Files
msime-windows/server/tests/src/test_jianpin_query.cpp
T
fanlusky a7779d2b05 refactor(engine): 将 engine 从 submodule 内嵌为仓内一等组件
`vendor/MetasequoiaImeEngine` submodule 改为 `engine/` 目录,导入的是
product-lock.json 原本钉住的提交 c810d201,逐文件核对过 blob,因此这是一次
搬迁而不是引擎升级。引擎自己的嵌套 submodule(googlepinyinime-rev、utfcpp、
miniaudio)按各自 gitlink 一并展开,来源与提交见 engine/UPSTREAM.md。

Windows 专门化裁掉了不属于本产品的部分:whisper.cpp 与本地推理 provider、
macOS 示例宿主、整个 dictionary/ 建库流水线、上游 CI 与 release-please 配置。
词库仍按 product-lock.json 在构建时取发布产物。

引擎不再进 product-lock.json:没有 gitlink 就没有可比对的东西,锁里写一个上游
commit 等于宣称这棵树仍等于上游,而第一个 Windows 专门化改动落地时这句话就假了。
verify-contracts / verify-checkout 随之移除,verify-published 保留。

CI 与工具链:引擎源码并入根 scripts/format.sh(删掉引擎自己的那份,避免两套排除
列表分叉);引擎 voice.yml 里的 silero_vad.onnx 摘要校验移到本仓 lock job;
engine-bump-triage 工作流与其测试删除;gitsubmodule 的 dependabot 频率由 daily
改为 monthly。

README、README.en、AGENTS.md、CONTRIBUTING、docs/product-release.md 及各子目录
文档同步说明新的位置与边界。
2026-09-12 03:40:48 +08:00

