1-- 2-- encoding-sensitive tests for json and jsonb 3-- 4 5-- We provide expected-results files for UTF8 (json_encoding.out) 6-- and for SQL_ASCII (json_encoding_1.out). Skip otherwise. 7SELECT getdatabaseencoding() NOT IN ('UTF8', 'SQL_ASCII') 8 AS skip_test \gset 9\if :skip_test 10\quit 11\endif 12 13SELECT getdatabaseencoding(); -- just to label the results files 14 15-- first json 16 17-- basic unicode input 18SELECT '"\u"'::json; -- ERROR, incomplete escape 19SELECT '"\u00"'::json; -- ERROR, incomplete escape 20SELECT '"\u000g"'::json; -- ERROR, g is not a hex digit 21SELECT '"\u0000"'::json; -- OK, legal escape 22SELECT '"\uaBcD"'::json; -- OK, uppercase and lower case both OK 23 24-- handling of unicode surrogate pairs 25 26select json '{ "a": "\ud83d\ude04\ud83d\udc36" }' -> 'a' as correct_in_utf8; 27select json '{ "a": "\ud83d\ud83d" }' -> 'a'; -- 2 high surrogates in a row 28select json '{ "a": "\ude04\ud83d" }' -> 'a'; -- surrogates in wrong order 29select json '{ "a": "\ud83dX" }' -> 'a'; -- orphan high surrogate 30select json '{ "a": "\ude04X" }' -> 'a'; -- orphan low surrogate 31 32--handling of simple unicode escapes 33 34select json '{ "a": "the Copyright \u00a9 sign" }' as correct_in_utf8; 35select json '{ "a": "dollar \u0024 character" }' as correct_everywhere; 36select json '{ "a": "dollar \\u0024 character" }' as not_an_escape; 37select json '{ "a": "null \u0000 escape" }' as not_unescaped; 38select json '{ "a": "null \\u0000 escape" }' as not_an_escape; 39 40select json '{ "a": "the Copyright \u00a9 sign" }' ->> 'a' as correct_in_utf8; 41select json '{ "a": "dollar \u0024 character" }' ->> 'a' as correct_everywhere; 42select json '{ "a": "dollar \\u0024 character" }' ->> 'a' as not_an_escape; 43select json '{ "a": "null \u0000 escape" }' ->> 'a' as fails; 44select json '{ "a": "null \\u0000 escape" }' ->> 'a' as not_an_escape; 45 46-- then jsonb 47 48-- basic unicode input 49SELECT '"\u"'::jsonb; -- ERROR, incomplete escape 50SELECT '"\u00"'::jsonb; -- ERROR, incomplete escape 51SELECT '"\u000g"'::jsonb; -- ERROR, g is not a hex digit 52SELECT '"\u0045"'::jsonb; -- OK, legal escape 53SELECT '"\u0000"'::jsonb; -- ERROR, we don't support U+0000 54-- use octet_length here so we don't get an odd unicode char in the 55-- output 56SELECT octet_length('"\uaBcD"'::jsonb::text); -- OK, uppercase and lower case both OK 57 58-- handling of unicode surrogate pairs 59 60SELECT octet_length((jsonb '{ "a": "\ud83d\ude04\ud83d\udc36" }' -> 'a')::text) AS correct_in_utf8; 61SELECT jsonb '{ "a": "\ud83d\ud83d" }' -> 'a'; -- 2 high surrogates in a row 62SELECT jsonb '{ "a": "\ude04\ud83d" }' -> 'a'; -- surrogates in wrong order 63SELECT jsonb '{ "a": "\ud83dX" }' -> 'a'; -- orphan high surrogate 64SELECT jsonb '{ "a": "\ude04X" }' -> 'a'; -- orphan low surrogate 65 66-- handling of simple unicode escapes 67 68SELECT jsonb '{ "a": "the Copyright \u00a9 sign" }' as correct_in_utf8; 69SELECT jsonb '{ "a": "dollar \u0024 character" }' as correct_everywhere; 70SELECT jsonb '{ "a": "dollar \\u0024 character" }' as not_an_escape; 71SELECT jsonb '{ "a": "null \u0000 escape" }' as fails; 72SELECT jsonb '{ "a": "null \\u0000 escape" }' as not_an_escape; 73 74SELECT jsonb '{ "a": "the Copyright \u00a9 sign" }' ->> 'a' as correct_in_utf8; 75SELECT jsonb '{ "a": "dollar \u0024 character" }' ->> 'a' as correct_everywhere; 76SELECT jsonb '{ "a": "dollar \\u0024 character" }' ->> 'a' as not_an_escape; 77SELECT jsonb '{ "a": "null \u0000 escape" }' ->> 'a' as fails; 78SELECT jsonb '{ "a": "null \\u0000 escape" }' ->> 'a' as not_an_escape; 79