/src/jsonnet/core/string_utils.cpp

Source (jump to first uncovered line)
/*
Copyright 2015 Google Inc. All rights reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

#include <iomanip>

#include "static_error.h"
#include "string_utils.h"

namespace jsonnet::internal {

UString jsonnet_string_unparse(const UString &str, bool single)
{
    UStringStream ss;
    ss << (single ? U'\'' : U'\"');
    ss << jsonnet_string_escape(str, single);
    ss << (single ? U'\'' : U'\"');
    return ss.str();
}

UString jsonnet_string_escape(const UString &str, bool single)
{
    UStringStream ss;
    for (std::size_t i = 0; i < str.length(); ++i) {
        char32_t c = str[i];
        switch (c) {
            case U'\"': ss << (single ? U"\"" : U"\\\""); break;
            case U'\'': ss << (single ? U"\\\'" : U"\'"); break;
            case U'\\': ss << U"\\\\"; break;
            case U'\b': ss << U"\\b"; break;
            case U'\f': ss << U"\\f"; break;
            case U'\n': ss << U"\\n"; break;
            case U'\r': ss << U"\\r"; break;
            case U'\t': ss << U"\\t"; break;
            case U'\0': ss << U"\\u0000"; break;
            default: {
                if (c < 0x20 || (c >= 0x7f && c <= 0x9f)) {
                    // Unprintable, use \u
                    std::stringstream ss8;
                    ss8 << "\\u" << std::hex << std::setfill('0') << std::setw(4)
                        << (unsigned long)(c);
                    ss << decode_utf8(ss8.str());
                } else {
                    // Printable, write verbatim
                    ss << c;
                }
            }
        }
    }
    return ss.str();
}

unsigned long jsonnet_string_parse_unicode(const LocationRange &loc, const char32_t *c)
{
    unsigned long codepoint = 0;
    // Expect 4 hex digits.
    for (unsigned i = 0; i < 4; ++i) {
        auto x = (unsigned char)(c[i]);
        unsigned digit;
        if (x == '\0') {
            auto msg = "Truncated unicode escape sequence in string literal.";
            throw StaticError(loc, msg);
        } else if (x >= '0' && x <= '9') {
            digit = x - '0';
        } else if (x >= 'a' && x <= 'f') {
            digit = x - 'a' + 10;
        } else if (x >= 'A' && x <= 'F') {
            digit = x - 'A' + 10;
        } else {
            std::stringstream ss;
            ss << "Malformed unicode escape character, "
               << "should be hex: '" << x << "'";
            throw StaticError(loc, ss.str());
        }
        codepoint *= 16;
        codepoint += digit;
    }
    return codepoint;
}

bool is_bmp_codepoint(const unsigned long codepoint)
{
    return codepoint < 0xd800 || (codepoint >= 0xe000 && codepoint < 0x10000);
}

char32_t decode_utf16_surrogates(const LocationRange &loc, const unsigned long high, const unsigned long low)
{
    if (high >= 0xd800 && high < 0xdc00 && low >= 0xdc00 && low < 0xe000) {
        return 0x10000 + ((high & 0x03ff) << 10) + (low & 0x03ff);
    } else {
        std::stringstream ss;
        ss << "Invalid UTF-16 bytes";
        throw StaticError(loc, ss.str());
    }
}

UString jsonnet_string_unescape(const LocationRange &loc, const UString &s)
{
    UString r;
    const char32_t *s_ptr = s.c_str();
    for (const char32_t *c = s_ptr; *c != U'\0'; ++c) {
        switch (*c) {
            case '\\':
                switch (*(++c)) {
                    case '"':
                    case '\'': r += *c; break;

                    case '\\': r += *c; break;

                    case '/': r += *c; break;

                    case 'b': r += '\b'; break;

                    case 'f': r += '\f'; break;

                    case 'n': r += '\n'; break;

                    case 'r': r += '\r'; break;

                    case 't': r += '\t'; break;

                    case 'u': {
                        ++c;  // Consume the 'u'.
                        unsigned long codepoint = jsonnet_string_parse_unicode(loc, c);

