.c.

{
    if (lex == NULL)
    {
        lex = palloc0(sizeof(JsonLexContext));
        lex->flags |= JSONLEX_FREE_STRUCT;
    }
    else
        memset(lex, 0, sizeof(JsonLexContext));
 
    lex->line_number = 1;
    lex->input_encoding = encoding;
    lex->incremental = true;
    lex->inc_state = palloc0(sizeof(JsonIncrementalState));
    initStringInfo(&(lex->inc_state->partial_token));
    lex->pstack = palloc(sizeof(JsonParserStack));
    lex->pstack->stack_size = JS_STACK_CHUNK_SIZE;
    lex->pstack->prediction = palloc(JS_STACK_CHUNK_SIZE * JS_MAX_PROD_LEN);
    lex->pstack->pred_index = 0;
    lex->pstack->fnames = palloc(JS_STACK_CHUNK_SIZE * sizeof(char *));
    lex->pstack->fnull = palloc(JS_STACK_CHUNK_SIZE * sizeof(bool));
    if (need_escapes)
    {
        lex->strval = makeStringInfo();
        lex->flags |= JSONLEX_FREE_STRVAL;
    }
    return lex;

References encoding, JsonLexContext::flags, JsonParserStack::fnames, JsonParserStack::fnull, JsonLexContext::inc_state, JsonLexContext::incremental, initStringInfo(), JsonLexContext::input_encoding, JS_MAX_PROD_LEN, JS_STACK_CHUNK_SIZE, JSONLEX_FREE_STRUCT, JSONLEX_FREE_STRVAL, JsonLexContext::line_number, makeStringInfo(), palloc(), palloc0(), JsonIncrementalState::partial_token, JsonParserStack::pred_index, JsonParserStack::prediction, JsonLexContext::pstack, JsonParserStack::stack_size, and JsonLexContext::strval.

Referenced by json_parse_manifest_incremental_init(), and main().

next_prediction()

static char next_prediction ( JsonParserStack *  pstack )

inlinestatic

Definition at line 438 of file jsonapi.c.

{
    Assert(pstack->pred_index > 0);
    return pstack->prediction[pstack->pred_index - 1];

References Assert, JsonParserStack::pred_index, and JsonParserStack::prediction.

Referenced by pg_parse_json_incremental().

parse_array()

static JsonParseErrorType parse_array ( JsonLexContext *  lex, const JsonSemAction *  sem  )

static

Definition at line 1231 of file jsonapi.c.

{
    /*
     * 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;
    JsonParseErrorType result;
 
#ifndef FRONTEND
    check_stack_depth();
#endif
 
    if (astart != NULL)
    {
        result = (*astart) (sem->semstate);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    /*
     * 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++;
 
    result = lex_expect(JSON_PARSE_ARRAY_START, lex, JSON_TOKEN_ARRAY_START);
    if (result == JSON_SUCCESS && lex_peek(lex) != JSON_TOKEN_ARRAY_END)
    {
        result = parse_array_element(lex, sem);
 
        while (result == JSON_SUCCESS && lex_peek(lex) == JSON_TOKEN_COMMA)
        {
            result = json_lex(lex);
            if (result != JSON_SUCCESS)
                break;
            result = parse_array_element(lex, sem);
        }
    }
    if (result != JSON_SUCCESS)
        return result;
 
    result = lex_expect(JSON_PARSE_ARRAY_NEXT, lex, JSON_TOKEN_ARRAY_END);
    if (result != JSON_SUCCESS)
        return result;
 
    lex->lex_level--;
 
    if (aend != NULL)
    {
        result = (*aend) (sem->semstate);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    return JSON_SUCCESS;

References JsonSemAction::array_end, JsonSemAction::array_start, check_stack_depth(), json_lex(), JSON_PARSE_ARRAY_NEXT, JSON_PARSE_ARRAY_START, JSON_SUCCESS, JSON_TOKEN_ARRAY_END, JSON_TOKEN_ARRAY_START, JSON_TOKEN_COMMA, lex_expect(), JsonLexContext::lex_level, lex_peek(), parse_array_element(), sem, and JsonSemAction::semstate.

Referenced by parse_array_element(), parse_object_field(), and pg_parse_json().

parse_array_element()

static JsonParseErrorType parse_array_element ( JsonLexContext *  lex, const JsonSemAction *  sem  )

static

Definition at line 1187 of file jsonapi.c.

