Created
October 18, 2017 21:49
-
-
Save joewiz/67b9264d9442c45f00e1f1b108b06615 to your computer and use it in GitHub Desktop.
Simple date processing, with XQuery
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
xquery version "3.1"; | |
module namespace dp = "http://history.state.gov/ns/xquery/date-processing"; | |
declare variable $dp:month-regex := "(January|February|March|April|May|June|July|August|September|October|November|December)"; | |
declare variable $dp:day-regex := "(\d{1,2})(?:st|d|nd|rd|th)?"; | |
declare variable $dp:regexes := | |
map { | |
"month": "(January|February|March|April|May|June|July|August|September|October|November|December)", | |
"day": "(\d{1,2})(?:st|d|nd|rd|th)?" | |
} | |
; | |
declare function dp:get-dates($source) { | |
let $day-regex := "(\d{1,2})(?:st|d|nd|rd|th)?" | |
let $day-range-regex := $day-regex || "[-–—]" || $day-regex | |
let $year-regex := "(\d{4})" | |
let $day-month-year-regex := $day-regex || '\s+' || $dp:month-regex || ',?\s+' || $year-regex | |
let $month-day-year-regex := $dp:month-regex || '\s+' || $day-regex || ',?\s+' || $year-regex | |
let $month-day-range-year-regex := | |
$dp:month-regex || '\s+' || $day-range-regex || ',?\s+' || $year-regex | |
return | |
( | |
analyze-string($source, $day-month-year-regex), | |
analyze-string($source, $month-day-year-regex), | |
analyze-string($source, $month-day-range-year-regex) | |
) | |
}; |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
xquery version "3.1"; | |
import module namespace dp = "http://history.state.gov/ns/xquery/date-processing" at "date-processing.xqm"; | |
let $source := | |
<head>The first date is <hi>February 2d, 1865</hi>. The next date is March 1, 2010. The <strong>third</strong> date is <hi rend="italic">31st July</hi> 2015.<note>This is a note with a date: September 1-4, 1929.</note></head>/string() | |
return | |
dp:get-dates($source) | |
(: | |
<date-hit> | |
<type>single</type> | |
<string>February 2d, 1865</string> | |
<date>1865-02-02</date> | |
</date-hit> | |
<date-hit> | |
<type>range</type> | |
<string>September 1-4, 1929</string> | |
<date-start>1929-09-01</date-end> | |
<date-start>1929-09-04</date-end> | |
</date-hit> | |
:) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment