octoleo/qpdf
2
1
Fork 0
mirror of https://github.com/qpdf/qpdf.git synced 2024-12-22 10:58:58 +00:00
Code Issues Releases Wiki Activity

In JSONParser::getToken handle structural and space chars early

Browse Source
This commit is contained in:
m-holger 2023-01-30 13:17:09 +00:00
parent ee32235f54
commit d3152869b6
3 changed files with 95 additions and 73 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

164
libqpdf/JSON.cc
View File

@ -791,7 +791,7 @@ JSONParser::tokenError()
void void
JSONParser::getToken() JSONParser::getToken()
{ {
enum { append, ignore, reread } action = append; enum { append, ignore } action = append;
bool ready = false; bool ready = false;
token.clear(); token.clear();
@ -820,13 +820,103 @@ JSONParser::getToken()
} else { } else {
break; break;
} }
} else { } else {
QTC::TC("libtests", "JSON parse null character"); QTC::TC("libtests", "JSON parse null character");
throw std::runtime_error( throw std::runtime_error(
"JSON: control or null character at offset " + "JSON: control or null character at offset " +
std::to_string(offset)); std::to_string(offset));
} }
} else if (*p == ',') {
if (lex_state == ls_top) {
++p;
++offset;
lex_state = ls_comma;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == ':') {
if (lex_state == ls_top) {
++p;
++offset;
lex_state = ls_colon;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == ' ') {
if (lex_state == ls_top) {
++p;
++offset;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == '{') {
if (lex_state == ls_top) {
token_start = offset;
++p;
++offset;
lex_state = ls_begin_dict;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == '}') {
if (lex_state == ls_top) {
++p;
++offset;
lex_state = ls_end_dict;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == '[') {
if (lex_state == ls_top) {
token_start = offset;
++p;
++offset;
lex_state = ls_begin_array;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else if (*p == ']') {
if (lex_state == ls_top) {
++p;
++offset;
lex_state = ls_end_array;
return;
} else if (lex_state == ls_string) {
token += *p;
++p;
++offset;
} else {
break;
}
} else { } else {
action = append; action = append;
switch (lex_state) { switch (lex_state) {
@ -835,36 +925,6 @@ JSONParser::getToken()
if (*p == '"') { if (*p == '"') {
lex_state = ls_string; lex_state = ls_string;
action = ignore; action = ignore;
} else if (*p == ' ') {
action = ignore;
} else if (*p == ',') {
lex_state = ls_comma;
action = ignore;
ready = true;
} else if (*p == ',') {
lex_state = ls_comma;
action = ignore;
ready = true;
} else if (*p == ':') {
lex_state = ls_colon;
action = ignore;
ready = true;
} else if (*p == '{') {
lex_state = ls_begin_dict;
action = ignore;
ready = true;
} else if (*p == '}') {
lex_state = ls_end_dict;
action = ignore;
ready = true;
} else if (*p == '[') {
lex_state = ls_begin_array;
action = ignore;
ready = true;
} else if (*p == ']') {
lex_state = ls_end_array;
action = ignore;
ready = true;
} else if ((*p >= 'a') && (*p <= 'z')) { } else if ((*p >= 'a') && (*p <= 'z')) {
lex_state = ls_alpha; lex_state = ls_alpha;
} else if (*p == '-') { } else if (*p == '-') {
@ -897,14 +957,6 @@ JSONParser::getToken()
case ls_number_leading_zero: case ls_number_leading_zero:
if (*p == '.') { if (*p == '.') {
lex_state = ls_number_point; lex_state = ls_number_point;
} else if (*p == ' ') {
lex_state = ls_number;
action = ignore;
ready = true;
} else if (strchr("{}[]:,", *p)) {
lex_state = ls_number;
action = reread;
ready = true;
} else if (*p == 'e' || *p == 'E') { } else if (*p == 'e' || *p == 'E') {
lex_state = ls_number_e; lex_state = ls_number_e;
} else { } else {
@ -920,14 +972,6 @@ JSONParser::getToken()
// continue // continue
} else if (*p == '.') { } else if (*p == '.') {
lex_state = ls_number_point; lex_state = ls_number_point;
} else if (*p == ' ') {
lex_state = ls_number;
action = ignore;
ready = true;
} else if (strchr("{}[]:,", *p)) {
lex_state = ls_number;
action = reread;
ready = true;
} else if (*p == 'e' || *p == 'E') { } else if (*p == 'e' || *p == 'E') {
lex_state = ls_number_e; lex_state = ls_number_e;
} else { } else {
@ -946,14 +990,6 @@ JSONParser::getToken()
case ls_number_after_point: case ls_number_after_point:
if ((*p >= '0') && (*p <= '9')) { if ((*p >= '0') && (*p <= '9')) {
// continue // continue
} else if (*p == ' ') {
lex_state = ls_number;
action = ignore;
ready = true;
} else if (strchr("{}[]:,", *p)) {
lex_state = ls_number;
action = reread;
ready = true;
} else if (*p == 'e' || *p == 'E') { } else if (*p == 'e' || *p == 'E') {
lex_state = ls_number_e; lex_state = ls_number_e;
} else { } else {
@ -983,12 +1019,6 @@ JSONParser::getToken()
// We only get here after we have seen an exponent. // We only get here after we have seen an exponent.
if ((*p >= '0') && (*p <= '9')) { if ((*p >= '0') && (*p <= '9')) {
// continue // continue
} else if (*p == ' ') {
action = ignore;
ready = true;
} else if (strchr("{}[]:,", *p)) {
action = reread;
ready = true;
} else { } else {
tokenError(); tokenError();
} }
@ -997,12 +1027,6 @@ JSONParser::getToken()
case ls_alpha: case ls_alpha:
if ((*p >= 'a') && (*p <= 'z')) { if ((*p >= 'a') && (*p <= 'z')) {
// okay // okay
} else if (*p == ' ') {
action = ignore;
ready = true;
} else if (strchr("{}[]:,", *p)) {
action = reread;
ready = true;
} else { } else {
tokenError(); tokenError();
} }
@ -1090,8 +1114,6 @@ JSONParser::getToken()
"JSONParser::getToken : trying to handle delimiter state"); "JSONParser::getToken : trying to handle delimiter state");
} }
switch (action) { switch (action) {
case reread:
break;
case append: case append:
token.append(1, *p); token.append(1, *p);
// fall through // fall through
@ -1107,7 +1129,7 @@ JSONParser::getToken()
} }
// We only get here if on end of input or if the last character was a // We only get here if on end of input or if the last character was a
// control character. // control character or other delimiter.
if (!token.empty()) { if (!token.empty()) {
switch (lex_state) { switch (lex_state) {

2
libtests/qtest/json_parse/bad-09.out
View File

@ -1 +1 @@
exception: bad-09.json: JSON: offset 3: expect string as dictionary key exception: bad-09.json: JSON: offset 2: expect string as dictionary key

2
libtests/qtest/json_parse/bad-31.out
View File

@ -1 +1 @@
exception: bad-31.json: JSON: offset 1: numeric literal: no digit after minus sign exception: bad-31.json: JSON: offset 1: numeric literal: incomplete number