{
    json_aelem_action astart = sem->array_element_start;
    json_aelem_action aend = sem->array_element_end;
    JsonTokenType tok = lex_peek(lex);
    JsonParseErrorType result;
    bool        isnull;
 
    isnull = tok == JSON_TOKEN_NULL;
 
    if (astart != NULL)
    {
        result = (*astart) (sem->semstate, isnull);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    /* an array element is any object, array or scalar */
    switch (tok)
    {
        case JSON_TOKEN_OBJECT_START:
            result = parse_object(lex, sem);
            break;
        case JSON_TOKEN_ARRAY_START:
            result = parse_array(lex, sem);
            break;
        default:
            result = parse_scalar(lex, sem);
    }
 
    if (result != JSON_SUCCESS)
        return result;
 
    if (aend != NULL)
    {
        result = (*aend) (sem->semstate, isnull);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    return JSON_SUCCESS;

References JsonSemAction::array_element_end, JsonSemAction::array_element_start, JSON_SUCCESS, JSON_TOKEN_ARRAY_START, JSON_TOKEN_NULL, JSON_TOKEN_OBJECT_START, lex_peek(), parse_array(), parse_object(), parse_scalar(), sem, and JsonSemAction::semstate.

Referenced by json_count_array_elements(), and parse_array().

parse_object()

static JsonParseErrorType parse_object ( JsonLexContext *  lex, const JsonSemAction *  sem  )

static

Definition at line 1113 of file jsonapi.c.

{
    /*
     * 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;
    JsonParseErrorType result;
 
#ifndef FRONTEND
    check_stack_depth();
#endif
 
    if (ostart != NULL)
    {
        result = (*ostart) (sem->semstate);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    /*
     * 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++;
 
    Assert(lex_peek(lex) == JSON_TOKEN_OBJECT_START);
    result = json_lex(lex);
    if (result != JSON_SUCCESS)
        return result;
 
    tok = lex_peek(lex);
    switch (tok)
    {
        case JSON_TOKEN_STRING:
            result = parse_object_field(lex, sem);
            while (result == JSON_SUCCESS && lex_peek(lex) == JSON_TOKEN_COMMA)
            {
                result = json_lex(lex);
                if (result != JSON_SUCCESS)
                    break;
                result = parse_object_field(lex, sem);
            }
            break;
        case JSON_TOKEN_OBJECT_END:
            break;
        default:
            /* case of an invalid initial token inside the object */
            result = report_parse_error(JSON_PARSE_OBJECT_START, lex);
    }
    if (result != JSON_SUCCESS)
        return result;
 
    result = lex_expect(JSON_PARSE_OBJECT_NEXT, lex, JSON_TOKEN_OBJECT_END);
    if (result != JSON_SUCCESS)
        return result;
 
    lex->lex_level--;
 
    if (oend != NULL)
    {
        result = (*oend) (sem->semstate);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    return JSON_SUCCESS;

References Assert, check_stack_depth(), json_lex(), JSON_PARSE_OBJECT_NEXT, JSON_PARSE_OBJECT_START, JSON_SUCCESS, JSON_TOKEN_COMMA, JSON_TOKEN_OBJECT_END, JSON_TOKEN_OBJECT_START, JSON_TOKEN_STRING, lex_expect(), JsonLexContext::lex_level, lex_peek(), JsonSemAction::object_end, JsonSemAction::object_start, parse_object_field(), report_parse_error(), sem, and JsonSemAction::semstate.

Referenced by parse_array_element(), parse_object_field(), and pg_parse_json().

parse_object_field()

static JsonParseErrorType parse_object_field ( JsonLexContext *  lex, const JsonSemAction *  sem  )

static

Definition at line 1051 of file jsonapi.c.

{
    /*
     * 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;
    JsonTokenType tok;
    JsonParseErrorType result;
 
    if (lex_peek(lex) != JSON_TOKEN_STRING)
        return report_parse_error(JSON_PARSE_STRING, lex);
    if ((ostart != NULL || oend != NULL) && lex->strval != NULL)
        fname = pstrdup(lex->strval->data);
    result = json_lex(lex);
    if (result != JSON_SUCCESS)
        return result;
 
    result = lex_expect(JSON_PARSE_OBJECT_LABEL, lex, JSON_TOKEN_COLON);
    if (result != JSON_SUCCESS)
        return result;
 
    tok = lex_peek(lex);
    isnull = tok == JSON_TOKEN_NULL;
 
    if (ostart != NULL)
    {
        result = (*ostart) (sem->semstate, fname, isnull);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    switch (tok)
    {
        case JSON_TOKEN_OBJECT_START:
            result = parse_object(lex, sem);
            break;
        case JSON_TOKEN_ARRAY_START:
            result = parse_array(lex, sem);
            break;
        default:
            result = parse_scalar(lex, sem);
    }
    if (result != JSON_SUCCESS)
        return result;
 
    if (oend != NULL)
    {
        result = (*oend) (sem->semstate, fname, isnull);
        if (result != JSON_SUCCESS)
            return result;
    }
 
    return JSON_SUCCESS;

References StringInfoData::data, json_lex(), JSON_PARSE_OBJECT_LABEL, JSON_PARSE_STRING, JSON_SUCCESS, JSON_TOKEN_ARRAY_START, JSON_TOKEN_COLON, JSON_TOKEN_NULL, JSON_TOKEN_OBJECT_START, JSON_TOKEN_STRING, lex_expect(), lex_peek(), JsonSemAction::object_field_end, JsonSemAction::object_field_start, parse_array(), parse_object(), parse_scalar(), pstrdup(), report_parse_error(), sem, JsonSemAction::semstate, and JsonLexContext::strval.

