XPath (transfer)

Now, some basic methods that use XPath to match nodes in XML.

The path matching path match is similar to the file path representation, which is better understood. There are several symbols:

Symbolic meaning case matching result / indicator Node path / A / C / D node "a" child node "d" "D", that is, the ID node / root node // all paths of the ID value of D2 Elements // e element ending after "//", the result is the E element for all three E element // c / e all parent nodes C, the result is that the ID value is E1 and E2. Two E-element * paths / A / B / C / * A element → B element → C element under the group B element, the NAME value is b, the D element and ID value of the ID value of D1 is The two E-elements / * / * / * of E1 and E2 have a D element of two-stage nodes. The matching result is a D element of the ID value of D2 // * All elements | Logic or // b | // C B-element and C element position match for each element, each of its sub-elements is orderly. Such as:

Examples Conflict Match Results / A / B / C [1] A element → b / b / c [Last ()] A / B / C [Last ()] a element → The final sub-element ID value of B element → C element is E element E element / A / B / C [position ()> 1] A element → B element → C element is greater than 1 element ID value of 1 D1 D element and two E-element properties and attribute values ​​with ID values ​​can utilize attributes and attribute values ​​to match elements in XPath. It is to be noted that there must be "@" prefix before the element's attribute name. E.g:

Example Configuration Match Result // B [@ID] All B-element ID values ​​with attribute IDs are B1 and B2 // b [@ *] all of the 2 Id properties with id properties B element and a Name property B element // b [NOT (@ *)] all non-attribute B element a elements → B element under C element // b [@ ID = "b1"] ID value is B1 B element B element under a element

The relative relationship matching the XML document can be attributed to the tree structure, so any of the nodes is not isolated. Usually, we will break the property relationship between nodes into a relative relationship, such as father, child, ancestors, offspring, brothers, etc. These concepts can also be used when matching elements. E.g:

Example Configuration Match Results // E / PARENT:: * All E-Node Parent Node Element ID Values ​​are A1 A1 Elements and ID Values ​​C1 C1 // F / AnceStor :: * All F Elements Ancestral Node The group ID value of the element ID value of the element is A1 is a C element of C2 of C2, the b1, b2 B element, the ID value of C2, and there is no attribute. Element / A / Descendant :: * A of all the descendants of the E-element except for all other elements other than the A element // f / self :: * All F element F element itself // f / Ancestor-or-self: * All F elements and its ancestral node elements F elements, F-element parent C elements and A elements / A / C / Descendant-OR-Self :: * All A elements → C elements and their descendants Id value C2 C element, sub-elements B, D, F element / A / c / following-sibling :: * a element → C-element adjacent back sequence All brothers Node elements do not have any attributes / A / C / preceding-sibling :: * A element → C element adjacent front All brothers Node element ID value of B1 and B2 2 B1 and B2 / A / B / C / FOLLOWING:: * a element → B element → c element All element IDs of the sequence are B2 B2, the C element without the attribute, the B element without the attribute, the ID of D2 D2 D element, the F element without attribute, and an E element without the attribute. / A / c / preceding :: * A element → The front of all element IDs of the C element is B2 B2 B2, the ID of E2 E element, the ID is E element E element, the ID of D1 D1, NAME is B B element, ID of C1 C1, b element of ID B1

The conditional matching condition match is the Boolean value of the calculation result of some functions to match the eligible node. Some functions that are commonly used for conditions match have four categories: node functions, string functions, numeric functions, and Boolean functions. For example, Last (), position (), etc. mentioned earlier. These functional functions can help us find the needs of the nodes.

Function and functional action count () function statistics count, return the number of Number () features in the eligible node to convert the text in the value of the value into a value substring () function

Syntax: Substring (Value, Start, Length)

Intercept string SUM () function summary

These features are only part of the XPath syntax, and there are a lot of functional functions that are not introduced, and the syntax of XPath is still growing. Through these functions we can achieve more complex queries and operations.

