3.4. Map order
This section discusses the object-relational mapping how to process the order.
3.4.1. Same order, level order, and document order
Sibling means "brothers and sisters." That is, the same level or PCDATA is an element or PCDATA with the same parent element. In other words, they appear in the same content model. For example, if the document is represented as a tree in the previous chapter, this is easy to display those elements: these elements are at the second level of this level, and there are A as their parent elements.
A
_______________________________________________
| | | | | | | | | | |
THIS TEXT C Makes B No Sense C Except as b An Example
| | | | | |
CC BBBB CCCC BB
Note that the element of the third level at the level is not as level because they do not share the same parent element. This also points out the order of the same level, which is the order of their parent element, and the order of hierarchy, which is the level of the child in the tree that represents the document, the difference between the two. Different documents, it is the order of elements and text appears in an XML document. E.g:
The order of the same level (only one of the places of the same level is not displayed):
A
_______________________________________________
| | | | | | | | | | |
THIS TEXT C Makes B No Sense C Except as b An Example
1 2 3 4 5 6 7 8 9
| | | | | |
CC BBBB CCCC BB
Hierarchical order:
1 a
_______________________________________________
| | | | | | | | | | |
2 THIS TEXT C Makes B No Sense C Except as b An EXAMPLE
| | | | | |
3 CC BBBB CCCC BB
Document order:
A
1
_______________________________________________
| | | | | | | | | | |
THIS TEXT C Makes B No Sense C Except as b An Example
2 3 5 6 8 9 11 12 14
| | | | | |
CC BBBB CCCC BB
4 7 10 13
According to XML, the same order is important. In fact, this depends on the application. For example, in a data-centric application, use an XML document to pass an object or table, the same order is usually irrelevant because the object-oriented language does not have the concept of the order between class properties. Similarly, the relational database is not the concept of the order between columns. Therefore, the same order is not important in the following documents:
They are all mapped into the following objects and rows in the table:
Object table
========================= ================================================================================================================================================================================================== ==========
Table Parts
Object Part {-------------------------------
Number = 123 ==> Number DESC PRICE
DESC = "Turkey Wrench" -----------------------
Price = 10.95 123 Turkey Wrench 10.95
(One of the main exceptions to this is when you must match a specific DTD in data-centric documents. This occurs when an application must verify the document, such as when they come from unknown or not trusted sources. Although in this case, the "all group" of XML Schemas can be homped by allowing a group of child elements to be helped in any order, but they do not support repeating sub-elements.)
On the other hand, in a document-centric application, the usual document is designed for people's consumption, and the order of the same level is very important. For example, I am likely to like the first comment rather than the second:
ronald bourret is a
Excellent writer .
Only fool
I don't read his work.
ronald bourret is a
Fool . only
Excellent writer
I don't read his work.
Object-Relational Mapping can keep the same level, which will be seen below, although it is actually rarely supports it. By mapping the column in the simple element type to the column in the table, and the reference to the complex element type is mapped to the primary key, the foreign key contact, its instinct is in order. The order of documents is retained in the order of hierarchical order and the same level. 3.4.2. Map of the same order
Because the object-oriented language has no concepts between the class properties, the relationship database does not have the order of the order between the columns, and the same order value must be stored independently with the data value. One way to do this is to intervene independent class properties and columns in which order values are stored. Another way to do this is stored in the mapping of itself.
3.4.2.1. Order properties and columns
Use the order class attributes and subsequences to store order values. They are opposite to data class properties and data columns. Each of the referenced elements or PCDATAs that are important for the order of the items require a class attribute or column. For example, consider an example of the above mixed content. The same order in the DTD is mapped to the order class attribute:
DTD class
================================================== =====
Class a {
String [] PCDATA;
int [] pcdataorder;
string [] b;
==> int [] border;
string [] C;
Int [] CORDER;
}
And then map to the sequence:
Class table
============================================================================================================================================= ==============
Table PCData
Class a {----- column a_fk
String [] pcdata; / column pcdata
int [] pcdataorder; / column pcdataorder
String [] B; Table A / Table B
Int [] border; ==> column a_pk -------- Column a_fkstring [] C; / Column B
int [] CORDER; / Column Border
} / Table C
/ ---- COLUMN A_FK
Column C
Column Corder
Note that the order class attribute stored in the standard is stripped in parallel.
The following example shows the use of the order attribute to retain the same order in the "Makes-NO-Sense" example. It is important thing to pay attention to here that all order class properties share the same order space. A sequence value that appears in one order class attribute does not appear in another order class property.