Referenced by parse_object().

parse_scalar()

static JsonParseErrorType parse_scalar ( JsonLexContext *  lex, const JsonSemAction *  sem  )

inlinestatic

Definition at line 1007 of file jsonapi.c.

{
    char       *val = NULL;
    json_scalar_action sfunc = sem->scalar;
    JsonTokenType tok = lex_peek(lex);
    JsonParseErrorType result;
 
    /* a scalar must be a string, a number, true, false, or null */
    if (tok != JSON_TOKEN_STRING && tok != JSON_TOKEN_NUMBER &&
        tok != JSON_TOKEN_TRUE && tok != JSON_TOKEN_FALSE &&
        tok != JSON_TOKEN_NULL)
        return report_parse_error(JSON_PARSE_VALUE, lex);
 
    /* if no semantic function, just consume the token */
    if (sfunc == NULL)
        return json_lex(lex);
 
    /* extract the de-escaped string value, or the raw lexeme */
    if (lex_peek(lex) == JSON_TOKEN_STRING)
    {
        if (lex->strval != NULL)
            val = pstrdup(lex->strval->data);
    }
    else
    {
        int         len = (lex->token_terminator - lex->token_start);
 
        val = palloc(len + 1);
        memcpy(val, lex->token_start, len);
        val[len] = '\0';
    }
 
    /* consume the token */
    result = json_lex(lex);
    if (result != JSON_SUCCESS)
        return result;
 
    /* invoke the callback */
    result = (*sfunc) (sem->semstate, val, tok);
 
    return result;

References StringInfoData::data, json_lex(), JSON_PARSE_VALUE, JSON_SUCCESS, JSON_TOKEN_FALSE, JSON_TOKEN_NULL, JSON_TOKEN_NUMBER, JSON_TOKEN_STRING, JSON_TOKEN_TRUE, len, lex_peek(), palloc(), pstrdup(), report_parse_error(), JsonSemAction::scalar, sem, JsonSemAction::semstate, JsonLexContext::strval, JsonLexContext::token_start, JsonLexContext::token_terminator, and val.

Referenced by parse_array_element(), parse_object_field(), and pg_parse_json().

pg_parse_json()

JsonParseErrorType pg_parse_json	(	JsonLexContext *	lex,
		const JsonSemAction *	sem
	)

Definition at line 521 of file jsonapi.c.

{
#ifdef FORCE_JSON_PSTACK
 
    lex->incremental = true;
    lex->inc_state = palloc0(sizeof(JsonIncrementalState));
 
    /*
     * We don't need partial token processing, there is only one chunk. But we
     * still need to init the partial token string so that freeJsonLexContext
     * works.
     */
    initStringInfo(&(lex->inc_state->partial_token));
    lex->pstack = palloc(sizeof(JsonParserStack));
    lex->pstack->stack_size = JS_STACK_CHUNK_SIZE;
    lex->pstack->prediction = palloc(JS_STACK_CHUNK_SIZE * JS_MAX_PROD_LEN);
    lex->pstack->pred_index = 0;
    lex->pstack->fnames = palloc(JS_STACK_CHUNK_SIZE * sizeof(char *));
    lex->pstack->fnull = palloc(JS_STACK_CHUNK_SIZE * sizeof(bool));
 
    return pg_parse_json_incremental(lex, sem, lex->input, lex->input_length, true);
 
#else
 
    JsonTokenType tok;
    JsonParseErrorType result;
 
    if (lex->incremental)
        return JSON_INVALID_LEXER_TYPE;
 
    /* get the initial token */
    result = json_lex(lex);
    if (result != JSON_SUCCESS)
        return result;
 
    tok = lex_peek(lex);
 
    /* parse by recursive descent */
    switch (tok)
    {
        case JSON_TOKEN_OBJECT_START:
            result = parse_object(lex, sem);
            break;
        case JSON_TOKEN_ARRAY_START:
            result = parse_array(lex, sem);
            break;
        default:
            result = parse_scalar(lex, sem);    /* json can be a bare scalar */
    }
 
    if (result == JSON_SUCCESS)
        result = lex_expect(JSON_PARSE_END, lex, JSON_TOKEN_END);
 
    return result;
#endif

References JsonParserStack::fnames, JsonParserStack::fnull, JsonLexContext::inc_state, JsonLexContext::incremental, initStringInfo(), JsonLexContext::input, JsonLexContext::input_length, JS_MAX_PROD_LEN, JS_STACK_CHUNK_SIZE, JSON_INVALID_LEXER_TYPE, json_lex(), JSON_PARSE_END, JSON_SUCCESS, JSON_TOKEN_ARRAY_START, JSON_TOKEN_END, JSON_TOKEN_OBJECT_START, lex_expect(), lex_peek(), palloc(), palloc0(), parse_array(), parse_object(), parse_scalar(), JsonIncrementalState::partial_token, pg_parse_json_incremental(), JsonParserStack::pred_index, JsonParserStack::prediction, JsonLexContext::pstack, sem, and JsonParserStack::stack_size.

