emchateau

xquery version "3.0";

(: find the shortest and longest article and get the average word count of a collection of TEI XML articles :)

declare namespace tei="http://www.tei-c.org/ns/1.0";

(: in our case, 'articles' are TEI divs that have @xml:id attributes and no child divs; 
   we filter out the foreward since they're not full articles. :)
let $milestone-articles := collection('/db/cms/apps/tei-content/data/milestones')//tei:div[@xml:id and not(.//tei:div)][@xml:id ne 'foreword']
let $article-infos := 

emchateau

xquery version "3.0";

declare namespace fn="http://www.w3.org/2005/xpath-functions";

(: Fix problems with mis-capitalized names. For example: 
    Before: MACARTHUR, Douglas II
    After:  MacArthur, Douglas II
:)
declare function local:fix-name-capitalization($name as xs:string) {
    (: 

emchateau

xquery version "1.0";

(: A naive approach to sentence tokenization inspired by http://stackoverflow.com/a/2103653/659732 
 : 
 : Works well with edited text like newspapers. Parameters like punctuation can/should be edited; 
 : see the section below called "criteria".
 :
 : For a more sophisticated approach, see Tibor Kiss and Jan Strunk, "Unsupervised Multilingual 
 : Sentence Boundary Detection", Computational Linguistics, Volume 32, Issue 4, December 2006, 
 : pp. 485-525.  Also, see these discussions of sentence tokenization: 

emchateau

javascript:(function({
  var INSTAPAPER=true,w=window,d=document,pageSelectedTxt=w.getSelection?w.getSelection():(d.getSelection)?d.getSelection():(d.selection?d.selection.createRange().text:0),pageTitle=d.title,pageUri=w.location.href,tmplt="";
  tmplt="From ["+pageTitle+"]("+pageUri+"):\n\n";
  if(pageSelectedTxt!="") {
    pageSelectedTxt=">%20"+pageSelectedTxt;
    pageSelectedTxt=pageSelectedTxt.replace(/(\r\n|\n+|\r)/gm,"\n");
    pageSelectedTxt=pageSelectedTxt.replace(/\n/g,"\n>%20\n>%20");
    w.location.href="nvalt://make/?txt="+encodeURIComponent(tmplt+pageSelectedTxt)+"&title="+encodeURIComponent(pageTitle)
  }
  else {

emchateau

<div>
  <h2>{$heading} - {count($actions/generate)}</h2>
  <p>Actions are processes that generate a new item from an existing item.</p>
   <div>
   {$partial("action1.xml","action",$actions/generate )}
   </div>
</div>

emchateau

# Requirements:
# - pen drive formatted in FAT
# - boot.img.gz downloaded from: http://ftp.debian.org/debian/dists/Debian6.0.2/main/installer-amd64/current/images/hd-media/
# - netinstall iso image http://cdimage.debian.org/debian-cd/6.0.2.1/amd64/iso-cd/
# - pen drive is on /dev/disk1 -> check with "diskutil list"
diskutil unmountDisk /dev/disk1
gzip -dc boot.img.gz >/dev/disk1
diskutil eject /dev/disk1
# mount pen drive again
# add iso image to the pen drive

emchateau

xquery version "1.0";
(:~
 : This function tokenize a string and returns each text item in a <text/> element
 :)
let $string := "test,test,test,test,test"
return
  for $text in tokenize($string,',')
  return
    <text>{concat('add some text: ', $text)}</text>

emchateau

<sales-qty-by-product>{
  for $pname in distinct-values(
    doc("sales-records.xml")/*/record/product-name)
  order by $pname
  return
    <product name="{$pname}">{
      sum(
        doc("sales-records.xml")/*/record[
          product-name eq $pname ]/qty)
    }</product>

emchateau

xquery version "3.1";

(: parse tweets using XQuery 3.1's JSON support
 : see http://www.w3.org/TR/xpath-functions-31/#json
 : sample JSON from https://dev.twitter.com/rest/reference/get/statuses/user_timeline 
 :)

let $json := json-doc('/db/user_timeline.json')
let $tweets := $json?*
return

emchateau

xquery version "3.1";

(: Utility functions for JSON as derived via XQuery 3.1's parse-json and json-doc functions.
 : See http://www.w3.org/TR/xpath-functions-31/#json :)

module namespace ju = "http://joewiz.org/ns/xquery/json-util";

declare namespace output="http://www.w3.org/2010/xslt-xquery-serialization";

(: Get the data type for a piece of JSON data :)