Fix erroneous splitting of Hungarian time names with additions
This commit is contained in:
parent
3e9f675fdc
commit
4a49c7a4e7
|
@ -42,7 +42,7 @@ final class NodaTimeSplitter {
|
|||
];
|
||||
|
||||
const MONTH_NAMES_HUNGARIAN = [
|
||||
"01" => ['január', 'jan'],
|
||||
"01" => ['január', 'januar', 'jan'],
|
||||
"02" => ['február', 'feb'],
|
||||
"03" => ['március', 'mar.', 'már.'],
|
||||
"04" => ['április', 'apr.', 'ápr.'],
|
||||
|
@ -521,20 +521,25 @@ final class NodaTimeSplitter {
|
|||
// Skip, if dates are too long and do not contain spaces (= no translatable names)
|
||||
if (str_contains($datum, " ") === false && strlen($datum) > 12) return [];
|
||||
|
||||
$unparsed = trim(strtolower(str_replace($year, '', $datum)), ' ,.');
|
||||
foreach (self::MONTH_NAMES_HUNGARIAN as $monthVal => $monthValidNames) {
|
||||
if (self::stri_occurs($datum, $monthValidNames)) {
|
||||
$monat = (string)$monthVal;
|
||||
foreach ($monthValidNames as $name) {
|
||||
$unparsed = str_replace($name, '', $unparsed);
|
||||
}
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if (strlen($unparsed) > 5) {
|
||||
return [];
|
||||
}
|
||||
|
||||
if (empty($monat) and self::is_numeric((string)\substr($datum, 5, 2))) $monat = \substr($datum, 5, 2);
|
||||
else if (empty($monat) and self::is_numeric((string)\substr($datum, 6, 2))) $monat = \substr($datum, 6, 2);
|
||||
|
||||
// Last four characters must contain at least one space or one dot
|
||||
$lastChars = substr($datum, -4);
|
||||
if (str_contains($lastChars, '.') === false && str_contains($lastChars, ' ') === false) return [];
|
||||
|
||||
$day = self::validateDateSubstr($datum, -2);
|
||||
if (empty($day)) $day = self::validateDateSubstr($datum, -3, 2);
|
||||
if (empty($day)) $day = self::validateDateSubstr($datum, -4, 2);
|
||||
|
@ -550,6 +555,14 @@ final class NodaTimeSplitter {
|
|||
}
|
||||
}
|
||||
|
||||
if ($datum === '1978. július 7 elött') {
|
||||
throw new Exception(var_export($monat, true));
|
||||
}
|
||||
|
||||
if (!empty($monat) && empty($day) && preg_match('~[0-9]+~', substr($datum, -3))) {
|
||||
return [];
|
||||
}
|
||||
|
||||
if (!empty($monat) and !empty($day)) {
|
||||
return [$year, $year, $monat, $day, '+', ""];
|
||||
}
|
||||
|
@ -932,6 +945,12 @@ final class NodaTimeSplitter {
|
|||
return $output;
|
||||
}
|
||||
}
|
||||
if (!empty(\preg_match("/^[0-9]{4}ig$/", $datum))) {
|
||||
if ($output = self::attempt_splitting(\substr($datum, 0, 4))) {
|
||||
$output[0] = "?";
|
||||
return $output;
|
||||
}
|
||||
}
|
||||
|
||||
return [];
|
||||
|
||||
|
@ -1102,7 +1121,7 @@ final class NodaTimeSplitter {
|
|||
|
||||
// Hungarian year and month until month
|
||||
// 2005.01.-02. => 2005.01.-2005.02.
|
||||
if ($inputLength === 12 && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\.\-[0-1][0-9]\.$/", $datum)) !== false) {
|
||||
if ($inputLength === 12 && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\.\-[0-1][0-9]\.$/", $datum))) {
|
||||
$reconstituted = substr($datum, 0, 8) . '-';
|
||||
$reconstituted .= substr($datum, 0, 4) . '.' . substr($datum, -3);
|
||||
return $reconstituted;
|
||||
|
@ -1111,7 +1130,7 @@ final class NodaTimeSplitter {
|
|||
// Hungarian year and month until month without a dot after the first YYYY-MM
|
||||
// 2005.01-02. => 2005.01.-2005.02.
|
||||
|
||||
if (in_array($inputLength, [10, 11], true) && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\-[0-1][0-9](\.|)$/", $datum)) !== false) {
|
||||
if (in_array($inputLength, [10, 11], true) && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\-[0-1][0-9](\.|)$/", $datum))) {
|
||||
$reconstituted = substr($datum, 0, 7) . '.-';
|
||||
$reconstituted .= substr($datum, 0, 4) . '.' . substr(rtrim($datum, '.'), -2) . '.';
|
||||
return $reconstituted;
|
||||
|
@ -1120,7 +1139,7 @@ final class NodaTimeSplitter {
|
|||
// Hungarian year and month until month
|
||||
// 2005.01.01.-02.02. => 2005.01.01-2005.02.02.
|
||||
// 2005.01.01-02.02 => 2005.01.01-2005.02.02.
|
||||
if ($inputLength >= 16 && $inputLength <= 18 && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\.[0-3][0-9](\.|)\-[0-1][0-9]\.[0-3][0-9](\.|)$/", $datum)) !== false) {
|
||||
if ($inputLength >= 16 && $inputLength <= 18 && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\.[0-3][0-9](\.|)\-[0-1][0-9]\.[0-3][0-9](\.|)$/", $datum))) {
|
||||
$parts = explode('-', $datum);
|
||||
if (count($parts) !== 2) return '';
|
||||
$reconstituted = substr($datum, 0, 10) . '.-';
|
||||
|
@ -1129,7 +1148,7 @@ final class NodaTimeSplitter {
|
|||
}
|
||||
|
||||
// Hungarian; without trailing dots: YYYY.MM.DD-DD
|
||||
if ($inputLength >= 13 && $inputLength <= 15 && (\preg_match("/^[0-9]{4}\.[0-1][0-9]\.[0-3][0-9](\.|)\-[0-3][0-9](\.|)$/", $datum)) !== false) {
|
||||
if ($inputLength >= 13 && $inputLength <= 15 && \preg_match("/^[0-9]{4}\.[0-1][0-9]\.[0-3][0-9](\.|)\-[0-3][0-9](\.|)$/", $datum)) {
|
||||
$parts = explode('-', $datum);
|
||||
if (count($parts) !== 2) return '';
|
||||
$reconstituted = substr($datum, 0, 10) . '.-';
|
||||
|
@ -1154,8 +1173,8 @@ final class NodaTimeSplitter {
|
|||
|
||||
// If es évek / as évek is contained in the string (e.g. 1880-1990-es évek), there
|
||||
// will be more than one hyphens
|
||||
if (MD_STD::stri_contains_any($datum, ['-as évek', '-es évek'])) {
|
||||
return strtr($datum, ['-as évek' => ' as évek', '-es évek' => ' es évek']);
|
||||
if (MD_STD::stri_contains_any($datum, ['-as évek', '-es-évek', '-es évek'])) {
|
||||
return strtr($datum, ['-as évek' => ' as évek', '-es-évek' => ' es évek', '-es évek' => ' es évek']);
|
||||
}
|
||||
|
||||
// 1981. július-augusztus > 1981.07-08
|
||||
|
|
|
@ -610,14 +610,14 @@ final class NodaTimeSplitterTest extends TestCase {
|
|||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 1925);
|
||||
|
||||
$output = NodaTimeSplitter::attempt_splitting("2020. Januar");
|
||||
self::assertEquals($output, [
|
||||
self::assertEquals([
|
||||
0 => "2020",
|
||||
1 => "2020",
|
||||
2 => "01",
|
||||
3 => "00",
|
||||
4 => "+",
|
||||
5 => "",
|
||||
]);
|
||||
], $output);
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToTimeName($output), "Januar 2020");
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 2020);
|
||||
|
||||
|
@ -729,6 +729,45 @@ final class NodaTimeSplitterTest extends TestCase {
|
|||
self::assertEquals(NodaTimeSplitter::timePartsToTimeName($output), "Bis 1801 v. Chr.");
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 1801);
|
||||
|
||||
/*
|
||||
* TODO
|
||||
$output = NodaTimeSplitter::attempt_splitting("1900 előtt");
|
||||
self::assertEquals($output, [
|
||||
0 => "?",
|
||||
1 => "1899",
|
||||
2 => "00",
|
||||
3 => "00",
|
||||
4 => "+",
|
||||
5 => "",
|
||||
]);
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToTimeName($output), "Vor 1900");
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 1899);
|
||||
*/
|
||||
|
||||
$output = NodaTimeSplitter::attempt_splitting("1900-ig");
|
||||
self::assertEquals($output, [
|
||||
0 => "?",
|
||||
1 => "1900",
|
||||
2 => "00",
|
||||
3 => "00",
|
||||
4 => "+",
|
||||
5 => "",
|
||||
]);
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToTimeName($output), "Bis 1900");
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 1900);
|
||||
|
||||
$output = NodaTimeSplitter::attempt_splitting("1900ig");
|
||||
self::assertEquals($output, [
|
||||
0 => "?",
|
||||
1 => "1900",
|
||||
2 => "00",
|
||||
3 => "00",
|
||||
4 => "+",
|
||||
5 => "",
|
||||
]);
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToTimeName($output), "Bis 1900");
|
||||
self::assertEquals(NodaTimeSplitter::timePartsToCountingYear($output), 1900);
|
||||
|
||||
}
|
||||
|
||||
/**
|
||||
|
@ -881,6 +920,9 @@ final class NodaTimeSplitterTest extends TestCase {
|
|||
$output = NodaTimeSplitter::attempt_splitting("1978. július7");
|
||||
self::assertEmpty($output);
|
||||
|
||||
$output = NodaTimeSplitter::attempt_splitting("1978. július 7 elött");
|
||||
self::assertEmpty($output);
|
||||
|
||||
$output = NodaTimeSplitter::attempt_splitting("Anfang September 1903");
|
||||
self::assertEmpty($output);
|
||||
|
||||
|
|
Loading…
Reference in New Issue
Block a user