File: generate_permission_element_grd.py

package info (click to toggle)
chromium 138.0.7204.183-1
  • links: PTS, VCS
  • area: main
  • in suites: trixie
  • size: 6,071,908 kB
  • sloc: cpp: 34,937,088; ansic: 7,176,967; javascript: 4,110,704; python: 1,419,953; asm: 946,768; xml: 739,971; pascal: 187,324; sh: 89,623; perl: 88,663; objc: 79,944; sql: 50,304; cs: 41,786; fortran: 24,137; makefile: 21,806; php: 13,980; tcl: 13,166; yacc: 8,925; ruby: 7,485; awk: 3,720; lisp: 3,096; lex: 1,327; ada: 727; jsp: 228; sed: 36
file content (233 lines) | stat: -rw-r--r-- 9,374 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
# Copyright 2024 The Chromium Authors
# Use of this source code is governed by a BSD-style license that can be
# found in the LICENSE file.
import glob
import os
import re
import sys
from xml.dom.minidom import parse, parseString

_SRC_PATH = os.path.abspath(
    os.path.join(os.path.dirname(__file__), '..', '..', '..', '..', '..'))
sys.path.append(os.path.join(_SRC_PATH, 'tools', 'grit'))
from grit.extern import tclib

kGrdTemplate = '''<?xml version="1.0" encoding="utf-8"?>
<!--
This file contains all Permission element strings in all locales.
This is a generated grd file.
The script to generate the grd file is located at
third_party/blink/renderer/build/scripts/generate_permission_element_grd.py
-->
<grit base_dir="." latest_public_release="0" current_release="1"
    source_lang_id="en" enc_check="möl">
<outputs>
    <output filename="grit/permission_element_generated_strings.h" type="rc_header">
      <emit emit_type='prepend'></emit>
    </output>
    <output filename="permission_element_generated_strings.pak" type="data_package" />
</outputs>
<release seq="1" allow_pseudo="false">
    <messages fallback_to_english="true">
    </messages>
</release>
</grit>
'''

kStringMapCcPrefix = '''// Copyright 2024 The Chromium Authors
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

// Generated by third_party/blink/renderer/build/scripts/generate_permission_element_grd.py

#include <stdint.h>

#include <optional>
#include <string_view>
#include <utility>

#include "base/containers/fixed_flat_map.h"
#include "base/containers/fixed_flat_set.h"
#include "base/strings/string_slice.h"
#include "third_party/blink/renderer/core/html/html_permission_element_strings_map.h"
#include "third_party/blink/public/strings/grit/permission_element_generated_strings.h"
#include "third_party/blink/public/strings/grit/permission_element_strings.h"

namespace blink {

namespace {

static constexpr char kLanguages[] =
'''
kStringMapCcMidfix = '''

using LangStringSlice = base::subtle::StringSlice<sizeof(kLanguages), kLanguages>;

// In C++20, pairs required exactly-matching types to be comparable, i.e.
// std::pair<T1, T2> cannot be compared against std::pair<U1, U2>, even if T1
// and T2 are comparable against U1 and U2. This was addressed in C++23 in
// https://cplusplus.github.io/LWG/issue3865. However, older versions of
// libstdc++ do not support this, so the lookup needs to be wrapped with a
// type that is heterogeneously comparable with the key pair type...
struct LookupHelper {
  std::string_view lang;
  uint16_t id;

  // The style guide requires operator== to be defined if operator<=> is
  // defined. However, operator== is never actually used, which causes a
  // compiler warning. Explicitly delete it instead... if it's ever needed,
  // the build will start failing.
  friend constexpr bool operator==(LookupHelper lhs,
                                   std::pair<LangStringSlice, uint16_t> rhs) = delete;
  friend constexpr auto operator<=>(LookupHelper lhs,
                                    std::pair<LangStringSlice, uint16_t> rhs) {
    return lhs.lang != rhs.first ?  lhs.lang <=> rhs.first : lhs.id <=> rhs.second;
  }
};

}  // namespace

std::optional<uint16_t> GetPermissionElementMessageId(
    std::string_view language_code,
    uint16_t base_message) {
  static constexpr auto kMessageIds =
      base::MakeFixedFlatMap<std::pair<LangStringSlice, uint16_t>, uint16_t>({
'''
kStringMapCcSuffix = '''
      });

  auto message = kMessageIds.find(
      LookupHelper{language_code, static_cast<uint16_t>(base_message)});
  return message == kMessageIds.end()
      ? std::nullopt
      : std::optional(message->second);
}