Referenced by json_parse_manifest(), json_validate(), main(), and pg_parse_json_or_errsave().

pg_parse_json_incremental()

JsonParseErrorType pg_parse_json_incremental	(	JsonLexContext *	lex,
		const JsonSemAction *	sem,
		const char *	json,
		size_t	len,
		bool	is_last
	)

Definition at line 649 of file jsonapi.c.

{
    JsonTokenType tok;
    JsonParseErrorType result;
    JsonParseContext ctx = JSON_PARSE_VALUE;
    JsonParserStack *pstack = lex->pstack;
 
 
    if (!lex->incremental)
        return JSON_INVALID_LEXER_TYPE;
 
    lex->input = lex->token_terminator = lex->line_start = json;
    lex->input_length = len;
    lex->inc_state->is_last_chunk = is_last;
 
    /* get the initial token */
    result = json_lex(lex);
    if (result != JSON_SUCCESS)
        return result;
 
    tok = lex_peek(lex);
 
    /* use prediction stack for incremental parsing */
 
    if (!have_prediction(pstack))
    {
        td_entry    goal = TD_ENTRY(JSON_PROD_GOAL);
 
        push_prediction(pstack, goal);
    }
 
    while (have_prediction(pstack))
    {
        char        top = pop_prediction(pstack);
        td_entry    entry;
 
        /*
         * these first two branches are the guts of the Table Driven method
         */
        if (top == tok)
        {
            /*
             * tok can only be a terminal symbol, so top must be too. the
             * token matches the top of the stack, so get the next token.
             */
            if (tok < JSON_TOKEN_END)
            {
                result = json_lex(lex);
                if (result != JSON_SUCCESS)
                    return result;
                tok = lex_peek(lex);
            }
        }
        else if (IS_NT(top) && (entry = td_parser_table[OFS(top)][tok]).prod != NULL)
        {
            /*
             * the token is in the director set for a production of the
             * non-terminal at the top of the stack, so push the reversed RHS
             * of the production onto the stack.
             */
            push_prediction(pstack, entry);
        }
        else if (IS_SEM(top))
        {
            /*
             * top is a semantic action marker, so take action accordingly.
             * It's important to have these markers in the prediction stack
             * before any token they might need so we don't advance the token
             * prematurely. Note in a couple of cases we need to do something
             * both before and after the token.
             */
            switch (top)
            {
                case JSON_SEM_OSTART:
                    {
                        json_struct_action ostart = sem->object_start;
 
                        if (lex->lex_level >= JSON_TD_MAX_STACK)
                            return JSON_NESTING_TOO_DEEP;
 
                        if (ostart != NULL)
                        {
                            result = (*ostart) (sem->semstate);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                        inc_lex_level(lex);
                    }
                    break;
                case JSON_SEM_OEND:
                    {
                        json_struct_action oend = sem->object_end;
 
                        dec_lex_level(lex);
                        if (oend != NULL)
                        {
                            result = (*oend) (sem->semstate);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_ASTART:
                    {
                        json_struct_action astart = sem->array_start;
 
                        if (lex->lex_level >= JSON_TD_MAX_STACK)
                            return JSON_NESTING_TOO_DEEP;
 
                        if (astart != NULL)
                        {
                            result = (*astart) (sem->semstate);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                        inc_lex_level(lex);
                    }
                    break;
                case JSON_SEM_AEND:
                    {
                        json_struct_action aend = sem->array_end;
 
                        dec_lex_level(lex);
                        if (aend != NULL)
                        {
                            result = (*aend) (sem->semstate);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_OFIELD_INIT:
                    {
                        /*
                         * all we do here is save out the field name. We have
                         * to wait to get past the ':' to see if the next
                         * value is null so we can call the semantic routine
                         */
                        char       *fname = NULL;
                        json_ofield_action ostart = sem->object_field_start;
                        json_ofield_action oend = sem->object_field_end;
 
                        if ((ostart != NULL || oend != NULL) && lex->strval != NULL)
                        {
                            fname = pstrdup(lex->strval->data);
                        }
                        set_fname(lex, fname);
                    }
                    break;
                case JSON_SEM_OFIELD_START:
                    {
                        /*
                         * the current token should be the first token of the
                         * value
                         */
                        bool        isnull = tok == JSON_TOKEN_NULL;
                        json_ofield_action ostart = sem->object_field_start;
 