                        // Leave us on the last char, ready for the ++c at
                        // the outer for loop.
                        c += 3;
                        if (!is_bmp_codepoint(codepoint)) {
                           if (*(++c) != '\\') {
                                std::stringstream ss;
                                ss << "Invalid non-BMP Unicode escape in string literal";
                                throw StaticError(loc, ss.str());
                           }
                           if (*(++c) != 'u') {
                                std::stringstream ss;
                                ss << "Invalid non-BMP Unicode escape in string literal";
                                throw StaticError(loc, ss.str());
                           }
                           ++c;
                           unsigned long codepoint2 = jsonnet_string_parse_unicode(loc, c);
                           c += 3;
                           codepoint = decode_utf16_surrogates(loc, codepoint, codepoint2);
                       }
                       r += codepoint;
                    } break;

                    case '\0': {
                        auto msg = "Truncated escape sequence in string literal.";
                        throw StaticError(loc, msg);
                    }

                    default: {
                        std::stringstream ss;
                        std::string utf8;
                        encode_utf8(*c, utf8);
                        ss << "Unknown escape sequence in string literal: '" << utf8 << "'";
                        throw StaticError(loc, ss.str());
                    }
                }
                break;

            default:
                // Just a regular letter.
                r += *c;
        }
    }
    return r;
}

