Add the json-checker

README.md

@@ -2,6 +2,13 @@
 
 然后需要使用`c++17`来编译。
 
+```
+$ clang++ -std=c++17 json-check.cc
+```
+
+[json-checker](https://www.json.org/JSON_checker/)用于测试parse是否能正常工作。
+
+---
 假如我们有以下`json`文件，就叫做`m.json`吧。
 ```
 {

json-check.cc

@@ -0,0 +1,36 @@
+#include "json.h"
+
+#include <iostream>
+
+json::value jv;
+
+bool check(const std::string& filename)
+{
+    if (!jv.parsefile(filename)) {
+        std::cout << "check failed: " << filename.c_str() << ": " << jv.as_string().c_str() << "\n";
+        return false;
+    }
+    std::cout << "check pass " << filename.c_str() << "\n";
+    if (filename.rfind("fail1.json") != std::string::npos) {
+        std::cout << "(Now it can be any JSON value)\n";
+    } else if (filename.rfind("fail18.json") != std::string::npos) {
+        std::cout << "(We have no maximum nesting depth limit)\n";
+    }
+    return true;
+}
+
+int main()
+{
+    char buf[32];
+    std::string check_dir = "json-checker/";
+    std::cout << "============CHECK fail1.json ~ fail33.json============\n";
+    for (int i = 1; i <= 33; i++) {
+        sprintf(buf, "fail%d.json", i);
+        check(check_dir + buf);
+    }
+    std::cout << "============CHECK pass1.json ~ pass3.json============\n";
+    for (int i = 1; i <= 3; i++) {
+        sprintf(buf, "pass%d.json", i);
+        check(check_dir + buf);
+    }
+}

json-checker/fail1.json

@@ -0,0 +1 @@
+"A JSON payload should be an object or array, not a string."

json-checker/fail10.json

@@ -0,0 +1 @@
+{"Extra value after close": true} "misplaced quoted value"

json-checker/fail11.json

@@ -0,0 +1 @@
+{"Illegal expression": 1 + 2}

json-checker/fail12.json

@@ -0,0 +1 @@
+{"Illegal invocation": alert()}

json-checker/fail13.json

@@ -0,0 +1 @@
+{"Numbers cannot have leading zeroes": 013}

json-checker/fail14.json

@@ -0,0 +1 @@
+{"Numbers cannot be hex": 0x14}

json-checker/fail15.json

@@ -0,0 +1 @@
+["Illegal backslash escape: \x15"]

json-checker/fail16.json

@@ -0,0 +1 @@
+[\naked]

json-checker/fail17.json

@@ -0,0 +1 @@
+["Illegal backslash escape: \017"]

json-checker/fail18.json

@@ -0,0 +1 @@
+[[[[[[[[[[[[[[[[[[[["Too deep"]]]]]]]]]]]]]]]]]]]]

json-checker/fail19.json

@@ -0,0 +1 @@
+{"Missing colon" null}

json-checker/fail2.json

@@ -0,0 +1 @@
+["Unclosed array"

json-checker/fail20.json

@@ -0,0 +1 @@
+{"Double colon":: null}

json-checker/fail21.json

@@ -0,0 +1 @@
+{"Comma instead of colon", null}

json-checker/fail22.json

@@ -0,0 +1 @@
+["Colon instead of comma": false]

json-checker/fail23.json

@@ -0,0 +1 @@
+["Bad value", truth]

json-checker/fail24.json

@@ -0,0 +1 @@
+['single quote']

json-checker/fail25.json

@@ -0,0 +1 @@
+["	tab	character	in	string	"]

json-checker/fail26.json

@@ -0,0 +1 @@
+["tab\   character\   in\  string\  "]

json-checker/fail27.json

@@ -0,0 +1,2 @@
+["line
+break"]

json-checker/fail28.json

@@ -0,0 +1,2 @@
+["line\
+break"]

json-checker/fail29.json

@@ -0,0 +1 @@
+[0e]

json-checker/fail3.json

@@ -0,0 +1 @@
+{unquoted_key: "keys must be quoted"}

json-checker/fail30.json

@@ -0,0 +1 @@
+[0e+]

json-checker/fail31.json

@@ -0,0 +1 @@
+[0e+-1]

json-checker/fail32.json

@@ -0,0 +1 @@
+{"Comma instead if closing brace": true,

json-checker/fail33.json

@@ -0,0 +1 @@
+["mismatch"}

json-checker/fail4.json

@@ -0,0 +1 @@
+["extra comma",]

json-checker/fail5.json

@@ -0,0 +1 @@
+["double extra comma",,]

json-checker/fail6.json

@@ -0,0 +1 @@
+[   , "<-- missing value"]

json-checker/fail7.json

@@ -0,0 +1 @@
+["Comma after the close"],

json-checker/fail8.json

@@ -0,0 +1 @@
+["Extra close"]]

json-checker/fail9.json

@@ -0,0 +1 @@
+{"Extra comma": true,}

json-checker/pass1.json

@@ -0,0 +1,58 @@
+[
+    "JSON Test Pattern pass1",
+    {"object with 1 member":["array with 1 element"]},
+    {},
+    [],
+    -42,
+    true,
+    false,
+    null,
+    {
+        "integer": 1234567890,
+        "real": -9876.543210,
+        "e": 0.123456789e-12,
+        "E": 1.234567890E+34,
+        "":  23456789012E66,
+        "zero": 0,
+        "one": 1,
+        "space": " ",