                        set_fnull(lex, isnull);
 
                        if (ostart != NULL)
                        {
                            char       *fname = get_fname(lex);
 
                            result = (*ostart) (sem->semstate, fname, isnull);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_OFIELD_END:
                    {
                        json_ofield_action oend = sem->object_field_end;
 
                        if (oend != NULL)
                        {
                            char       *fname = get_fname(lex);
                            bool        isnull = get_fnull(lex);
 
                            result = (*oend) (sem->semstate, fname, isnull);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_AELEM_START:
                    {
                        json_aelem_action astart = sem->array_element_start;
                        bool        isnull = tok == JSON_TOKEN_NULL;
 
                        set_fnull(lex, isnull);
 
                        if (astart != NULL)
                        {
                            result = (*astart) (sem->semstate, isnull);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_AELEM_END:
                    {
                        json_aelem_action aend = sem->array_element_end;
 
                        if (aend != NULL)
                        {
                            bool        isnull = get_fnull(lex);
 
                            result = (*aend) (sem->semstate, isnull);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                case JSON_SEM_SCALAR_INIT:
                    {
                        json_scalar_action sfunc = sem->scalar;
 
                        pstack->scalar_val = NULL;
 
                        if (sfunc != NULL)
                        {
                            /*
                             * extract the de-escaped string value, or the raw
                             * lexeme
                             */
                            /*
                             * XXX copied from RD parser but looks like a
                             * buglet
                             */
                            if (tok == JSON_TOKEN_STRING)
                            {
                                if (lex->strval != NULL)
                                    pstack->scalar_val = pstrdup(lex->strval->data);
                            }
                            else
                            {
                                ptrdiff_t   tlen = (lex->token_terminator - lex->token_start);
 
                                pstack->scalar_val = palloc(tlen + 1);
                                memcpy(pstack->scalar_val, lex->token_start, tlen);
                                pstack->scalar_val[tlen] = '\0';
                            }
                            pstack->scalar_tok = tok;
                        }
                    }
                    break;
                case JSON_SEM_SCALAR_CALL:
                    {
                        /*
                         * We'd like to be able to get rid of this business of
                         * two bits of scalar action, but we can't. It breaks
                         * certain semantic actions which expect that when
                         * called the lexer has consumed the item. See for
                         * example get_scalar() in jsonfuncs.c.
                         */
                        json_scalar_action sfunc = sem->scalar;
 
                        if (sfunc != NULL)
                        {
                            result = (*sfunc) (sem->semstate, pstack->scalar_val, pstack->scalar_tok);
                            if (result != JSON_SUCCESS)
                                return result;
                        }
                    }
                    break;
                default:
                    /* should not happen */
                    break;
            }
        }
        else
        {
            /*
             * The token didn't match the stack top if it's a terminal nor a
             * production for the stack top if it's a non-terminal.
             *
             * Various cases here are Asserted to be not possible, as the
             * token would not appear at the top of the prediction stack
             * unless the lookahead matched.
             */
            switch (top)
            {
                case JSON_TOKEN_STRING:
                    if (next_prediction(pstack) == JSON_TOKEN_COLON)
                        ctx = JSON_PARSE_STRING;
                    else
                    {
                        Assert(false);
                        ctx = JSON_PARSE_VALUE;
                    }
                    break;
                case JSON_TOKEN_NUMBER:
                case JSON_TOKEN_TRUE:
                case JSON_TOKEN_FALSE:
                case JSON_TOKEN_NULL:
                case JSON_TOKEN_ARRAY_START:
                case JSON_TOKEN_OBJECT_START:
                    Assert(false);
                    ctx = JSON_PARSE_VALUE;
                    break;
                case JSON_TOKEN_ARRAY_END:
                    Assert(false);
                    ctx = JSON_PARSE_ARRAY_NEXT;
                    break;
                case JSON_TOKEN_OBJECT_END:
                    Assert(false);
                    ctx = JSON_PARSE_OBJECT_NEXT;
                    break;
                case JSON_TOKEN_COMMA:
                    Assert(false);
                    if (next_prediction(pstack) == JSON_TOKEN_STRING)
                        ctx = JSON_PARSE_OBJECT_NEXT;
                    else
                        ctx = JSON_PARSE_ARRAY_NEXT;
                    break;
                case JSON_TOKEN_COLON:
                    ctx = JSON_PARSE_OBJECT_LABEL;
                    break;
                case JSON_TOKEN_END:
                    ctx = JSON_PARSE_END;
                    break;
                case JSON_NT_MORE_ARRAY_ELEMENTS:
                    ctx = JSON_PARSE_ARRAY_NEXT;
                    break;
                case JSON_NT_ARRAY_ELEMENTS:
                    ctx = JSON_PARSE_ARRAY_START;
                    break;
                case JSON_NT_MORE_KEY_PAIRS:
                    ctx = JSON_PARSE_OBJECT_NEXT;
                    break;
                case JSON_NT_KEY_PAIRS:
                    ctx = JSON_PARSE_OBJECT_START;
                    break;
                default:
                    ctx = JSON_PARSE_VALUE;
            }
            return report_parse_error(ctx, lex);
        }
    }
 
    return JSON_SUCCESS;

