Skip to content

fix: enhance Chinese numeral time parsing #360

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
deepin-bot merged 1 commit into from
Jul 20, 2026
Merged

fix: enhance Chinese numeral time parsing #360

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
59 changes: 59 additions & 0 deletions autotests/dfm-search-tests/tst_chinese_nlp.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -959,6 +959,20 @@ void tst_ChineseNLP::timeCustom_month()
m_parser->parse(QStringLiteral("5月份的图片"), intent2);
QCOMPARE(intent2.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent2.timeConstraint().customStart().date().month(), 5);

// "七月" — Chinese numeral month should work too
ParsedIntent intent3;
m_parser->parse(QStringLiteral("七月的文件"), intent3);
QCOMPARE(intent3.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent3.timeConstraint().customStart().date().month(), 7);
QCOMPARE(intent3.timeConstraint().customEnd().date().month(), 7);

// "十二月" — leading-ten Chinese numeral month should work too
ParsedIntent intent4;
m_parser->parse(QStringLiteral("十二月的文件"), intent4);
QCOMPARE(intent4.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent4.timeConstraint().customStart().date().month(), 12);
QCOMPARE(intent4.timeConstraint().customEnd().date().month(), 12);
}

void tst_ChineseNLP::timeCustom_yearMonth()
Expand Down Expand Up @@ -988,6 +1002,22 @@ void tst_ChineseNLP::timeCustom_yearMonth()
QCOMPARE(intent3.timeConstraint().customStart().date().year(), 2025);
QCOMPARE(intent3.timeConstraint().customStart().date().month(), 12);
QCOMPARE(intent3.timeConstraint().customStart().date().day(), 1);

// "2025年七月" — Chinese numeral month should also be parsed
ParsedIntent intent4;
m_parser->parse(QStringLiteral("2025年七月的文档"), intent4);
QCOMPARE(intent4.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent4.timeConstraint().customStart().date().year(), 2025);
QCOMPARE(intent4.timeConstraint().customStart().date().month(), 7);
QCOMPARE(intent4.timeConstraint().customStart().date().day(), 1);

// "2025年十一月" — leading-ten Chinese numeral month in year-month form
ParsedIntent intent5;
m_parser->parse(QStringLiteral("2025年十一月的文档"), intent5);
QCOMPARE(intent5.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent5.timeConstraint().customStart().date().year(), 2025);
QCOMPARE(intent5.timeConstraint().customStart().date().month(), 11);
QCOMPARE(intent5.timeConstraint().customStart().date().day(), 1);
}

void tst_ChineseNLP::timeCustom_yearMonth_separators()
Expand Down Expand Up @@ -1035,6 +1065,20 @@ void tst_ChineseNLP::timeCustom_dateSpoken()
QCOMPARE(intent.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent.timeConstraint().customStart().date().month(), 3);
QCOMPARE(intent.timeConstraint().customStart().date().day(), 8);

// "七月八号" — Chinese numeral month/day should work too
ParsedIntent intent2;
m_parser->parse(QStringLiteral("七月八号的图片"), intent2);
QCOMPARE(intent2.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent2.timeConstraint().customStart().date().month(), 7);
QCOMPARE(intent2.timeConstraint().customStart().date().day(), 8);

// "十二月十五号" — leading-ten Chinese numeral month/day should work too
ParsedIntent intent3;
m_parser->parse(QStringLiteral("十二月十五号的图片"), intent3);
QCOMPARE(intent3.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent3.timeConstraint().customStart().date().month(), 12);
QCOMPARE(intent3.timeConstraint().customStart().date().day(), 15);
}

void tst_ChineseNLP::timeCustom_fullDate()
Expand All @@ -1060,6 +1104,14 @@ void tst_ChineseNLP::timeCustom_fullDate()
QCOMPARE(intent2.timeConstraint().customStart().date().year(), 2025);
QCOMPARE(intent2.timeConstraint().customStart().date().month(), 12);
QCOMPARE(intent2.timeConstraint().customStart().date().day(), 30);

