{"id":83055,"date":"2020-05-28T01:42:15","date_gmt":"2020-05-27T23:42:15","guid":{"rendered":"https:\/\/prohoster.info\/blog\/administrirovanie\/kak-linuxovskij-sort-sortiruet-stroki"},"modified":"2020-05-28T01:42:15","modified_gmt":"2020-05-27T23:42:15","slug":"kak-linuxovskij-sort-sortiruet-stroki","status":"publish","type":"post","link":"https:\/\/prohoster.info\/cs\/blog\/administrirovanie\/kak-linuxovskij-sort-sortiruet-stroki","title":{"rendered":"Jak linuxov\u00fd sort \u0159ad\u00ed \u0159et\u011bzce","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\u00davod<\/h1>\n

<\/p>\n

V\u0161e za\u010dalo kr\u00e1tk\u00fdm skriptem, kter\u00fd m\u011bl slou\u010dit informace o adres\u00e1ch e-mail<\/em> zam\u011bstnanc\u016f, z\u00edskan\u00fdch ze seznamu u\u017eivatel\u016f e-mailov\u00e9ho zpravodaje, s pozicemi zam\u011bstnanc\u016f, kter\u00e9 byly z\u00edsk\u00e1ny z datab\u00e1ze person\u00e1ln\u00edho odd\u011blen\u00ed. Ob\u011b seznamy byly exportov\u00e1ny do textov\u00fdch soubor\u016f v k\u00f3dov\u00e1n\u00ed Unicode UTF-8<\/em> a ulo\u017eeny s unixov\u00fdmi konci \u0159\u00e1dk\u016f.<\/p>\n