In these matching methods, the most important path matching is used. Relying on the position of the node to give a subpath relative to the current path. The XQuery ---- The revolutionary data inquiry language (Zeng Yi) in the XML era (Zeng Yi) has quickly received support and concerns from all parties and research institutions since the birth of the documentation and data. Now, many vendors developed using XML to transmit messages (such as SOAP or XML-RPC messages) or as a permanent storage of data (such as XML database). Data query language in the relational database SQL, XQuery will become the mainstream data query language of the XML era. XQuery overview After the popularity of computer networks, the Internet has become one of the important means of obtaining information. With the current chance of technology, people inquiry data on the Internet are also more powerful, and given the query of unstructured data and the relational database Self-defects, people need a more optimized data inquiry language, and XQuery played this role with its good design and powerful features. XQuery is a query language that gets data from an XML format, originated from the XML data query language quilt. And put XPath version 2.0 as its subset. Quilt has a lot of very good features, integrates SQL, ODMG, XPath 1.0, XQL, and XML-QL many features, however, as the amount of information stored in the XML document, for efficient access and Query XML information, Quilt shows its shortcomings. So the new XQuery data query language was born. The XQuery specification launched the Query Language of the W3C in 1998 Boston Semiological Seminar, and the members of the meeting should be divided into two categories. The first category is the person used by the XML as a document, and the second class is the XML as a The person used by data. The invited representative from the industry, academia and research groups uses this opportunity to publish their own views, explain the characteristics and needs of XML query language they think. Query Language Working Group was established after the discussion. This working group is very large, consisting of more than 30 members, XML query language standards represent the needs and views of two types of users. Although XQuery standardized work has been launched in 1998, the time spent in this work is quite long due to differences in XML applications. Until in February 2001, the issuance of the Working Group began to carry out steps, and a large number of documents began to launch. In June and December 2001, an important update was conducted in August 2002 and in May 2003. After joining an XQuery serialization and two full-text related drafts in May 2003, the 12 documents were basically completed, and the draft work is approaching the end. These 12 important documents include: XML Query Requirements (Latest version posted on November 12, 2003) Planning documents for this document Construction team. XQuery demand list. XML Query Use Cases (Latest Version Published on November 12, 2003) This document provides several actual solutions to solve specific problems and the XQuery code snippet. XQuery 1.0: An XML Query Language (Latest Version Published on November 12, 2003) This document is the core document in the draft, introduces the language itself, and an overview of most other content. XQuery 1.0 and XPath 2.0 Data Model (Latest Version Published on November 12, 2003) This document is an extension of the XML information set. Describe the basis of data items and form semantics that the query must be understood.

XQuery 1.0 and XPath 2.0 Formal Semantics (Latest Version Published on November 12, 2003) This document defines the underlying algebra of the language from the form. XML SYNTAX for XQuery 1.0 (Latest Version Published on December 19, 2003) This document is another syntax provided by people who like to use XML. The machines can be used anywhere. XQuery 1.0 and XPath 2.0 Functions and Operators Version 1.0 (Latest Version Published on November 12, 2003) This document describes the basic functions and operators of the Schema data type, the xQuery node, and node sequences. XML Path Language (XPATH) 2.0 (Latest Version Published on November 12, 2003) This document is a separate XPath documentation. XPath Requirements Version 2.0 (Latest Version Published on August 22, 2003) This document is a demand document for XPath. XSLT 2.0 and xquery 1.0 Serialization (Latest Version published on November 12, 2003) This document describes the details of XSLT 2.0 and XQuery 1.0. XML Query and XPath Full-Text Requirements (Latest Version Published on May 2, 2003) This document describes the functional requirements that Full-text Recommendation needs to be achieved. XML Query and XPath Full-Text Use Cases (Latest Version Published on February 14, 2003) This document provides the actual situation that the Full-Text specification is expected to process. XQuery specification work, the latest version of the date of the author is: November 12, 2003, you can get the latest work by the following address http://www.w3.org/tr/xQuery/. XQuery's module structure XQuery module includes three parts: namespace and pattern declaration, function definition, query expressions. The pattern declaration and function definition is not required. (1) The definition of name space is similar to: namespace xs = "http://www.w3.org/2001/xmlschema"(2) The concept of function definition is not unfamiliar with program designers, the following is an example of a function definition (Later we will use this example):

Define Function Depth Returns XS: Integer {{- An Empty Element Has Depth 1 -} {- OtherWise, Add 1 To Max_u100? Epth of Children -} IF (EMPTY ($ E / *) Then 1 Else Max (for $ c in $ E / * RETURN DEPTH ($ C)) 1} The first two parts are called Query ProLog. (3) Query expression is the most important part of the module structure, for example: for $ e in document ("emp.xml") // name return $ E