References JsonSemAction::array_element_end, JsonSemAction::array_element_start, JsonSemAction::array_end, JsonSemAction::array_start, Assert, StringInfoData::data, dec_lex_level(), get_fname(), get_fnull(), have_prediction(), inc_lex_level(), JsonLexContext::inc_state, JsonLexContext::incremental, JsonLexContext::input, JsonLexContext::input_length, JsonIncrementalState::is_last_chunk, IS_NT, IS_SEM, JSON_INVALID_LEXER_TYPE, json_lex(), JSON_NESTING_TOO_DEEP, JSON_NT_ARRAY_ELEMENTS, JSON_NT_KEY_PAIRS, JSON_NT_MORE_ARRAY_ELEMENTS, JSON_NT_MORE_KEY_PAIRS, JSON_PARSE_ARRAY_NEXT, JSON_PARSE_ARRAY_START, JSON_PARSE_END, JSON_PARSE_OBJECT_LABEL, JSON_PARSE_OBJECT_NEXT, JSON_PARSE_OBJECT_START, JSON_PARSE_STRING, JSON_PARSE_VALUE, JSON_PROD_GOAL, JSON_SEM_AELEM_END, JSON_SEM_AELEM_START, JSON_SEM_AEND, JSON_SEM_ASTART, JSON_SEM_OEND, JSON_SEM_OFIELD_END, JSON_SEM_OFIELD_INIT, JSON_SEM_OFIELD_START, JSON_SEM_OSTART, JSON_SEM_SCALAR_CALL, JSON_SEM_SCALAR_INIT, JSON_SUCCESS, JSON_TD_MAX_STACK, JSON_TOKEN_ARRAY_END, JSON_TOKEN_ARRAY_START, JSON_TOKEN_COLON, JSON_TOKEN_COMMA, JSON_TOKEN_END, JSON_TOKEN_FALSE, JSON_TOKEN_NULL, JSON_TOKEN_NUMBER, JSON_TOKEN_OBJECT_END, JSON_TOKEN_OBJECT_START, JSON_TOKEN_STRING, JSON_TOKEN_TRUE, len, JsonLexContext::lex_level, lex_peek(), JsonLexContext::line_start, next_prediction(), JsonSemAction::object_end, JsonSemAction::object_field_end, JsonSemAction::object_field_start, JsonSemAction::object_start, OFS, palloc(), pop_prediction(), JsonLexContext::pstack, pstrdup(), push_prediction(), report_parse_error(), JsonSemAction::scalar, JsonParserStack::scalar_tok, JsonParserStack::scalar_val, sem, JsonSemAction::semstate, set_fname(), set_fnull(), JsonLexContext::strval, TD_ENTRY, td_parser_table, JsonLexContext::token_start, and JsonLexContext::token_terminator.

Referenced by json_parse_manifest_incremental_chunk(), main(), and pg_parse_json().

pop_prediction()

static char pop_prediction ( JsonParserStack *  pstack )

inlinestatic

Definition at line 431 of file jsonapi.c.

{
    Assert(pstack->pred_index > 0);
    return pstack->prediction[--pstack->pred_index];

References Assert, JsonParserStack::pred_index, and JsonParserStack::prediction.

Referenced by pg_parse_json_incremental().

push_prediction()

static void push_prediction ( JsonParserStack *  pstack, td_entry  entry  )

inlinestatic

Definition at line 424 of file jsonapi.c.

{
    memcpy(pstack->prediction + pstack->pred_index, entry.prod, entry.len);
    pstack->pred_index += entry.len;

References td_entry::len, JsonParserStack::pred_index, JsonParserStack::prediction, and td_entry::prod.

Referenced by pg_parse_json_incremental().

report_parse_error()

static JsonParseErrorType report_parse_error ( JsonParseContext  ctx, JsonLexContext *  lex  )

static

Definition at line 2052 of file jsonapi.c.

