Decimal data type

Similar Questions

• Probable bug with decimal data type mapping in storage of XmlType relational object

  Hello

  I'm having a problem with what appears to be a bug me when Oracle creates the mapping object / relational schema for a data type "XSD: Decimal":

  I want to store geo codes 6 fractional digits, so I need to 3 digits to the left and 6 digits to the right of the decimal point.

  In the XSD data type restrictions, which would give me a 'fractionDigits' and a 'totalDigits' 9 6.

  Oracle must convert this number (9, 6).

  However, what Oracle made out that is a NUMBER (15.6)

  In order to get an Oracle to generate a NUMBER (9, 6), I need to qualify the 'totalDigits' 3, which is accepted by Oracle, but makes my schema invalid for virtually any other application XML, because it violates the official specs of XSD:

  http://www.w3.org/TR/xmlschema-2/#RF-fractionDigits:

  4.3.12.4 constraints on fractionDigits components of schema

  Schema component constraint: fractionDigits less than or equal to totalDigits

  It is an error for fractionDigits is greater than totalDigits.

  Is it possible to have the good totalDigits of 9 in my scheme and force the Oracle to translate that into NUMBER (9,6)?

  Maybe some attribute oraxdb I do not know yet?

  I'm surprised that this has not been found earlier.

  There was an announcement about this before, but he did not get the real answers:

  TotalDigits and fractionDigits to number conversion problem

  Here is my complete code example:

  -- Registering the schema
  BEGIN
  DBMS_XMLSCHEMA.REGISTERSCHEMA(
  'http://localhost/decimaltest.xsd',
  '<?xml version="1.0" encoding="UTF-8"?>
  <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" elementFormDefault="qualified" xmlns:oraxdb="http://xmlns.oracle.com/xdb" oraxdb:storeVarrayAsTable="true" oraxdb:schemaURL="http://localhost/decimaltest.xsd">
    <xs:element name="dataTypeTests" oraxdb:SQLName="ZZZ_DATA_TYPE_TEST" oraxdb:SQLType="ZZZ_DATA_TYPE_TEST_TYPE" oraxdb:defaultTable="" oraxdb:tableProps="TABLESPACE CATALOG NOLOGGING" oraxdb:maintainDOM="false">
    <xs:complexType>
    <xs:sequence>
    <xs:element name="geoCode" type="geoCodeType" oraxdb:SQLName="GEO_CODE" oraxdb:maintainDOM="false"/>
    </xs:sequence>
    </xs:complexType>
    </xs:element>
    <xs:simpleType name="geoCodeType">
    <xs:restriction base="xs:decimal">
    <xs:totalDigits value="9"/>
    <xs:fractionDigits value="6"/>
    </xs:restriction>
    </xs:simpleType>
  </xs:schema>',
  gentables=>FALSE);
  end;
  /
    
    
  -- Display the auto-generated Oracle custom type
  describe ZZZ_DATA_TYPE_TEST_TYPE;
    
  -- Clean up
    
  begin
    DBMS_XMLSCHEMA.DELETESCHEMA(
    'http://localhost/decimaltest.xsd',
    DBMS_XMLSCHEMA.DELETE_CASCADE
    );
  end;
  /
  

  In the XSD data type restrictions, which would give me a 'fractionDigits' and a 'totalDigits' 9 6.

  Oracle must convert this number (9, 6).

  And that's where you're wrong.

  fractionDigits = 6 means at least 6 digits located to the right of the decimal point.

  This is different from the semantics of NUMBER (m, n), where n is exactly n fractional digits stored.

  123.456789 is a valid number by the XSD.

  123456789 is as valid a number, but of course it does not fit in a NUMBER (9.6).

  Oracle must create a NUMBER (15.6) to store any number of valid schema, it is a necessary condition.

  But this is not a sufficient condition: obviously, any number in a NUMBER (15.6) is not valid per the XSD, and this is where the schema validation occurs to enforce the constraints of two facets.

  Here are some examples:

  BEGIN

  DBMS_XMLSCHEMA. REGISTERSCHEMA)

  schemaURL-online "decimaltest.xsd."

  schemaDoc =>

  '

  http://www.w3.org/2001/XMLSchema"elementFormDefault ="qualified"xmlns:oraxdb ="http://xmlns.oracle.com/xdb"oraxdb:storeVarrayAsTable ="true"oraxdb:schemaURL ="http://localhost/decimaltest.xsd">" "

  '

  genTypes-online true

  genTables-online fake

  enableHierarchy-online dbms_xmlschema. ENABLE_HIERARCHY_NONE

  );

  end;

  /

  SQL > declare

  doc 2 xmltype: = xmltype)

  3'123456789"

  .createSchemaBasedXML('decimaltest.xsd') 4);

  5. start

  6 doc.schemaValidate ();

  7 end;

  8.

  PL/SQL procedure successfully completed.

  SQL > declare

  doc 2 xmltype: = xmltype)

  3'123456789.1'

  .createSchemaBasedXML('decimaltest.xsd') 4);

  5. start

  6 doc.schemaValidate ();

  7 end;

  8.

  declare

  *

  ERROR on line 1:

  ORA-31154: invalid XML document

  ORA-19202: an error has occurred in the processing of XML

  LSX-00215: value '123456789.1' whose total numbers exceed 9

  ORA-06512: at "SYS." XMLTYPE", line 354

  ORA-06512: at line 6

  SQL > declare

  doc 2 xmltype: = xmltype)

  3'12.3456789'

  .createSchemaBasedXML('decimaltest.xsd') 4);

  5. start

  6 doc.schemaValidate ();

  7 end;

  8.

  declare

  *

  ERROR on line 1:

  ORA-31154: invalid XML document

  ORA-19202: an error has occurred in the processing of XML

  LSX-00211: "12.3456789", including the fractional numbers exceeds 6

  ORA-06512: at "SYS." XMLTYPE", line 354

  ORA-06512: at line 6

• Calculate w / Numeric or Decimal Data Types

  Working in ASP VBscript, connected to a SQL Server 2000 database.
  