Class table
================================= ================================================================================================================== ====================
Table PCData
A_fk pcdata pcdataorder
---- ---------------------
1 THIS TEXT 1
Object a {1 Makes 3
Pcdata = {"this text", 1 no Sense 5
"Makes", 1 Except as 7
"No Sense", Table A 1 An Example. 9
"eXCEPT AS", A_PK
"an esample."} ==> ---- TABLE B
PcdataRDER = {1, 3, 5, 7, 9} 1 a_fk b border
B = {"bbbb", "bb"} ---- ---- ------
Border = {4, 8} 1 BBBB 4
C = {"cc", "cccc"} 1 BB 8
CORDER = {2, 6}
} Table CA_FK C CORDER
---- ---- ----------
1 CC 2
1 CCCC 6
Although the order class attribute is most commonly used to maintain the order in the mixed content, they can also be used with the element content. For example, consider the following element type definition. Because B can in A, it can appear once, it is stored in a stand-alone attribute table. There is no sequence class attribute, which will not be able to determine how the B child element is seized. (Note that this order cannot be used, because the relational database does not guarantee a row in any particular order.)
3.4.2.2. Store in the mapping
In many cases, the same order is only important when verifying; in addition to verifying a document, it is not concerned about the same order. Especially in data-centric documents. In this case, it is sufficient to store the order information in the mapping itself.
For example, the following content model is given, and the order of the sub-elements that can store A is B and then the information of d:
In particular, it is limited to store order information in the mapping. For example, consider the following content model:
Texture Matching this content model The document requires that software first decides how much data can be obtained to construct B elements. If there is only enough data to construct a B element, it cannot be the first B element because the second B element is necessary.
Let most software so troublesome are unlikely. Turns, only those content models of all the same elements of the same element type together provide reasonable restrictions. It is sufficient for many data-centric content models and can be implemented by storing each element in the mapping.
For example, the order of the peer elements in the following elements can be mapped. Note that both of Author and Editor can be given the same order value or different values in the third content model; if they are given different values, all elements of one type will appear in other types of any elements. Front.
When the order information is only stored in the mapping, as long as the content model contains more than one of the same type of element, the non-destructive flow of the document is not possible. For example, consider the following content model:
Although the mapping can tell the software that all B elements must appear in front of the C element, it cannot specify the order of the B element. So, if the data is transmitted from the document containing this content model and then transfer it back, the B element cannot be guaranteed in the order in the original document. Fortunately, this is usually not a problem with data-centric documents.
3.5. Mapping properties
As seen in the previous, map the attribute map to the scalar property. This section will discuss details of this mapping, there are other points.
3.5.1. Map single value and multi-value properties
There are two different properties: single value (CDATA, ID, IDREF, NMTOKEN, Entity, Notation, and Enumeration) and multiple values (idRefs, NMTOKENS, and Entities). It is expected that it is possible to map them into single-value class properties (the terminal is column) and multi-value class properties (then attribute table). E.g:
DTD class ======================================== ======= ====
Class A {Table A
COLUMN B; Column B
D cdata #Required> ==> string c; ==> Column C
string d; Column D
}
with:
DTD class table
======================================= ======================================================================================================================================= ==
Class A {Table A
D idrefs #implied> ==> string c; ==> Column B
string [] d; color
} TABLE D
Column a_fk
Column D
It is not important in the order of XML information set properties. For example, the following two XMLs are considered to be equal. Therefore, there is no need to maintain the order of the properties to maintain the order of properties, although it is entirely possible.
C = "CCC" D = "ddd" />
B = "bbb" D = "ddd" /> On the other hand, the order in which the value appears in the multi-value attribute is considered to be important. This is the case of the same level element and PCDATA, and the order class attribute can be used to maintain information about the order of the value in the multi-value attribute. However, there is an important difference between the order class attribute used for the same level and PCData, and the order class attribute for multi-value attributes: each order class attribute for multi-value attribute has its own order. space. This can be seen in the examples below: XML object ======================================================================================================================================= Object a { ID / IDREF (s) attribute mapping is the primary key, foreign key contact. For example, the above document can be stored in the database as the following list and columns: Table a Column a_pk ... / / / / TABLE B TABLE C Column a_fk column a_fk Column Ref_d Column Ref_D ... / / / / Table D Column a_fk COLUMN ID ... When you store ID / IDREF (s) attributes in the database, the data transfer software requires a careful thing that the ID only guarantees that it is unique in a given XML document. Therefore, if data from more than one document is stored in the same table, the ID is unique. The solution to this problem is "Modified" ID in some way. You can map the property to two columns, one column contains a value for each document, and the other column contains ID, or modified the ID itself, such as a prefix with a unique value. There are similar problems when data is transmitted from the database to the XML document. If the data retrieved origines from more than one document, the data transfer software needs to make sure the ID value is unique. This may involve changing one or more values, together with all the values of all the idRef (s) properties that reference them. Currently, most products do not support ID / IDREF as an attribute that is different from other properties. 3.5.3. Mapping Note Attribute Use a notation in an XML document to remind the application how to handle an element or an unsatisfactory entity. For example, the following "XHTML" notes can tell to apply this element containing XHTML and should use a browser to display:
