… | |
… | |
56 | does so, and even documents what "correct" means. |
56 | does so, and even documents what "correct" means. |
57 | |
57 | |
58 | * round-trip integrity |
58 | * round-trip integrity |
59 | |
59 | |
60 | When you serialise a perl data structure using only data types |
60 | When you serialise a perl data structure using only data types |
61 | supported by JSON, the deserialised data structure is identical on |
61 | supported by JSON and Perl, the deserialised data structure is |
62 | the Perl level. (e.g. the string "2.0" doesn't suddenly become "2" |
62 | identical on the Perl level. (e.g. the string "2.0" doesn't suddenly |
63 | just because it looks like a number). There minor *are* exceptions |
63 | become "2" just because it looks like a number). There *are* minor |
64 | to this, read the MAPPING section below to learn about those. |
64 | exceptions to this, read the MAPPING section below to learn about |
|
|
65 | those. |
65 | |
66 | |
66 | * strict checking of JSON correctness |
67 | * strict checking of JSON correctness |
67 | |
68 | |
68 | There is no guessing, no generating of illegal JSON texts by |
69 | There is no guessing, no generating of illegal JSON texts by |
69 | default, and only JSON is accepted as input by default (the latter |
70 | default, and only JSON is accepted as input by default (the latter |
… | |
… | |
635 | calls). |
636 | calls). |
636 | |
637 | |
637 | JSON::XS will only attempt to parse the JSON text once it is sure it has |
638 | JSON::XS will only attempt to parse the JSON text once it is sure it has |
638 | enough text to get a decisive result, using a very simple but truly |
639 | enough text to get a decisive result, using a very simple but truly |
639 | incremental parser. This means that it sometimes won't stop as early as |
640 | incremental parser. This means that it sometimes won't stop as early as |
640 | the full parser, for example, it doesn't detect parenthese mismatches. |
641 | the full parser, for example, it doesn't detect mismatched parentheses. |
641 | The only thing it guarantees is that it starts decoding as soon as a |
642 | The only thing it guarantees is that it starts decoding as soon as a |
642 | syntactically valid JSON text has been seen. This means you need to set |
643 | syntactically valid JSON text has been seen. This means you need to set |
643 | resource limits (e.g. "max_size") to ensure the parser will stop parsing |
644 | resource limits (e.g. "max_size") to ensure the parser will stop parsing |
644 | in the presence if syntax errors. |
645 | in the presence if syntax errors. |
645 | |
646 | |
… | |
… | |
669 | otherwise. For this to work, there must be no separators between the |
670 | otherwise. For this to work, there must be no separators between the |
670 | JSON objects or arrays, instead they must be concatenated |
671 | JSON objects or arrays, instead they must be concatenated |
671 | back-to-back. If an error occurs, an exception will be raised as in |
672 | back-to-back. If an error occurs, an exception will be raised as in |
672 | the scalar context case. Note that in this case, any |
673 | the scalar context case. Note that in this case, any |
673 | previously-parsed JSON texts will be lost. |
674 | previously-parsed JSON texts will be lost. |
|
|
675 | |
|
|
676 | Example: Parse some JSON arrays/objects in a given string and return |
|
|
677 | them. |
|
|
678 | |
|
|
679 | my @objs = JSON::XS->new->incr_parse ("[5][7][1,2]"); |
674 | |
680 | |
675 | $lvalue_string = $json->incr_text |
681 | $lvalue_string = $json->incr_text |
676 | This method returns the currently stored JSON fragment as an lvalue, |
682 | This method returns the currently stored JSON fragment as an lvalue, |
677 | that is, you can manipulate it. This *only* works when a preceding |
683 | that is, you can manipulate it. This *only* works when a preceding |
678 | call to "incr_parse" in *scalar context* successfully returned an |
684 | call to "incr_parse" in *scalar context* successfully returned an |
… | |
… | |
893 | Numbers containing a fractional or exponential part will always be |
899 | Numbers containing a fractional or exponential part will always be |
894 | represented as numeric (floating point) values, possibly at a loss |
900 | represented as numeric (floating point) values, possibly at a loss |
895 | of precision (in which case you might lose perfect roundtripping |
901 | of precision (in which case you might lose perfect roundtripping |
896 | ability, but the JSON number will still be re-encoded as a JSON |
902 | ability, but the JSON number will still be re-encoded as a JSON |
897 | number). |
903 | number). |
|
|
904 | |
|
|
905 | Note that precision is not accuracy - binary floating point values |
|
|
906 | cannot represent most decimal fractions exactly, and when converting |
|
|
907 | from and to floating point, JSON::XS only guarantees precision up to |
|
|
908 | but not including the leats significant bit. |
898 | |
909 | |
899 | true, false |
910 | true, false |
900 | These JSON atoms become "JSON::XS::true" and "JSON::XS::false", |
911 | These JSON atoms become "JSON::XS::true" and "JSON::XS::false", |
901 | respectively. They are overloaded to act almost exactly like the |
912 | respectively. They are overloaded to act almost exactly like the |
902 | numbers 1 and 0. You can check whether a scalar is a JSON boolean by |
913 | numbers 1 and 0. You can check whether a scalar is a JSON boolean by |
… | |
… | |
979 | |
990 | |
980 | You can not currently force the type in other, less obscure, ways. |
991 | You can not currently force the type in other, less obscure, ways. |
981 | Tell me if you need this capability (but don't forget to explain why |
992 | Tell me if you need this capability (but don't forget to explain why |
982 | it's needed :). |
993 | it's needed :). |
983 | |
994 | |
|
|
995 | Note that numerical precision has the same meaning as under Perl (so |
|
|
996 | binary to decimal conversion follows the same rules as in Perl, |
|
|
997 | which can differ to other languages). Also, your perl interpreter |
|
|
998 | might expose extensions to the floating point numbers of your |
|
|
999 | platform, such as infinities or NaN's - these cannot be represented |
|
|
1000 | in JSON, and it is an error to pass those in. |
|
|
1001 | |
984 | ENCODING/CODESET FLAG NOTES |
1002 | ENCODING/CODESET FLAG NOTES |
985 | The interested reader might have seen a number of flags that signify |
1003 | The interested reader might have seen a number of flags that signify |
986 | encodings or codesets - "utf8", "latin1" and "ascii". There seems to be |
1004 | encodings or codesets - "utf8", "latin1" and "ascii". There seems to be |
987 | some confusion on what these do, so here is a short comparison: |
1005 | some confusion on what these do, so here is a short comparison: |
988 | |
1006 | |
… | |
… | |
1125 | characters as well - using "eval" naively simply *will* cause problems. |
1143 | characters as well - using "eval" naively simply *will* cause problems. |
1126 | |
1144 | |
1127 | Another problem is that some javascript implementations reserve some |
1145 | Another problem is that some javascript implementations reserve some |
1128 | property names for their own purposes (which probably makes them |
1146 | property names for their own purposes (which probably makes them |
1129 | non-ECMAscript-compliant). For example, Iceweasel reserves the |
1147 | non-ECMAscript-compliant). For example, Iceweasel reserves the |
1130 | "__proto__" property name for it's own purposes. |
1148 | "__proto__" property name for its own purposes. |
1131 | |
1149 | |
1132 | If that is a problem, you could parse try to filter the resulting JSON |
1150 | If that is a problem, you could parse try to filter the resulting JSON |
1133 | output for these property strings, e.g.: |
1151 | output for these property strings, e.g.: |
1134 | |
1152 | |
1135 | $json =~ s/"__proto__"\s*:/"__proto__renamed":/g; |
1153 | $json =~ s/"__proto__"\s*:/"__proto__renamed":/g; |
… | |
… | |
1183 | (which is not that difficult or long) and finally make YAML |
1201 | (which is not that difficult or long) and finally make YAML |
1184 | compatible to it, and educating users about the changes, instead of |
1202 | compatible to it, and educating users about the changes, instead of |
1185 | spreading lies about the real compatibility for many *years* and |
1203 | spreading lies about the real compatibility for many *years* and |
1186 | trying to silence people who point out that it isn't true. |
1204 | trying to silence people who point out that it isn't true. |
1187 | |
1205 | |
1188 | Addendum/2009: the YAML 1.2 spec is still incomaptible with JSON, |
1206 | Addendum/2009: the YAML 1.2 spec is still incompatible with JSON, |
1189 | even though the incompatibilities have been documented (and are |
1207 | even though the incompatibilities have been documented (and are |
1190 | known to Brian) for many years and the spec makes explicit claims |
1208 | known to Brian) for many years and the spec makes explicit claims |
1191 | that YAML is a superset of JSON. It would be so easy to fix, but |
1209 | that YAML is a superset of JSON. It would be so easy to fix, but |
1192 | apparently, bullying and corrupting userdata is so much easier. |
1210 | apparently, bullying people and corrupting userdata is so much |
|
|
1211 | easier. |
1193 | |
1212 | |
1194 | SPEED |
1213 | SPEED |
1195 | It seems that JSON::XS is surprisingly fast, as shown in the following |
1214 | It seems that JSON::XS is surprisingly fast, as shown in the following |
1196 | tables. They have been generated with the help of the "eg/bench" program |
1215 | tables. They have been generated with the help of the "eg/bench" program |
1197 | in the JSON::XS distribution, to make it easy to compare on your own |
1216 | in the JSON::XS distribution, to make it easy to compare on your own |
… | |
… | |
1201 | single-line JSON string (also available at |
