XPath $\subset$ XQuery

language for querying XML (and sorting and structuring query results)

parsing getting semantics of text

analyzing static errors: searching syntax (independent of document), ...

evaluation dynamic errors: missing input document, division by zero, ...

pronounced: “flower expressions”

generalized version of select from having where in SQL

• Example of FLWOR

  for $d in doc("departments.xml")//dept_no
  let $e := doc("employees.xml")//employee[dept_no = $d]
  where count($e) >= 10
  order by avg($e/salary) descending
  return
  <large_dept>
  {$d}
  <size> {count($e)} </size>
  <avg_salary> {avg($e/salary)} </avg_salary>
  </large_dept>

  • for generates bindings of dept_no values to $d

  • let generates another binding to $e : employees with dept_no as $d

    →forandletcan be used multiple times

  • where filters that list to keep only the desired pairs (at least 10 employees)

  • order by sorts that lists by the given criteria (sorted desceding by average salary)

  • return constructs for each pair a resulting value

One can create new elements and add them in the result, by wrapping query into elements.

• Example of Element Constructor

  Here we create an element and attribute.

  <sorted_departments> {
  for $d in doc("departments.xml")//dept_no
  let $e := doc("employees.xml")//employee[dept_no = $d]
  where count($e) >= 10
  order by avg($e/salary) descending
  return
  <large_dept name = "{$d}">
  {$d}
  <size> {count($e)} </size>
  <avg_salary> {avg($e/salary)} </avg_salary>
  </large_dept>
  } </sorted_departments>

XQuery expressions manupulate lists of items.

if-then-else expressions

( else is required but can be left empty with () )

exists, forall expressed with