+        "quote": "\"",
+        "backslash": "\\",
+        "controls": "\b\f\n\r\t",
+        "slash": "/ & \/",
+        "alpha": "abcdefghijklmnopqrstuvwyz",
+        "ALPHA": "ABCDEFGHIJKLMNOPQRSTUVWYZ",
+        "digit": "0123456789",
+        "0123456789": "digit",
+        "special": "`1~!@#$%^&*()_+-={':[,]}|;.</>?",
+        "hex": "\u0123\u4567\u89AB\uCDEF\uabcd\uef4A",
+        "true": true,
+        "false": false,
+        "null": null,
+        "array":[  ],
+        "object":{  },
+        "address": "50 St. James Street",
+        "url": "http://www.JSON.org/",
+        "comment": "// /* <!-- --",
+        "# -- --> */": " ",
+        " s p a c e d " :[1,2 , 3
+
+,
+
+4 , 5        ,          6           ,7        ],"compact":[1,2,3,4,5,6,7],
+        "jsontext": "{\"object with 1 member\":[\"array with 1 element\"]}",
+        "quotes": "&#34; \u0022 %22 0x22 034 &#x22;",
+        "\/\\\"\uCAFE\uBABE\uAB98\uFCDE\ubcda\uef4A\b\f\n\r\t`1~!@#$%^&*()_+-=[]{}|;:',./<>?"
+: "A key can be any string"
+    },
+    0.5 ,98.6
+,
+99.44
+,
+
+1066,
+1e1,
+0.1e1,
+1e-1,
+1e00,2e+00,2e-00
+,"rosebud"]

json-checker/pass2.json

@@ -0,0 +1 @@
+[[[[[[[[[[[[[[[[[[["Not too deep"]]]]]]]]]]]]]]]]]]]

json-checker/pass3.json

@@ -0,0 +1,6 @@
+{
+    "JSON Test Pattern pass3": {
+        "The outermost value": "must be an object or array.",
+        "In this test": "It is an object."
+    }
+}

json.h

@@ -7,8 +7,6 @@
 #include <variant>
 #include <fstream>
 
-#include <iostream>
-
 namespace json {
 
 enum ValueType {
@@ -265,6 +263,8 @@ enum error_code {
     invalid_value_type,
     invalid_object,
     invalid_array,
+    invalid_string_tab,
+    invalid_string_break,
     invalid_escape,
     invalid_unicode,
     invalid_unicode_surrogate,
@@ -342,9 +342,11 @@ class parser {
     object parse_object()
     {
         object o;
+        skipspace();
+        if (c == '}') return o;
+        backward();
         while (true) {
             skipspace();
-            if (c == '}') break;
             if (c != '"') throw invalid_object;
             auto key = parse_string();
             skipspace();
@@ -395,6 +397,8 @@ class parser {
             switch (nextchar()) {
             case '\\': parse_escape(s); break;
             case '\"': return s;
+            case '\t': throw invalid_string_tab;
+            case '\n': throw invalid_string_break;
             default: s.push_back(c); break;
             }
         }
@@ -559,6 +563,8 @@ const char *parser::get_error_string(error_code code)
         "invalid_value_type",
         "invalid_object",
         "invalid_array",
+        "invalid_string_tab",
+        "invalid_string_break",
         "invalid_escape",
         "invalid_unicode",
         "invalid_unicode_surrogate",
@@ -708,15 +714,15 @@ class writer {
                 // 1 bit is treated as an ASCII character,
                 // in which case UTF-8 is compatible with ASCII.
                 buf.push_back(c);
-            } else if (!(c & 0x20)) {
+            } else if ((c & 0xE0) == 0xC0) { // 110xxxxx
                 if (i + 1 >= s.size()) throw invalid_unicode;
                 char c2 = s[++i];
                 // xxx|xx xx|xxxx
                 put_unicode(0,
                             (c & 0x1C) >> 2,
                             ((c & 0x03) << 2) | ((c2 & 0x30) >> 4),
                             (c2 & 0x0F));
-            } else if (!(c & 0x10)) {
+            } else if ((c & 0xF0) == 0xE0) { // 1110xxxx
                 if (i + 2 >= s.size()) throw invalid_unicode;
                 char c2 = s[++i], c3 = s[++i];
                 // xxxx |xxxx|xx xx|xxxx
@@ -725,7 +731,7 @@ class writer {
                             ((c2 & 0x03) << 2) | ((c3 & 0x30) >> 4),
                             (c3 & 0x0F));
             } else { // SURROGATE-PAIR
-                assert(!(c & 0x08));
+                assert((c & 0xF8) == 0xF0); // 11110xxx
                 if (i + 3 >= s.size()) throw invalid_unicode;
                 char c2 = s[++i], c3 = s[++i], c4 = s[++i];
                 unsigned u = 0;