{
    /* Handle case where the input ended prematurely. */
    if (lex->token_start == NULL || lex->token_type == JSON_TOKEN_END)
        return JSON_EXPECTED_MORE;
 
    /* Otherwise choose the error type based on the parsing context. */
    switch (ctx)
    {
        case JSON_PARSE_END:
            return JSON_EXPECTED_END;
        case JSON_PARSE_VALUE:
            return JSON_EXPECTED_JSON;
        case JSON_PARSE_STRING:
            return JSON_EXPECTED_STRING;
        case JSON_PARSE_ARRAY_START:
            return JSON_EXPECTED_ARRAY_FIRST;
        case JSON_PARSE_ARRAY_NEXT:
            return JSON_EXPECTED_ARRAY_NEXT;
        case JSON_PARSE_OBJECT_START:
            return JSON_EXPECTED_OBJECT_FIRST;
        case JSON_PARSE_OBJECT_LABEL:
            return JSON_EXPECTED_COLON;
        case JSON_PARSE_OBJECT_NEXT:
            return JSON_EXPECTED_OBJECT_NEXT;
        case JSON_PARSE_OBJECT_COMMA:
            return JSON_EXPECTED_STRING;
    }
 
    /*
     * We don't use a default: case, so that the compiler will warn about
     * unhandled enum values.
     */
    Assert(false);
    return JSON_SUCCESS;        /* silence stupider compilers */

References Assert, JSON_EXPECTED_ARRAY_FIRST, JSON_EXPECTED_ARRAY_NEXT, JSON_EXPECTED_COLON, JSON_EXPECTED_END, JSON_EXPECTED_JSON, JSON_EXPECTED_MORE, JSON_EXPECTED_OBJECT_FIRST, JSON_EXPECTED_OBJECT_NEXT, JSON_EXPECTED_STRING, JSON_PARSE_ARRAY_NEXT, JSON_PARSE_ARRAY_START, JSON_PARSE_END, JSON_PARSE_OBJECT_COMMA, JSON_PARSE_OBJECT_LABEL, JSON_PARSE_OBJECT_NEXT, JSON_PARSE_OBJECT_START, JSON_PARSE_STRING, JSON_PARSE_VALUE, JSON_SUCCESS, JSON_TOKEN_END, JsonLexContext::token_start, and JsonLexContext::token_type.

Referenced by lex_expect(), parse_object(), parse_object_field(), parse_scalar(), and pg_parse_json_incremental().

set_fname()

static void set_fname ( JsonLexContext *  lex, char *  fname  )

inlinestatic

Definition at line 451 of file jsonapi.c.

{
    lex->pstack->fnames[lex->lex_level] = fname;

References JsonParserStack::fnames, JsonLexContext::lex_level, and JsonLexContext::pstack.

Referenced by pg_parse_json_incremental().

set_fnull()

static void set_fnull ( JsonLexContext *  lex, bool  fnull  )

inlinestatic

Definition at line 463 of file jsonapi.c.

{
    lex->pstack->fnull[lex->lex_level] = fnull;

References JsonParserStack::fnull, JsonLexContext::lex_level, and JsonLexContext::pstack.

Referenced by pg_parse_json_incremental().

Variable Documentation

JSON_PROD_ARRAY

char JSON_PROD_ARRAY[] = {JSON_SEM_AEND, JSON_TOKEN_ARRAY_END, JSON_NT_ARRAY_ELEMENTS, JSON_TOKEN_ARRAY_START, JSON_SEM_ASTART, 0}

static

Definition at line 146 of file jsonapi.c.

JSON_PROD_ARRAY_ELEMENTS

char JSON_PROD_ARRAY_ELEMENTS[] = {JSON_NT_MORE_ARRAY_ELEMENTS, JSON_SEM_AELEM_END, JSON_NT_JSON, JSON_SEM_AELEM_START, 0}

static

Definition at line 149 of file jsonapi.c.

JSON_PROD_EPSILON

char JSON_PROD_EPSILON[] = {0}

static

Definition at line 125 of file jsonapi.c.

JSON_PROD_GOAL

char JSON_PROD_GOAL[] = {JSON_TOKEN_END, JSON_NT_JSON, 0}

static

Definition at line 211 of file jsonapi.c.

Referenced by pg_parse_json_incremental().

JSON_PROD_KEY_PAIRS

char JSON_PROD_KEY_PAIRS[] = {JSON_NT_MORE_KEY_PAIRS, JSON_SEM_OFIELD_END, JSON_NT_JSON, JSON_SEM_OFIELD_START, JSON_TOKEN_COLON, JSON_TOKEN_STRING, JSON_SEM_OFIELD_INIT, 0}

static

Definition at line 155 of file jsonapi.c.

JSON_PROD_MORE_ARRAY_ELEMENTS

char JSON_PROD_MORE_ARRAY_ELEMENTS[] = {JSON_NT_MORE_ARRAY_ELEMENTS, JSON_SEM_AELEM_END, JSON_NT_JSON, JSON_SEM_AELEM_START, JSON_TOKEN_COMMA, 0}

static

Definition at line 152 of file jsonapi.c.

