Oracle advanced (XIV) explanation of escape characters

One 、 Preface

stay DAO layer , Relating to an entity class SQL Statements are usually configured in XML In file , as follows SQL Statement configuration ：

<select id=&quot;queryUsrnmTypOprorder&quot; parameterClass=&quot;java.util.HashMap&quot;
resultMap=&quot;queryCasprjstfsetListMap&quot;> 
<![CDATA[ 
 select 
 a.usrnm, 
 a.typ, 
 a.oprorder 
 from fset a
 inner join casprjstfapr 
 using (stfaprid) 
 where casprjstfapr.jugsumid = #jugsumid# 
 and casprjstfapr.blnorg = #blnorg# 
 and isvld = '1' 
 order by a.typ asc, a.oprorder asc 
]]> 
</select>


By

![CDATA[]]

The content of this tag will be represented as plain text , such as

![CDATA[]]

Represent text content “

<

”.

This tag is used for xml In the document , Let's first look at the use of escape characters . We know , stay xml in ,”<”、”>”、”&” Etc. characters cannot be stored directly , otherwise xml Error will be reported in grammar check , If you want to xml Use these symbols , It must be escaped as an entity , Such as ”

<

”、”

>

”、”

&

”, In this way, it can be saved into xml file .

When using program reading , The parser will automatically convert these entities back to ”<”、”>”、”&”. for instance ：

Two 、 Be careful

• There cannot be spaces between escape sequence characters ;

• The escape sequence must be in ”
  ;
  ” end ;

• Appearing alone ”
  &
  ” Will not be considered the beginning of the escape ;

• Case sensitive .

stay XML in , The characters that need to be escaped are ：

But strictly speaking , stay XML There are only ”

<

” and ”

&

” It's illegal. , The other three are legal , however , It's a good habit to escape them all .

Anyway , The characters before escape are also good , The escaped characters are good , Will be xml Parser parsing , For convenience , Use

![CDATA[]]

To include not being xml What the parser parses . But here's the thing ：

• This part can no longer contain ”
  ]]>
  ”;

• Nesting is not allowed ;

• ”
  ]]>
  ” This part cannot contain spaces or line breaks .

Last , say something

![CDATA[]]

and xml Transfer character relationships , Do the two of them seem to have the same sensory function ？

3、 ... and 、 Application scenarios and requirements

• ![CDATA[]]
  Not in all cases , Escape characters can ;

• For short strings
  ![CDATA[]]
  It's wordy , For long string escape characters, the readability is poor ;

• ![CDATA[]]
  Express xml Parsers ignore parsing , So faster .

Four 、 Expanding reading  DELETE and TRUNCATE  The difference between

4.1  grammar

delete from table aa 
truncate table aa 

4.2  difference

• delete from
  The conditions can be written later ,
  truncate
  Can not be .

• delete from
  The records are deleted one by one , Every deleted record will be entered into the log , and
  truncate
  Delete the whole page at one time , Therefore, only page release is recorded in the solstice , in short
  ,delete from
  Update log ,
  truncate
  Basically not , Less transaction log space is used .

• delete from
  After deleting the empty table , Will keep an empty page ,
  truncate
  There will be no pages left in the table .

• When using row locks to execute  
  DELETE
    When the sentence is , The rows in the table will be locked for deletion .
  truncate
  Always lock tables and pages , Instead of locking rows .

• If there is
  identity
  The resulting self accretion id Column ,
  delete from
  After that, it still increased from the last number , That is, the seed does not change , and
  truncate
  after , The seed will return to its original state .

• truncate
  Not trigger
  delete
  The trigger of , because
  truncate
  The operation does not record the deletion of each line .

4.3  summary

• truncate
  and  
  delete
  Delete data only , Do not delete the structure of the table ( Definition );

• drop
  Statement will delete the constraint that the structure of the table is dependent on (
  constrain
  ), trigger (
  trigger
  ), Indexes (
  index
  );  Stored procedures that depend on the table / The function will keep , But it became
  invalid
  state .

• delete
  The sentence is
  dml
  , This operation will put
  rollback segement
  in ,
  The transaction does not take effect until it is committed
  ; If there is a corresponding
  trigger
  , Execution will be triggered ;
  truncate
  ,
  drop
  yes
  ddl
  ,  The operation takes effect immediately , The original data will not be put into
  rollback segment
  in , Cannot be rolled back , Operation does not trigger
  trigger
  .

• delete
  Statement does not affect the
  extent
  ,  High waterline (high watermark) Keep in place ; obviously
  drop
  Statement to free all the space occupied by the table .
  truncate
    Statement releases space to by default  
  minextents
  individual  
  extent,
  Unless used
  reuse storage
  ; 
  truncate
  Will reset the high waterline ( Go back to the beginning ).

• Speed , Generally speaking : 
  drop> truncate > delete
  .

• Security : Careful use
  drop
    and
  truncate
  , Especially when there is no backup , Otherwise, it's too late to cry .

• On use , Want to delete some data lines
  delete
  , Pay attention to take
  where
  Clause .  The rollback segment should be large enough ; Want to delete table . Of course with
  drop
  ; Want to keep the table and delete all the data . If it's not about business , use
  truncate
  that will do .  If it's about business , Or want to trigger
  trigger
  , Or use it
  delete
  ; If it's defragmenting the table , It can be used
  truncate
  keep pace with
  reuse stroage
  , Re import / insert data .