XQuery syntax and instance each xQuery query includes one or more query expressions. Common XQuery syntax: Path Expression, arithmetic expression with Boolean expressions, FLWR expressions, conditional expression, element constructors, and function calls (Function Call). Path Expression XQuery, the path expression follows the syntax of XPath 2.0. They use the PATH tag to select the node of interest from the XML document. Here, it is actually seen as a tree with nodes. The path expression provides a method of selecting a node from the node tree. Some examples of path expressions are mentioned below: • PARA selection for the paragraph element of the context node? * Select all elements of the context node? Text () Select all the text nodes of the context node? @Name Selects the Name property of the context node? @ * Select all attributes of the context node? Para [1] selection of the first PARA element node of the context node? Para [Last ()] The last PARA element node of the context node is selected? * / Para selection as a PARA element of the grandson node? / Doc / chapter [5] / section [2] Select DOC Chapter 5 Part III? Chapter // Para sends the grace of the child's node of the context Node CHAPTER to choose? // Para's grandson's grandson's node PARA of all documents, and the same as the context PARA elements in a document are selected? // ilist /: item Select all item elements in the same document as a context node with an OLIST dual-proceeding point? Select the context node? .//Para selection as the context Point grandson node's para element? .. Choose a double-pro-junction point for context nodes? ./@Lang selection of LANG attributes of the double-proof point of the context node? Para [@ type = "warning"] For all PARA elements of the child node with a Type property value for Warning? Chapter [title = "introduction"] Select Chapter with Title's Introduction.? Chapter [Title] Select one or more Title's Chapter? Employee [@secretary and @ssistant] Select all the child nodes of the child's node with the secondary attribute and the assistant property of EMPLOYEE. • Book / (chapter | appendix) / section Select Book in Book is part of Chapter or Appendix. Book / XF: ID (Publisher) / Name and XF: ID (Book / Publisher) / Name Returns the same result. Arithmetic expression and boolean expression (AritiTic and Boolean Expressions): This is the easiest part, the rules are similar to the grammar. We may wish to experience a few examples: Expressions: 10 * 5 Return to the result 50.0 Expression: if (10 * 5 = 50) THEN 55 ELSE 56 Return Result: 55 Expression:

NumFormat ("####", IF (10 * 5 = 50) THEN 55 ELSE 56) Return Result: 55 Expression:

NumFormat ("###. ##", IF (10 * 5 = 50) The "55.5555" Else "56.4444")

Please note that double quotes are needed. The calculation results will return: 55.56FLWR Expression FLWR (read "FLOWER") is one of the most important syntax types of XQuery. They look similar to SQL SELECT statements and have similar functions. FLWR "Flower", represents for, let, where, and returnge expressions. Where: (1) The FOR statement combines one or more variables together, each variable will be given a series of values, which is the corresponding expression, the entire content is worth it. Cartesian product. (2) Let statement binds the variable directly to a complete expression. If the for statement exists, the variable bindings created by letters will be added to the tuple generated by the for statement. If there is no for statement, the Let statement will generate a TUPLE with all variables. (3) Where statement applies to Tuple generated by the for statement and the Let statement. (4) The return statement contains an expression that is used to generate the result of the FLWR expression. Although a complete query statement can be completed in one line, XQuery still uses a grinding program design style to increase readability. In our data query, we use the indented program design mainly echo the nesting nested. For example, the "for" statement in the input and "for" statement in the output is aligned in the vertical row. Of course, this is our programming style, you can keep your own program design style. Example: For $ e in document ("Emp.xml") // name return $ E The above example returns a element for each element. The output of the output is as follows. The actual output will depend on the specific situation of the database. Different database output formats will be different. Hari john ... leu example: Let $ E: = Document ("Emp.xml") // name return $ E An example returns a forest consisting of elements and is enclosed in a element. HARI John ... leu Need to note that the above let statement returns a correct XML document. But for the FOR statement. If this is the user's hopes, it is very good. But if the user wants to get a format neat XML document, the FOR query will be able to modify it.

for $ e in document ("Emp.xml") // name return $ E The expression on the above will return a correct XML document. Please note that the label and in this example are returned. At the same time, it should be noted that in order to make the return statement properly, a single entity must be included. HARI john ... Leu

Note that the above query is returned is the element. If we want to display the value of the name without displaying any tags? We need to use the text () function. In the above query, "// Name" is changed to "// name / text ()", "$ E" is changed to "$ E / text ()". This method is used in the following query. For $ e in document ("emp.xml") // name return $ E / text () This query returns a series of names with no labels. This shows that the xQuery query can be customized according to the situation expected by the user or the application. XQuery is developed from a language called quilt. Harijohn ... Leu has another example.

