Initial data commit

This commit is contained in:
Ingy döt Net 2023-07-01 11:58:00 -04:00
parent 72d218235f
commit f23f22d71c
199087 changed files with 3378941 additions and 0 deletions

View file

@ -0,0 +1,20 @@
let $maxentries := 10,
$uri := 'https://www.gutenberg.org/files/135/135-0.txt'
return
<words in="{$uri}" top="{$maxentries}"> {
(
let $doc := unparsed-text($uri),
$tokens := (
tokenize($doc, '\W+')[normalize-space()]
! lower-case(.)
! normalize-unicode(., 'NFC')
)
return
for $token in $tokens
let $key := $token
group by $key
let $count := count($token)
order by $count descending
return <word key="{$key}" count="{$count}"/>
)[position()=(1 to $maxentries)]
}</words>

View file

@ -0,0 +1,12 @@
<words in="https://www.gutenberg.org/files/135/135-0.txt" top="10">
<word key="the" count="41092"/>
<word key="of" count="19954"/>
<word key="and" count="14943"/>
<word key="a" count="14545"/>
<word key="to" count="13953"/>
<word key="in" count="11219"/>
<word key="he" count="9649"/>
<word key="was" count="8622"/>
<word key="that" count="7924"/>
<word key="it" count="6661"/>
</words>