}  // namespace jsonnet::internal

Line	Count	Source (jump to first uncovered line)
1		/*
2		Copyright 2015 Google Inc. All rights reserved.
3
4		Licensed under the Apache License, Version 2.0 (the "License");
5		you may not use this file except in compliance with the License.
6		You may obtain a copy of the License at
7
8		http://www.apache.org/licenses/LICENSE-2.0
9
10		Unless required by applicable law or agreed to in writing, software
11		distributed under the License is distributed on an "AS IS" BASIS,
12		WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13		See the License for the specific language governing permissions and
14		limitations under the License.
15		*/
16
17		#include <iomanip>
18
19		#include "static_error.h"
20		#include "string_utils.h"
21
22		namespace jsonnet::internal {
23
24		UString jsonnet_string_unparse(const UString &str, bool single)
25	975k	{
26	975k	UStringStream ss;
27	975k	ss << (single ? U'\'' : U'\"');
28	975k	ss << jsonnet_string_escape(str, single);
29	975k	ss << (single ? U'\'' : U'\"');
30	975k	return ss.str();
31	975k	}
32
33		UString jsonnet_string_escape(const UString &str, bool single)
34	975k	{
35	975k	UStringStream ss;
36	1.60G	for (std::size_t i = 0; i < str.length(); ++i) {
37	1.60G	char32_t c = str[i];
38	1.60G	switch (c) {
39	1.54M	case U'\"': ss << (single ? U"\"" : U"\\\""); break;
40	2.15M	case U'\'': ss << (single ? U"\\\'" : U"\'"); break;
41	1.30G	case U'\\': ss << U"\\\\"; break;
42	1.08M	case U'\b': ss << U"\\b"; break;
43	519k	case U'\f': ss << U"\\f"; break;
44	37.4M	case U'\n': ss << U"\\n"; break;
45	3.48M	case U'\r': ss << U"\\r"; break;
46	1.44M	case U'\t': ss << U"\\t"; break;
47	7.51k	case U'\0': ss << U"\\u0000"; break;
48	254M	default: {
49	254M	if (c < 0x20 \|\| (c >= 0x7f && c <= 0x9f)) {
50		// Unprintable, use \u
51	78.7M	std::stringstream ss8;
52	78.7M	ss8 << "\\u" << std::hex << std::setfill('0') << std::setw(4)
53	78.7M	<< (unsigned long)(c);
54	78.7M	ss << decode_utf8(ss8.str());
55	175M	} else {
56		// Printable, write verbatim
57	175M	ss << c;
58	175M	}
59	254M	}
60	1.60G	}
61	1.60G	}
62	975k	return ss.str();
63	975k	}
64
65		unsigned long jsonnet_string_parse_unicode(const LocationRange &loc, const char32_t *c)
66	54.5k	{
67	54.5k	unsigned long codepoint = 0;
68		// Expect 4 hex digits.
69	272k	for (unsigned i = 0; i < 4; ++i) {
70	218k	auto x = (unsigned char)(c[i]);
71	218k	unsigned digit;
72	218k	if (x == '\0') {
73	58	auto msg = "Truncated unicode escape sequence in string literal.";
74	58	throw StaticError(loc, msg);
75	217k	} else if (x >= '0' && x <= '9') {
76	173k	digit = x - '0';
77	173k	} else if (x >= 'a' && x <= 'f') {
78	16.3k	digit = x - 'a' + 10;
79	28.5k	} else if (x >= 'A' && x <= 'F') {
80	28.3k	digit = x - 'A' + 10;
81	28.3k	} else {
82	174	std::stringstream ss;
83	174	ss << "Malformed unicode escape character, "
84	174	<< "should be hex: '" << x << "'";
85	174	throw StaticError(loc, ss.str());
86	174	}
87	217k	codepoint *= 16;
88	217k	codepoint += digit;
89	217k	}
90	54.3k	return codepoint;
91	54.5k	}
92
93		bool is_bmp_codepoint(const unsigned long codepoint)
94	52.1k	{
95	52.1k	return codepoint < 0xd800 \|\| (codepoint >= 0xe000 && codepoint < 0x10000);
96	52.1k	}
97
98		char32_t decode_utf16_surrogates(const LocationRange &loc, const unsigned long high, const unsigned long low)
99	2.18k	{
100	2.18k	if (high >= 0xd800 && high < 0xdc00 && low >= 0xdc00 && low < 0xe000) {
101	2.14k	return 0x10000 + ((high & 0x03ff) << 10) + (low & 0x03ff);
102	2.14k	} else {
103	48	std::stringstream ss;
104	48	ss << "Invalid UTF-16 bytes";
105	48	throw StaticError(loc, ss.str());
106	48	}
107	2.18k	}
108
109		UString jsonnet_string_unescape(const LocationRange &loc, const UString &s)
110	22.4M	{
111	22.4M	UString r;
112	22.4M	const char32_t *s_ptr = s.c_str();
113	290M	for (const char32_t c = s_ptr; c != U'\0'; ++c) {
114	267M	switch (*c) {
115	1.09M	case '\\':
116	1.09M	switch (*(++c)) {
117	42.0k	case '"':
118	44.7k	case '\'': r += *c; break;
119
120	270k	case '\\': r += *c; break;
121
122	5.38k	case '/': r += *c; break;
123
124	26.6k	case 'b': r += '\b'; break;
125
126	62.6k	case 'f': r += '\f'; break;
127
128	516k	case 'n': r += '\n'; break;
129
130	39.2k	case 'r': r += '\r'; break;
131
132	78.0k	case 't': r += '\t'; break;
133
134	52.3k	case 'u': {
135	52.3k	++c; // Consume the 'u'.
136	52.3k	unsigned long codepoint = jsonnet_string_parse_unicode(loc, c);
137
138		// Leave us on the last char, ready for the ++c at
139		// the outer for loop.
140	52.3k	c += 3;
141	52.3k	if (!is_bmp_codepoint(codepoint)) {
142	2.27k	if (*(++c) != '\\') {
143	53	std::stringstream ss;
144	53	ss << "Invalid non-BMP Unicode escape in string literal";
145	53	throw StaticError(loc, ss.str());
146	53	}
147	2.22k	if (*(++c) != 'u') {
148	12	std::stringstream ss;
149	12	ss << "Invalid non-BMP Unicode escape in string literal";
150	12	throw StaticError(loc, ss.str());
151	12	}
152	2.21k	++c;
153	2.21k	unsigned long codepoint2 = jsonnet_string_parse_unicode(loc, c);
154	2.21k	c += 3;
155	2.21k	codepoint = decode_utf16_surrogates(loc, codepoint, codepoint2);
156	2.21k	}
157	52.3k	r += codepoint;
158	52.3k	} break;
159
160	4	case '\0': {
161	4	auto msg = "Truncated escape sequence in string literal.";
162	4	throw StaticError(loc, msg);
163	52.3k	}
164
165	308	default: {
166	308	std::stringstream ss;
167	308	std::string utf8;
168	308	encode_utf8(*c, utf8);
169	308	ss << "Unknown escape sequence in string literal: '" << utf8 << "'";
170	308	throw StaticError(loc, ss.str());
171	52.3k	}
172	1.09M	}
173	1.09M	break;
174
175	266M	default:
176		// Just a regular letter.
177	266M	r += *c;
178	267M	}
179	267M	}
180	22.4M	return r;
181	22.4M	}
182
183		} // namespace jsonnet::internal

Coverage Report

Created: 2023-09-25 07:10