<\/p>\n

Obsah mail.txt<\/em><\/p>\n

<\/p>\n

Ivanov Andrej;ia@example.com<\/code><\/pre>\n
<\/p>\n
Obsah buhg.txt<\/em><\/p>\n
<\/p>\n
Ivanova Alla;mal\u00ed\u0159\nJolkina Ella;je\u0159\u00e1bnice\nIvanov Andrej;zedn\u00edk\nAbakanov Michail;mal\u00ed\u0159<\/code><\/pre>\n
<\/p>\n
Pro slou\u010den\u00ed byly soubory se\u0159azeny unixov\u00fdm p\u0159\u00edkazem sort<\/em> a p\u0159ed\u00e1ny do unixov\u00e9ho programu p\u0159ipojit<\/em>, kter\u00fd n\u00e1hle skon\u010dil s chybou: <\/p>\n
<\/p>\n
$>> sort buhg.txt > buhg.srt\n$>> sort mail.txt > mail.srt\n$>> join buhg.srt mail.srt > result\njoin: buhg.srt:4: nen\u00ed se\u0159azeno: Ivanov Andrej;zedn\u00edk<\/code><\/pre>\n
<\/p>\n
N\u00e1hled v\u00fdsledku t\u0159\u00edd\u011bn\u00ed uk\u00e1zal, \u017ee celkov\u011b je t\u0159\u00edd\u011bn\u00ed spr\u00e1vn\u00e9, ale v p\u0159\u00edpad\u011b shody mu\u017esk\u00fdch a \u017eensk\u00fdch p\u0159\u00edjmen\u00ed, \u017eensk\u00e1 p\u0159\u00edjmen\u00ed jsou p\u0159ed mu\u017esk\u00fdmi:<\/p>\n
<\/p>\n
$>> sort buhg.txt\nAbakanov Michail;mal\u00ed\u0159\nJolkina Ella;je\u0159\u00e1bnice\nIvanova Alla;mal\u00ed\u0159\nIvanov Andrej;zedn\u00edk<\/code><\/pre>\n
<\/p>\n
Vypad\u00e1 to jako chyba t\u0159\u00edd\u011bn\u00ed v Unicode nebo jako projev feminismu v algoritmu t\u0159\u00edd\u011bn\u00ed. Prvn\u00ed mo\u017enost je samoz\u0159ejm\u011b pravd\u011bpodobn\u011bj\u0161\u00ed.<\/p>\n
<\/a><\/noindex><\/p>\n
Na chv\u00edli to odlo\u017e\u00edme p\u0159ipojit<\/em> a soust\u0159ed\u00edme se na sort<\/em>. Zkus\u00edme vy\u0159e\u0161it probl\u00e9m metodou pokus-omyl. Pro za\u010d\u00e1tek zm\u011bn\u00edme lokalitu z en_US<\/em> na ru_RU<\/em>. Pro t\u0159\u00edd\u011bn\u00ed by sta\u010dilo nastavit prom\u011bnnou prost\u0159ed\u00ed LC_COLLATE<\/em>, ale nebudeme se \u0161et\u0159it:<\/p>\n
<\/p>\n
$>> LANG=ru_RU.UTF-8 sort buhg.txt\nAbakanov Michail;mal\u00ed\u0159\nJolkina Ella;je\u0159\u00e1bnice\nIvanova Alla;mal\u00ed\u0159\nIvanov Andrej;zedn\u00edk<\/code><\/pre>\n
<\/p>\n
Nic se nezm\u011bnilo.<\/p>\n
<\/p>\n
Zkus\u00edme p\u0159ek\u00f3dovat soubory do jednovbajtov\u00e9ho k\u00f3dov\u00e1n\u00ed: <\/p>\n
<\/p>\n
$>> iconv -f UTF-8 -t KOI8-R buhg.txt \n | LANG=ru_RU.KOI8-R sort \n | iconv -f KOI8-R -t UTF8<\/code><\/pre>\n
<\/p>\n
Op\u011bt se nic nezm\u011bnilo.<\/p>\n
<\/p>\n
Nezb\u00fdv\u00e1 ne\u017e hledat \u0159e\u0161en\u00ed na internetu. O rusk\u00fdch p\u0159\u00edjmen\u00edch sice nic nen\u00ed, ale existuj\u00ed ot\u00e1zky ohledn\u011b jin\u00fdch podivnost\u00ed v t\u0159\u00edd\u011bn\u00ed. Nap\u0159\u00edklad takov\u00fd probl\u00e9m: unix sort pova\u017euje znaky \u201e-\u201c (poml\u010dku) za neviditeln\u00e9<\/a><\/noindex>. Stru\u010dn\u011b \u0159e\u010deno, \u0159et\u011bzce \"a-b\", \"aa\", \"ac\" jsou se\u0159azeny jako \"aa\", \"a-b\", \"ac\".<\/p>\n
<\/p>\n
Odpov\u011b\u010f je v\u0161ude standardn\u00ed: pou\u017eijte program\u00e1torskou lokalitu \"C\"<\/em> a bude v\u00e1m zle. Zkou\u0161\u00edme:<\/p>\n
<\/p>\n
$>> LANG=C sort buhg.txt\nJolkina Ella;je\u0159\u00e1bnice\nAbakanov Michail;mal\u00ed\u0159\nIvanov Andrej;zedn\u00edk\nIvanova Alla;advok\u00e1t<\/code><\/pre>\n
<\/p>\n
N\u011bco se zm\u011bnilo. Ivanovovi se \u0159adili ve spr\u00e1vn\u00e9m po\u0159ad\u00ed, ale Jolkina n\u011bkam zmizela. Vrac\u00edme se k p\u016fvodn\u00edmu \u00fakolu:<\/p>\n
<\/p>\n
$> LANG=C sort buhg.txt > buhg.srt\n$> LANG=C sort mail.txt > mail.srt\n$> LANG=C join buhg.srt mail.srt > result<\/code><\/pre>\n
<\/p>\n
Pracovalo to bez chyb, tak jak sliboval internet. A to i p\u0159es Jolkina v prvn\u00edm \u0159\u00e1dku.<\/p>\n
<\/p>\n
Probl\u00e9m se zd\u00e1 b\u00fdt vy\u0159e\u0161en, ale pro jistotu zkus\u00edme je\u0161t\u011b jedno rusk\u00e9 k\u00f3dov\u00e1n\u00ed \u2014 Windows. CP1251<\/em>:<\/p>\n
<\/p>\n
$> iconv -f UTF-8 -t CP1251 buhg.txt \n | LANG=ru_RU.CP1251 sort \n | iconv -f CP1251 -t UTF8 <\/code><\/pre>\n
<\/p>\n
V\u00fdsledek t\u0159\u00edd\u011bn\u00ed, jak se ukazuje, odpov\u00edd\u00e1 lokalit\u011b. \"C\"<\/em>, a cel\u00fd p\u0159\u00edklad proch\u00e1z\u00ed bez chyb. Je to n\u011bjak\u00e1 mystika.<\/p>\n
<\/p>\n
Nem\u00e1m r\u00e1d mystiku v programov\u00e1n\u00ed, proto\u017ee obvykle maskuje chyby. Budu se muset v\u00e1\u017en\u011b zab\u00fdvat ot\u00e1zkou, jak to funguje sort<\/em> a na co to m\u00e1 vliv. LC_COLLATE<\/em> .<\/p>\n
<\/p>\n
Na konci se pokus\u00edm odpov\u011bd\u011bt na ot\u00e1zky:<\/p>\n
<\/p>\n