1220 | single-line JSON string (also available at |
1202 | <http://dist.schmorp.de/misc/json/short.json>). |
1221 | <http://dist.schmorp.de/misc/json/short.json>). |
1203 | |
1222 | |
1204 | {"method": "handleMessage", "params": ["user1", |
1223 | {"method": "handleMessage", "params": ["user1", |
1205 | "we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7, |
1224 | "we were just talking"], "id": null, "array":[1,11,234,-5,1e5,1e7, |
1206 | true, false]} |
1225 | 1, 0]} |
1207 | |
1226 | |
1208 | It shows the number of encodes/decodes per second (JSON::XS uses the |
1227 | It shows the number of encodes/decodes per second (JSON::XS uses the |
1209 | functional interface, while JSON::XS/2 uses the OO interface with |
1228 | functional interface, while JSON::XS/2 uses the OO interface with |
1210 | pretty-printing and hashkey sorting enabled, JSON::XS/3 enables shrink). |
1229 | pretty-printing and hashkey sorting enabled, JSON::XS/3 enables shrink. |
1211 | Higher is better: |
1230 | JSON::DWIW/DS uses the deserialise function, while JSON::DWIW::FJ uses |
|
|
1231 | the from_json method). Higher is better: |
1212 | |
1232 | |
1213 | module | encode | decode | |
1233 | module | encode | decode | |
1214 | -----------|------------|------------| |
1234 | --------------|------------|------------| |
1215 | JSON 1.x | 4990.842 | 4088.813 | |
1235 | JSON::DWIW/DS | 86302.551 | 102300.098 | |
1216 | JSON::DWIW | 51653.990 | 71575.154 | |
1236 | JSON::DWIW/FJ | 86302.551 | 75983.768 | |
1217 | JSON::PC | 65948.176 | 74631.744 | |
1237 | JSON::PP | 15827.562 | 6638.658 | |
1218 | JSON::PP | 8931.652 | 3817.168 | |
1238 | JSON::Syck | 63358.066 | 47662.545 | |
1219 | JSON::Syck | 24877.248 | 27776.848 | |
1239 | JSON::XS | 511500.488 | 511500.488 | |
1220 | JSON::XS | 388361.481 | 227951.304 | |
1240 | JSON::XS/2 | 291271.111 | 388361.481 | |
1221 | JSON::XS/2 | 227951.304 | 218453.333 | |
1241 | JSON::XS/3 | 361577.931 | 361577.931 | |
1222 | JSON::XS/3 | 338250.323 | 218453.333 | |
1242 | Storable | 66788.280 | 265462.278 | |
1223 | Storable | 16500.016 | 135300.129 | |
|
|
1224 | -----------+------------+------------+ |
1243 | --------------+------------+------------+ |
1225 | |
1244 | |
1226 | That is, JSON::XS is about five times faster than JSON::DWIW on |
1245 | That is, JSON::XS is almost six times faster than JSON::DWIW on |
1227 | encoding, about three times faster on decoding, and over forty times |
1246 | encoding, about five times faster on decoding, and over thirty to |
1228 | faster than JSON, even with pretty-printing and key sorting. It also |
1247 | seventy times faster than JSON's pure perl implementation. It also |
1229 | compares favourably to Storable for small amounts of data. |
1248 | compares favourably to Storable for small amounts of data. |
1230 | |
1249 | |
1231 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
1250 | Using a longer test string (roughly 18KB, generated from Yahoo! Locals |
1232 | search API (<http://dist.schmorp.de/misc/json/long.json>). |
1251 | search API (<http://dist.schmorp.de/misc/json/long.json>). |
1233 | |
1252 | |
1234 | module | encode | decode | |
1253 | module | encode | decode | |
1235 | -----------|------------|------------| |
1254 | --------------|------------|------------| |
1236 | JSON 1.x | 55.260 | 34.971 | |
1255 | JSON::DWIW/DS | 1647.927 | 2673.916 | |
1237 | JSON::DWIW | 825.228 | 1082.513 | |
1256 | JSON::DWIW/FJ | 1630.249 | 2596.128 | |
1238 | JSON::PC | 3571.444 | 2394.829 | |
|
|
1239 | JSON::PP | 210.987 | 32.574 | |
1257 | JSON::PP | 400.640 | 62.311 | |
1240 | JSON::Syck | 552.551 | 787.544 | |
1258 | JSON::Syck | 1481.040 | 1524.869 | |
1241 | JSON::XS | 5780.463 | 4854.519 | |
1259 | JSON::XS | 20661.596 | 9541.183 | |
1242 | JSON::XS/2 | 3869.998 | 4798.975 | |
1260 | JSON::XS/2 | 10683.403 | 9416.938 | |
1243 | JSON::XS/3 | 5862.880 | 4798.975 | |
1261 | JSON::XS/3 | 20661.596 | 9400.054 | |
1244 | Storable | 4445.002 | 5235.027 | |
1262 | Storable | 19765.806 | 10000.725 | |
1245 | -----------+------------+------------+ |
1263 | --------------+------------+------------+ |
1246 | |
1264 | |
1247 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
1265 | Again, JSON::XS leads by far (except for Storable which non-surprisingly |
1248 | decodes faster). |
1266 | decodes a bit faster). |
1249 | |
1267 | |
1250 | On large strings containing lots of high Unicode characters, some |
1268 | On large strings containing lots of high Unicode characters, some |
1251 | modules (such as JSON::PC) seem to decode faster than JSON::XS, but the |
1269 | modules (such as JSON::PC) seem to decode faster than JSON::XS, but the |
1252 | result will be broken due to missing (or wrong) Unicode handling. Others |
1270 | result will be broken due to missing (or wrong) Unicode handling. Others |
1253 | refuse to decode or encode properly, so it was impossible to prepare a |
1271 | refuse to decode or encode properly, so it was impossible to prepare a |