JSON_PROD_MORE_KEY_PAIRS

char JSON_PROD_MORE_KEY_PAIRS[] = {JSON_NT_MORE_KEY_PAIRS, JSON_SEM_OFIELD_END, JSON_NT_JSON, JSON_SEM_OFIELD_START, JSON_TOKEN_COLON, JSON_TOKEN_STRING, JSON_SEM_OFIELD_INIT, JSON_TOKEN_COMMA, 0}

static

Definition at line 158 of file jsonapi.c.

JSON_PROD_OBJECT

char JSON_PROD_OBJECT[] = {JSON_SEM_OEND, JSON_TOKEN_OBJECT_END, JSON_NT_KEY_PAIRS, JSON_TOKEN_OBJECT_START, JSON_SEM_OSTART, 0}

static

Definition at line 143 of file jsonapi.c.

JSON_PROD_SCALAR_FALSE

char JSON_PROD_SCALAR_FALSE[] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_FALSE, JSON_SEM_SCALAR_INIT, 0}

static

Definition at line 137 of file jsonapi.c.

JSON_PROD_SCALAR_NULL

char JSON_PROD_SCALAR_NULL[] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_NULL, JSON_SEM_SCALAR_INIT, 0}

static

Definition at line 140 of file jsonapi.c.

JSON_PROD_SCALAR_NUMBER

char JSON_PROD_SCALAR_NUMBER[] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_NUMBER, JSON_SEM_SCALAR_INIT, 0}

static

Definition at line 131 of file jsonapi.c.

JSON_PROD_SCALAR_STRING

char JSON_PROD_SCALAR_STRING[] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_STRING, JSON_SEM_SCALAR_INIT, 0}

static

Definition at line 128 of file jsonapi.c.

JSON_PROD_SCALAR_TRUE

char JSON_PROD_SCALAR_TRUE[] = {JSON_SEM_SCALAR_CALL, JSON_TOKEN_TRUE, JSON_SEM_SCALAR_INIT, 0}

static

Definition at line 134 of file jsonapi.c.

nullSemAction

const JsonSemAction nullSemAction

Initial value:

=
{
    NULL, NULL, NULL, NULL, NULL,
    NULL, NULL, NULL, NULL, NULL
}

Definition at line 224 of file jsonapi.c.

Referenced by json_count_array_elements(), json_in(), json_recv(), json_validate(), and main().

td_parser_table

td_entry td_parser_table[JSON_NUM_NONTERMINALS][JSON_NUM_TERMINALS]

static

Initial value:

=
{
    
    [OFS(JSON_NT_JSON)][JSON_TOKEN_STRING] = TD_ENTRY(JSON_PROD_SCALAR_STRING),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_NUMBER] = TD_ENTRY(JSON_PROD_SCALAR_NUMBER),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_TRUE] = TD_ENTRY(JSON_PROD_SCALAR_TRUE),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_FALSE] = TD_ENTRY(JSON_PROD_SCALAR_FALSE),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_NULL] = TD_ENTRY(JSON_PROD_SCALAR_NULL),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_ARRAY_START] = TD_ENTRY(JSON_PROD_ARRAY),
    [OFS(JSON_NT_JSON)][JSON_TOKEN_OBJECT_START] = TD_ENTRY(JSON_PROD_OBJECT),
    
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_ARRAY_START] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_OBJECT_START] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_STRING] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_NUMBER] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_TRUE] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_FALSE] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_NULL] = TD_ENTRY(JSON_PROD_ARRAY_ELEMENTS),
    [OFS(JSON_NT_ARRAY_ELEMENTS)][JSON_TOKEN_ARRAY_END] = TD_ENTRY(JSON_PROD_EPSILON),
    
    [OFS(JSON_NT_MORE_ARRAY_ELEMENTS)][JSON_TOKEN_COMMA] = TD_ENTRY(JSON_PROD_MORE_ARRAY_ELEMENTS),
    [OFS(JSON_NT_MORE_ARRAY_ELEMENTS)][JSON_TOKEN_ARRAY_END] = TD_ENTRY(JSON_PROD_EPSILON),
    
    [OFS(JSON_NT_KEY_PAIRS)][JSON_TOKEN_STRING] = TD_ENTRY(JSON_PROD_KEY_PAIRS),
    [OFS(JSON_NT_KEY_PAIRS)][JSON_TOKEN_OBJECT_END] = TD_ENTRY(JSON_PROD_EPSILON),
    
    [OFS(JSON_NT_MORE_KEY_PAIRS)][JSON_TOKEN_COMMA] = TD_ENTRY(JSON_PROD_MORE_KEY_PAIRS),
    [OFS(JSON_NT_MORE_KEY_PAIRS)][JSON_TOKEN_OBJECT_END] = TD_ENTRY(JSON_PROD_EPSILON),
}

Definition at line 180 of file jsonapi.c.

Referenced by pg_parse_json_incremental().