For $ d in distinct ("Emp.xml") // DNAME) Let $ E: = Document ("Emp.xml") // entry [DNAME = $ D] Return Concat ($ d / text (), "Department HAS", NumFormat ("##", count ($ E)), "Employee (s).")

The following results will be output: Credit Department Has 1 Employee (s) .toys Department Has 2 Employee (s). SHOES Department Has 1 Employee (s) .Audit Department Has 1 Employee (s).

Conditional Expression In the previous example we have initially contacted some of the IF THEN ELSE statement usage, let's take a specific look at their grammar format: if Then Else < Expression II> In condition expression, the conditional expression should assign a value of the Boolean, or a type that can be converted to the Boolean value. If the value is true, the value of the result of the entire condition is the same as the value of the expression. Otherwise and the expression of the expression. The following is a simple example of the reader:

for $ e in document ("emp.xml") // entry return $ E / NAME, IF ($ E / SALARY.> =. 60000) THEN 100000 ELSE 200000

Element Constructor When writing query statements, you may need to create a new element as part of the output. Element constructors provide such functions. An example of a simple element constructor is as follows: xQuery language The content of the new element can be specified, or it can be calculated. Function Call This is identical to the concept of functions here and function concepts in other programming languages. Let's take an example: The following function returns the maximum depth of the XML document. Then a caller of calculating a depth of PartsList.xml document. Define Function Depth Returns XS: Integer {{- An Empty Element Has Depth 1 -} {- OtherWise, Add 1 To Max Depth of Children -} IF (EMPTY ($ E / *)) THEN 1 Else Max (for $ c in $ E / * RETURN DEPTH ($ C)) 1}

Depth (Document ("Partlist.xml")))

Advanced topics and application requirements and in xQuery, ask for easy completion through Let and SUM, in SQL, use "Group By" and "Having" statement. However, in SQL, "Group By" and "Having" can only correspond to an from statement, and multiple Let statements can be used in XQuery. The following example is to return the total amount of salary in accordance with the department.

For $ d in distinct (Document ("Emp.xml") // DNAME) Let $ s: = Document ("Emp.xml") // entry [DNAME = $ D] / SALARY RETURN $ D, SUM ($ S)

If you want to get wages greater than or equal to 70000, we should do the following query. for $ d in distinct ("Emp.xml") // DNAME) Where every $ s in document ("emp.xml") // entry [DNAME = $ D] Satisfies ($ S / SALARY> = 70000) Return $ D, avg (document ("emp.xml") // entry [DNAME = $ D] / SALARY)

The most important thing worth noting is. In KWEELT, we use ".> =." Rather than use "> =" e usees ".> =." The biggest obstacle is that Kweelt does not support the word. Therefore, "Every X Satisfies C (x)" does not work in Kweelt. Imagine an equivalent expression "NOT (Some X Satisfies Not (C (X))" Although this still contains the word, it provides us with an indirect way: First we collect NOT in $ s (C (C) x) Elements X, use "not exists" in the WHERE clause to filter out some of the through For statement but does not meet this requirements. The remaining part is required, and will execute after the return statement It was released. So we got the following query. for $ d in distinct ("Emp.xml") // DNAME) Let $ s: = document ("emp.xml") // entry [DNAME = $ D and Salary. <. 70000] / Salary WHERE NOT (exists) Return $ D, avg (Document ("Emp.xml") // entry [DNAME = $ D] / salary)

This also shows the imperfectness of SQL syntax, and students in Computer Science and Technology must attach great importance to this understanding. When a thing is standardized, how much is it. The XQuery is very good at the beginning of the beginning. The imperfectness of KWeelT may be derived from the pre-standard version of XQuery. Sort sorting is performed by default. As shown in the following example: for $ I in (1, 2), $ J in (3, 4) Return {$ i} {$ j} < / tuple>