}  // namespace blink
'''


def get_message_id_map_and_orderings(input_base_dir):
    dom = parse(input_base_dir + "permission_element_strings.grd")
    dic = {}
    messages = dom.getElementsByTagName("message")
    for message in messages:
        dic[tclib.GenerateMessageId(
            message.firstChild.data.strip())] = message.getAttribute("name")
    # The returned orderings are used to help sort keys for the fixed flat map.
    # Empirically, resource IDs appear to be allocated in the order the messages
    # are listed in the grd.
    return (dic,
            dict((message.getAttribute("name"), i)
                 for (i, message) in enumerate(messages)))


def generate_grd_file(id_map, file_list, output_file_path):
    doc = parseString(kGrdTemplate)
    messages_node = doc.getElementsByTagName("messages")[0]
    for file in file_list:
        translated_file = parse(file)
        translated_messages = translated_file.getElementsByTagName(
            "translation")
        if translated_messages.length == 0:
            continue
        message_name_suffix = file.rsplit('.',
                                          1)[0].rsplit('_',
                                                       1)[1].replace('-', '_')
        for translated_message in translated_messages:
            message_name_prefix = id_map[translated_message.getAttribute("id")]
            generated_message_name = message_name_prefix + "_" + message_name_suffix
            message = translated_message.firstChild.data.strip()
            new_message_node = doc.createElement("message")
            new_message_node.setAttribute("name", generated_message_name)
            new_message_node.setAttribute("translateable", "false")
            messages_node.appendChild(new_message_node)
            new_message_node.appendChild(doc.createTextNode(message))
            messages_node.appendChild(doc.createTextNode('\n      '))

    with open(output_file_path, 'wb') as output_file:
        output_file.write(doc.toxml(encoding='UTF-8'))


# Generate the shortest string containing both `long` and `short` as
# substrings. The full n-string problem is NP-complete, but we only use a crude
# greedy heuristic.
def superstring(long, short):
    if long.find(short) >= 0:
        return long
    for i in range(len(short), 0, -1):
        if long[-i:] == short[:i]:
            return long + short[i:]
        if short[-i:] == long[:i]:
            return short + long[i:]
    return long + short


def generate_cpp_mapping(orderings, input_file_path, output_file_path):
    doc = parse(input_file_path)
    messages = doc.getElementsByTagName("message")
    with open(output_file_path, 'w') as output_file:
        # This is to add language-only versions for the only three languages for
        # which we do not have language-only locales available in our translation
        # lists. The language only version of the string is needed for the case
        # when the combination of language and country is unknown. E.g. for the
        # `pt-AO` (Portuguese Angola) lang setting, we will use `pt`, which via
        # this code will use `pt-pt` (Portuguese from Portugal).
        custom_locale_mappings = {"en-gb": "en", "pt-pt": "pt", "zh-cn": "zh"}

        locales = set()
        message_map = []
        for message in messages:
            message_name = message.getAttribute('name')
            base_message = re.split('_[a-z]', message_name)[0]
            locale = message_name.split(base_message)[1].split(
                '_', 1)[1].lower().replace("_", "-")
            if locale in custom_locale_mappings:
                locale = custom_locale_mappings[locale]
            # Add all locales first since iteration order is non-deterministic.
            locales.add(locale)
            message_map.append((locale, base_message, message_name))

        langs = ''
        locales = sorted(locales, key=lambda x: (-len(x), x))
        # Sort by length so that we add `ab-cd` before `ab` and `cd`
        for locale in locales:
            langs = superstring(langs, locale)

        lang_map = {}
        for locale in locales:
            lang_map[locale] = langs.find(locale), len(locale)

        output_file.write(kStringMapCcPrefix)
        output_file.write(f'    "{langs}";\n')
        output_file.write(kStringMapCcMidfix)
        # Pre-sorting is important here to avoid running into constexpr
        # evaluation limits at compile time, since the translation tables can be
        # quite large.
        for (lang, base_message,
             message_name) in sorted(message_map,
                                     key=lambda x: (x[0], orderings[x[1]])):
            output_file.write(
                f'        {{{{{{{lang_map[lang][0]}, {lang_map[lang][1]}}}, {base_message}}}, {message_name}}},\n'
            )
        # Argh
        output_file.write(kStringMapCcSuffix)


def main(argv):
    output_grd_file_position = argv.index('--output_grd')
    output_map_file_position = argv.index('--output_map')
    input_base_dir_position = argv.index('--input_base_dir')
    input_base_dir = argv[input_base_dir_position + 1]
    id_map, orderings = get_message_id_map_and_orderings(input_base_dir)
    translated_files = list(
        glob.glob(input_base_dir +
                  "translations/permission_element_strings_*"))
    generate_grd_file(id_map, translated_files,
                      argv[output_grd_file_position + 1])
    generate_cpp_mapping(orderings, argv[output_grd_file_position + 1],
                         argv[output_map_file_position + 1])


if __name__ == '__main__':
    sys.exit(main(sys.argv))