// "2025年七月三十日" — Chinese numeral month/day should work too
ParsedIntent intent3;
m_parser->parse(QStringLiteral("2025年七月三十日的文档"), intent3);
QCOMPARE(intent3.timeConstraint().kind(), TimeConstraintKind::Custom);
QCOMPARE(intent3.timeConstraint().customStart().date().year(), 2025);
QCOMPARE(intent3.timeConstraint().customStart().date().month(), 7);
QCOMPARE(intent3.timeConstraint().customStart().date().day(), 30);
}

void tst_ChineseNLP::timeCustom_fullDate_separators()
Expand Down Expand Up @@ -1912,6 +1964,13 @@ void tst_ChineseNLP::timeDynamic_chineseNumerals()
QCOMPARE(intent5.timeConstraint().kind(), TimeConstraintKind::Relative);
QCOMPARE(intent5.timeConstraint().relativeValue(), 7);

// "最近十五天" — leading-ten Chinese numeral in dynamic relative time
ParsedIntent intent6;
m_parser->parse(QStringLiteral("最近十五天"), intent6);
QCOMPARE(intent6.timeConstraint().kind(), TimeConstraintKind::Relative);
QCOMPARE(intent6.timeConstraint().relativeValue(), 15);
QCOMPARE(intent6.timeConstraint().relativeUnit(), TimeUnit::Days);

// Mixed: Arabic + Chinese should still work
// "最近3天" already tested above
}
Expand Down
12 changes: 10 additions & 2 deletions src/dfm-search/dfm-search-lib/semantic/extractors/timeextractor.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -270,10 +270,18 @@ int TimeExtractor::localeAwareToInt(const QString &input,
return -1;
}

// Two-character pattern: "XY" where Y is the tens unit (e.g., "十五" = 15)
// Leading tens-unit pattern: "十X" (e.g., "十一" = 11, "十五" = 15)
if (input.size() == 2 && input.left(1) == tensUnit) {
int suffix = digitMap.value(input.right(1), -1);
if (suffix >= 0) {
return 10 + suffix;
}
}

// Two-character pattern: "XY" where Y is the tens unit (e.g., "二十" = 20)
if (input.size() == 2 && input.mid(1) == tensUnit) {
int prefix = digitMap.value(input.left(1), -1);
if (prefix > 1) {
if (prefix > 0) {
return prefix * 10;
}
// "十" alone = 10
Expand Down
8 changes: 4 additions & 4 deletions src/dfm-search/dfm-search-lib/semantic/rules/zh_CN/time_rules.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,7 +132,7 @@
},
{
"id": "time_exact_month_current_year",
"pattern": "(?<month>\\d{1,2})月份?",
"pattern": "(?<month>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))月份?",
"description": "Month this year (e.g. 12月, 5月份)",
"enabled": true,
"priority": 120,
Expand All @@ -157,7 +157,7 @@
},
{
"id": "time_exact_year_month",
"pattern": "(?<year>\\d{2,4})[年\\./\\-](?<month>\\d{1,2})(?:月份?)?",
"pattern": "(?<year>\\d{2,4})[年\\./\\-](?<month>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))(?:月份?)?",
"description": "Exact year-month (e.g. 2025年12月, 2025-12)",
"enabled": true,
"priority": 140,
Expand All @@ -182,7 +182,7 @@
},
{
"id": "time_exact_date_current_year",
"pattern": "(?<month>\\d{1,2})月(?<day>\\d{1,2})[日号]",
"pattern": "(?<month>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))月(?<day>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))[日号]",
"description": "Exact date this year (e.g. 12月5日, 3月8号)",
"enabled": true,
"priority": 140,
Expand All @@ -207,7 +207,7 @@
},
{
"id": "time_exact_full_date",
"pattern": "(?<year>\\d{2,4})[年\\./\\-](?<month>\\d{1,2})[月\\./\\-](?<day>\\d{1,2})[日号]?",
"pattern": "(?<year>\\d{2,4})[年\\./\\-](?<month>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))[月\\./\\-](?<day>(?:\\d{1,2}|[零一二两三四五六七八九十]{1,3}))[日号]?",
"description": "Exact full date (e.g. 2025年12月5日, 2025-12-05)",
"enabled": true,
"priority": 160,
Expand Down
Loading