/src/jsonnet/core/string_utils.cpp

Source
/*
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
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	1.29M	{
26	1.29M	UStringStream ss;
27	1.29M	ss << (single ? U'\'' : U'\"');
28	1.29M	ss << jsonnet_string_escape(str, single);
29	1.29M	ss << (single ? U'\'' : U'\"');
30	1.29M	return ss.str();
31	1.29M	}
32
33		UString jsonnet_string_escape(const UString &str, bool single)
34	1.29M	{
35	1.29M	UStringStream ss;
36	2.32G	for (std::size_t i = 0; i < str.length(); ++i) {
37	2.32G	char32_t c = str[i];
38	2.32G	switch (c) {
39	1.28M	case U'\"': ss << (single ? U"\"" : U"\\\""); break;
40	1.26M	case U'\'': ss << (single ? U"\\\'" : U"\'"); break;
41	2.15G	case U'\\': ss << U"\\\\"; break;
42	1.41M	case U'\b': ss << U"\\b"; break;
43	1.66M	case U'\f': ss << U"\\f"; break;
44	19.5M	case U'\n': ss << U"\\n"; break;
45	721k	case U'\r': ss << U"\\r"; break;
46	2.12M	case U'\t': ss << U"\\t"; break;
47	169k	case U'\0': ss << U"\\u0000"; break;
48	142M	default: {
49	142M	if (c < 0x20 \|\| (c >= 0x7f && c <= 0x9f)) {
50		// Unprintable, use \u
51	27.8M	std::stringstream ss8;
52	27.8M	ss8 << "\\u" << std::hex << std::setfill('0') << std::setw(4)
53	27.8M	<< (unsigned long)(c);
54	27.8M	ss << decode_utf8(ss8.str());
55	114M	} else {
56		// Printable, write verbatim
57	114M	ss << c;
58	114M	}
59	142M	}
60	2.32G	}
61	2.32G	}
62	1.29M	return ss.str();
63	1.29M	}
64
65		unsigned long jsonnet_string_parse_unicode(const LocationRange &loc, const char32_t *c)
66	70.4k	{
67	70.4k	unsigned long codepoint = 0;
68		// Expect 4 hex digits.
69	351k	for (unsigned i = 0; i < 4; ++i) {
70	281k	auto x = (unsigned char)(c[i]);
71	281k	unsigned digit;
72	281k	if (x == '\0') {
73	54	auto msg = "Truncated unicode escape sequence in string literal.";
74	54	throw StaticError(loc, msg);
75	281k	} else if (x >= '0' && x <= '9') {
76	220k	digit = x - '0';
77	220k	} else if (x >= 'a' && x <= 'f') {
78	25.9k	digit = x - 'a' + 10;
79	35.1k	} else if (x >= 'A' && x <= 'F') {
80	34.9k	digit = x - 'A' + 10;
81	34.9k	} else {
82	117	std::stringstream ss;
83	117	ss << "Malformed unicode escape character, "
84	117	<< "should be hex: '" << x << "'";
85	117	throw StaticError(loc, ss.str());
86	117	}
87	281k	codepoint *= 16;
88	281k	codepoint += digit;
89	281k	}
90	70.3k	return codepoint;
91	70.4k	}
92
93		bool is_bmp_codepoint(const unsigned long codepoint)
94	69.2k	{
95	69.2k	return codepoint < 0xd800 \|\| (codepoint >= 0xe000 && codepoint < 0x10000);
96	69.2k	}
97
98		char32_t decode_utf16_surrogates(const LocationRange &loc, const unsigned long high, const unsigned long low)
99	1.10k	{
100	1.10k	if (high >= 0xd800 && high < 0xdc00 && low >= 0xdc00 && low < 0xe000) {
101	1.06k	return 0x10000 + ((high & 0x03ff) << 10) + (low & 0x03ff);
102	1.06k	} else {
103	41	std::stringstream ss;
104	41	ss << "Invalid UTF-16 bytes";
105	41	throw StaticError(loc, ss.str());
106	41	}
107	1.10k	}
108
109		UString jsonnet_string_unescape(const LocationRange &loc, const UString &s)
110	10.3M	{
111	10.3M	UString r;
112	10.3M	const char32_t *s_ptr = s.c_str();
113	190M	for (const char32_t c = s_ptr; c != U'\0'; ++c) {
114	180M	switch (*c) {
115	1.26M	case '\\':
116	1.26M	switch (*(++c)) {
117	46.8k	case '"':
118	60.5k	case '\'': r += *c; break;
119
120	260k	case '\\': r += *c; break;
121
122	7.27k	case '/': r += *c; break;
123
124	31.8k	case 'b': r += '\b'; break;
125
126	85.3k	case 'f': r += '\f'; break;
127
128	605k	case 'n': r += '\n'; break;
129
130	44.7k	case 'r': r += '\r'; break;
131
132	99.2k	case 't': r += '\t'; break;
133
134	69.3k	case 'u': {
135	69.3k	++c; // Consume the 'u'.
136	69.3k	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	69.3k	c += 3;
141	69.3k	if (!is_bmp_codepoint(codepoint)) {
142	1.17k	if (*(++c) != '\\') {
143	42	std::stringstream ss;
144	42	ss << "Invalid non-BMP Unicode escape in string literal";
145	42	throw StaticError(loc, ss.str());
146	42	}
147	1.13k	if (*(++c) != 'u') {
148	17	std::stringstream ss;
149	17	ss << "Invalid non-BMP Unicode escape in string literal";
150	17	throw StaticError(loc, ss.str());
151	17	}
152	1.11k	++c;
153	1.11k	unsigned long codepoint2 = jsonnet_string_parse_unicode(loc, c);
154	1.11k	c += 3;
155	1.11k	codepoint = decode_utf16_surrogates(loc, codepoint, codepoint2);
156	1.11k	}
157	69.3k	r += codepoint;
158	69.3k	} break;
159
160	11	case '\0': {
161	11	auto msg = "Truncated escape sequence in string literal.";
162	11	throw StaticError(loc, msg);
163	69.3k	}
164
165	397	default: {
166	397	std::stringstream ss;
167	397	std::string utf8;
168	397	encode_utf8(*c, utf8);
169	397	ss << "Unknown escape sequence in string literal: '" << utf8 << "'";
170	397	throw StaticError(loc, ss.str());
171	69.3k	}
172	1.26M	}
173	1.26M	break;
174
175	179M	default:
176		// Just a regular letter.
177	179M	r += *c;
178	180M	}
179	180M	}
180	10.3M	return r;
181	10.3M	}
182
183		} // namespace jsonnet::internal

Coverage Report

Created: 2026-05-30 06:23