Introduction to Semistructured Data and XML. Management of XML and Semistructured Data. Path Expressions

Transcription

1 Introduction to Semistructured Data and XML Chapter 27, Part E Based on slides by Dan Suciu University of Washington Database Management Systems, R. Ramakrishnan 1 Management of XML and Semistructured Data Based upon slides by Dan Suciu Database Management Systems, R. Ramakrishnan 2 Path Expressions Examples: Bib.paper Bib.book.publisher Bib.paper.author.lastname Given an OEM instance, the value of a path expression p is a set of objects Database Management Systems, R. Ramakrishnan 3 1

2 2 Path Expressions Examples: DB firstname Bib &o1 paper paper book references &o12 &o24 &o29 references references author author year author title http title author publisher author title author &o43 &o44 &o45 &o46 &o52 &96 lastname 1997 page &o47 &o48 &o49 &o50 &o51 firstname lastname first &25 last &o70 &o71 &243 &206 Serge Abiteboul Victor Vianu Bib.paper{&o12,&o29} Bib.book.publisher{&o51} Bib.paper.author.lastname{&o71,&206} Database Management Systems, R. Ramakrishnan 4 Summary: --WHERE-ORDERBY- FLWOR / Clauses List of tuples WHERE Clause ORDERBY/ Clause List of tuples Instance of Xquery data model Database Management Systems, R. Ramakrishnan 5 $x in expr -- binds $x to each value in the list expr $x expr -- binds $x to the entire list expr Useful for common subexpressions and for aggregations Database Management Systems, R. Ramakrishnan 6

3 v.s. $x $x document("bib.xml")/bib/book $x $x Returns: <book>...</book> <book>...</book> <book>...</book>... $x $x document("bib.xml")/bib/book $x $x Returns: <book>...</book> <book>...</book> <book>...</book>... Database Management Systems, R. Ramakrishnan 7 Path Expressions Abbreviated Syntax /bib/paper[2]/author[1] /bib//author paper[author/lastname Vianu"] /bib/(paper book)/title Unabbreviated Syntax child::bib/descendant::author child::bib/descendant-or-self::*/child::author parent, self, descendant-or-self, attribute Database Management Systems, R. Ramakrishnan 8 Find all book titles published after 1995: $x $x document("bib.xml")/bib/book WHERE $x/year > 1995 $x/title Result: <title> abc </title> <title> def </title> <title> ghi </title> Database Management Systems, R. Ramakrishnan 9 3

4 4 For each author of a book by Morgan Kaufmann, list all books she published: distinct(document("bib.xml") /bib/book[publisher Morgan Kaufmann ]/author),, $t $t /bib/book[author]/title $t $t distinct a function that eliminates duplicates Database Management Systems, R. Ramakrishnan 10 Result: <author>jones</author> <title> abc </title> <title> def </title> <author> Smith </author> <title> ghi </title> Database Management Systems, R. Ramakrishnan 11 <big_publishers> $p $p distinct(document("bib.xml")//publisher) : : document("bib.xml")/book[publisher $p] $p] WHERE > $p $p </big_publishers> count a (aggregate) function that returns the number of elms Database Management Systems, R. Ramakrishnan 12

5 Find books whose price is larger than average: avg(document("bib.xml")/bib/book/price) in in document("bib.xml")/bib/book WHERE /price > Database Management Systems, R. Ramakrishnan 13 v.s. Binds node variables iteration Binds collection variables one value Database Management Systems, R. Ramakrishnan 14 Collections in Ordered and unordered collections /bib/book/author an ordered collection Distinct(/bib/book/author) an unordered collection /bib/book is a collection /author a collection (several authors...) /author Returns: <author>...</author> <author>...</author> <author>...</author>... Database Management Systems, R. Ramakrishnan 15 5

6 6 Collections in What about collections in expressions? /price list of n prices /price * 0.7 list of n numbers?? /price * /quantity list of n x m numbers?? Valid only if the two sequences have at most one element Atomization ook1/author eq "Kennedy" - Value Comparison ook1/author "Kennedy" - General Comparison Database Management Systems, R. Ramakrishnan 16 Sorting in <publisher_list> $p $p distinct(document("bib.xml")//publisher) ORDERBY $p $p <publisher> <name> $p/text() </name>,, document("bib.xml")//book[publisher $p] $p] ORDERBY /price DESCENDG <book> /title,, /price </book> </publisher> </publisher_list> Database Management Systems, R. Ramakrishnan 17 If-Then-Else $h $h //holding ORDERBY $h/title <holding> $h/title, IF IF $h/@type "Journal" THEN $h/editor ELSE $h/author </holding> Database Management Systems, R. Ramakrishnan 18

7 Existential Quantifiers //book WHERE SOME $p $p //para SATISFIES contains($p, "sailing") AND contains($p, "windsurfing") /title Database Management Systems, R. Ramakrishnan 19 Universal Quantifiers //book WHERE EVERY $p $p //para SATISFIES contains($p, "sailing") /title Database Management Systems, R. Ramakrishnan 20 Other Stuff in If-then-else Universal and existential quantifiers Sorting Before and After for dealing with order in the input Filter deletes some edges in the result tree Recursive functions Database Management Systems, R. Ramakrishnan 21 7

8 8 No GROUPBY currently in A recent proposal (next) What do YOU think? Database Management Systems, R. Ramakrishnan 22 document(" /@year WHERE WHERE /publisher"morgan Kaufmann" GROUPBY WHERE > </year> </year> with GROUPBY SELECT SELECTyear year FROM FROMBib Bib Equivalent SQL WHERE WHEREBib.publisher"Morgan Kaufmann" Kaufmann" GROUPBY GROUPBYyear year HAVG HAVGcount(*) > Database Management Systems, R. Ramakrishnan 23 document(" /author, /author, /@year /@year GROUPBY GROUPBY,,,, Without GROUPBY with GROUPBY document( )/ document( )/ bib/book/author, bib/book/author, /../@year /../@year document(" document(" Correct if the GROUPBY is node-identity based Not equivalent if the GROUPBY is value-based Database Management Systems, R. Ramakrishnan 24

9 9 document(" /author, /author, GROUPBY GROUPBY,,,, with GROUPBY distinct(document( )/ distinct(document( )/ bib/book/author) bib/book/author) distinct(document( distinct(document( IF IF > > 0 0 THEN THEN,, Database Management Systems, R. Ramakrishnan 25 Without GROUPBY document(" /author, /author, /@year /@year GROUPBY GROUPBY,,,, with GROUPBY $Tup $Tup distinct( distinct( document(" document(" /author, /author, /@year /@year <Tup> <Tup> <a> <a> </a> </a> <y> <y> </y> </y> </Tup>), </Tup>), $Tup/a/node(), $Tup/a/node(), $Tup/y/node() $Tup/y/node() document(" document(" Database Management Systems, R. Ramakrishnan 26 Without GROUPBY document(" /author, /@year, $t /title, $p /publisher GROUPBY $p, $p, GROUPBY <authorentry>, GROUPBY $t $t <authorentry> Nested GROUPBY s Database Management Systems, R. Ramakrishnan 27