169 lines
5.9 KiB
C++

#include "tests/includes/test_framework.h"
#include "engine/local_modes/jianpin_query.h"
#include "engine/shuangpin/shuangpin_profile.h"
#include "utils/common_utils.h"
#include <filesystem>
#include <string>
#include <utf8.h>
#include <vector>
namespace
{
bool DictionaryAvailable()
{
return std::filesystem::exists(CommonUtils::get_ime_data_path() + "\\msime.db");
}
bool Contains(const std::vector<WordItem> &items, const std::string &word)
{
for (const auto &item : items)
if (item.word == word)
return true;
return false;
}
size_t Utf8Length(const std::string &text)
{
return static_cast<size_t>(utf8::distance(text.begin(), text.end()));
}
} // namespace
TEST_CASE(jianpin_ranking_context_joins_each_letter)
{
REQUIRE_EQ(metasequoia::local_modes::jianpin_ranking_context("nh"), std::string("n'h"));
REQUIRE_EQ(metasequoia::local_modes::jianpin_ranking_context("NH"), std::string("n'h"));
REQUIRE_EQ(metasequoia::local_modes::jianpin_ranking_context("a"), std::string("a"));
REQUIRE(metasequoia::local_modes::jianpin_ranking_context("n1").empty());
REQUIRE(metasequoia::local_modes::jianpin_ranking_context("").empty());
}
TEST_CASE(jianpin_query_rejects_empty_and_non_letters)
{
REQUIRE(metasequoia::local_modes::query_jianpin("", SchemeType::Quanpin).candidates.empty());
REQUIRE(metasequoia::local_modes::query_jianpin("n'h", SchemeType::Quanpin).candidates.empty());
}
TEST_CASE(jianpin_query_matches_two_letter_word)
{
if (!DictionaryAvailable())
return;
const auto results = metasequoia::local_modes::query_jianpin("nh", SchemeType::Quanpin, 50).candidates;
REQUIRE(Contains(results, "你好"));
for (const auto &item : results)
{
REQUIRE_EQ(Utf8Length(item.word), 2);
REQUIRE(item.source == CandidateSource::Database);
REQUIRE(!item.canonical_pinyin.empty());
}
}
TEST_CASE(jianpin_query_does_not_treat_letters_as_full_syllable)
{
if (!DictionaryAvailable())
return;
const auto results = metasequoia::local_modes::query_jianpin("an", SchemeType::Quanpin, 50).candidates;
REQUIRE(!results.empty());
REQUIRE(!Contains(results, "安"));
for (const auto &item : results)
REQUIRE_EQ(Utf8Length(item.word), 2);
}
std::string SyllableInitialForTest(const std::string &syllable)
{
if (syllable.size() >= 2)
{
const auto prefix = syllable.substr(0, 2);
if (prefix == "zh" || prefix == "ch" || prefix == "sh")
return prefix;
}
return syllable.empty() ? std::string{} : syllable.substr(0, 1);
}
std::vector<std::string> SplitKeyForTest(const std::string &key)
{
std::vector<std::string> parts;
size_t start = 0;
while (true)
{
const size_t pos = key.find('\'', start);
if (pos == std::string::npos)
{
parts.push_back(key.substr(start));
return parts;
}
parts.push_back(key.substr(start, pos - start));
start = pos + 1;
}
}
TEST_CASE(jianpin_shuangpin_expands_zh_ch_sh_keys_per_scheme)
{
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("nu", SchemeType::Shuangpin, GetShuangpinProfile("xiaohe")),
std::string("n'sh"));
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("ns", SchemeType::Shuangpin, GetShuangpinProfile("xiaohe")),
std::string("n's"));
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("vi", SchemeType::Shuangpin, GetShuangpinProfile("xiaohe")),
std::string("zh'ch"));
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("ne", SchemeType::Shuangpin, GetShuangpinProfile("shoudao")),
std::string("n'sh"));
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("nu", SchemeType::Shuangpin, GetShuangpinProfile("shoudao")),
std::string("n'u"));
REQUIRE_EQ(
metasequoia::local_modes::jianpin_ranking_context("nu", SchemeType::Shuangpin, GetShuangpinProfile("ziranma")),
std::string("n'sh"));
REQUIRE_EQ(metasequoia::local_modes::jianpin_ranking_context("nu", SchemeType::Shuangpin,
GetShuangpinProfile("microsoft")),
std::string("n'sh"));
REQUIRE_EQ(metasequoia::local_modes::jianpin_ranking_context("nu"), std::string("n'u"));
}
TEST_CASE(jianpin_shuangpin_nu_matches_n_sh_not_n_s)
{
if (!DictionaryAvailable())
return;
const auto nu =
metasequoia::local_modes::query_jianpin("nu", SchemeType::Shuangpin, 50, GetShuangpinProfile("xiaohe"))
.candidates;
const auto ns =
metasequoia::local_modes::query_jianpin("ns", SchemeType::Shuangpin, 50, GetShuangpinProfile("xiaohe"))
.candidates;
REQUIRE(!nu.empty());
REQUIRE(!ns.empty());
REQUIRE(Contains(nu, "你说"));
REQUIRE(!Contains(ns, "你说"));
for (const auto &item : nu)
{
const auto syllables = SplitKeyForTest(item.canonical_pinyin);
REQUIRE_EQ(syllables.size(), 2);
REQUIRE_EQ(SyllableInitialForTest(syllables[0]), std::string("n"));
REQUIRE_EQ(SyllableInitialForTest(syllables[1]), std::string("sh"));
}
for (const auto &item : ns)
{
const auto syllables = SplitKeyForTest(item.canonical_pinyin);
REQUIRE_EQ(syllables.size(), 2);
REQUIRE_EQ(SyllableInitialForTest(syllables[0]), std::string("n"));
REQUIRE_EQ(SyllableInitialForTest(syllables[1]), std::string("s"));
}
}
TEST_CASE(jianpin_shoudao_uses_e_for_sh)
{
if (!DictionaryAvailable())
return;
const auto ne =
metasequoia::local_modes::query_jianpin("ne", SchemeType::Shuangpin, 50, GetShuangpinProfile("shoudao"))
.candidates;
REQUIRE(Contains(ne, "你说"));
const auto nu =
metasequoia::local_modes::query_jianpin("nu", SchemeType::Shuangpin, 50, GetShuangpinProfile("shoudao"))
.candidates;
REQUIRE(!Contains(nu, "你说"));
}