  Trying to average values returned in a Recordset into a session variable, using the following:
  
  < %
  Dim VComm
  Dim vResp
  Dim vLead
  Dim vQual
  Dim vPack
  Dim vTotal
  
  vComm = rsQuality.Fields.Item ("Comm"). Value
  vResp = rsQuality.Fields.Item ("Resp"). Value
  vLead = rsQuality.Fields.Item ("Lead"). Value
  vQual = rsQuality.Fields.Item ("Qual"). Value
  vPack = rsQuality.Fields.Item ("Pack"). Value
  % >
  
  < % Session("AVG") = (vComm + vResp + vLead + vQual + vPack)/5.00% >)
  
  It works fine as long as the data type for the columns of the source in the SQL Server database is "Integer", numbers of i.e.whole. When the data type is 'comma' or 'digital', the following error is generated when the page is served:
  
  Error running Microsoft VBScript '800a000d '.
  
  Type mismatch
  
  Samples/pages/report_views/quattro_overview.asp, line 216
  
  Part of what confuses me, is that the error only occurs when you try to calculate in expression... A test, I set the session variable = to each of the variables (with the decimal or numeric data type) indidually and the session variable dspilays fine, IE as a value(e.g.") 1.7 ") decimal. But when I try to calculate these decimal values, as shown in the expression above, I get the error message.
  
  Any help is greatly appreciated.
  
  Kind regards
  
  Chris Robinson
  
  

  18 Sep 2006 macromedia.dreamweaver.appdev, stratcat900 wrote:

  > Part of what confuses me, is that the error only occurs when you try
  > to calculate in expression... A test, I set the session
  > variable = to each of the indidually of variables (with the data type as
  (> decimal or numeric) and the session variable dspilays fine, as
  > a decimal value (e.g." 1.7 "). But when I try to calculate these
  > decimal values, as shown in the expression above, I get the error
  > message.
  >
  > Any help is greatly appreciated.

  Try their casting as Double (CDbl()):

  vComm = CDbl (rsQuality.Fields.Item("Comm"). Value)
  vResp = CDbl (rsQuality.Fields.Item("Resp"). Value)
  vLead = CDbl (rsQuality.Fields.Item("Lead"). Value)
  vQual = CDbl (rsQuality.Fields.Item("Qual"). Value)
  vPack = CDbl (rsQuality.Fields.Item("Pack"). Value)

  http://DevGuru.com/technologies/VBScript/13905.asp

  --
  Joe Makowiec
  http://Makowiec.NET/
  E-mail: http://makowiec.net/email.php

• Problem SQL Loader: load decimal data type

  I'm trying to load the following data to my oracle db using sqlldr.
  
  The DATA are
  ===============================================================
  10500158 919602961512 15014122109153571044261 BadSong_MessageSent sorry, we could not understand the song that you have recorded. Please place your phone near the music and try again. To listen to songs in your playlist, call the 578787 (Re.1 per minute) 2010-06-26 15:36:22 Rajasthan 9602961512_2010_06_26_15_35_57_IST_727298.wav 10.0227 geetweb 10839 songID is null raj 0.96 true 40,8225 0
  
  ===============================================================
  
  It is separated from the tab (although it appears so after validation)
  
  and my control file looks like this.
  ===============================================================
  load data
  INFILE ' / home/oracle/sonam/data/SCfilter_backend_dump.test.csv'
  TRUNCATE
  IN THE SCfilter_backend_dump TABLE
  FIELDS CANCEL BY X ' 09 "trailing nullcols)
  SERIALNO fill tank (20),
  UserID char(20),
  TxnId tank (45),
  SONGID char(20),
  vcode char(20),
  title char (100),
  album tank (100),
  status tank (45),
  MessageSent tank (4000).
  DateTime TIMESTAMP "yyyy-mm-dd hh24:mi:ss"
  circle char(20),
  name of file (100) tank.
  EXTERNAL DECIMAL FileLength
  Server char (20),
  EXTERNAL DECIMAL ResponseTime,
  error char (100),
  language char(20),
  DECIMAL EXTERNAL, speed
  signalquality DECIMAL EXTERNAL,
  BlankCount INTEGER (40),
  isfilevalid char (20))
  
  ===============================================================
  
  
  While my table of looks like this
  ===============================================================
  SQL > desc SCFILTER_BACKEND_DUMP;
  Name Null? Type
  ----------------------------------------- -------- ----------------------------
  USERID NOT NULL VARCHAR2 (20)
  TXNID NOT NULL VARCHAR2 (100)
  SONGID VARCHAR2 (20)
  VCODE VARCHAR2 (20)
  TITLE VARCHAR2 (100)
  ALBUM VARCHAR2 (100)
  STATUS VARCHAR2 (45)
  DATETIME NOT NULL TIMESTAMP (6)
  CIRCLE VARCHAR2 (45)
  FILENAME VARCHAR2 (100)
  FILELENGTH NUMBER (7.2)
  SERVER NOT NULL VARCHAR2 (20)
  RESPONSETIME NUMBER (7.2)
  ERROR VARCHAR2 (100)
  LANGUAGE VARCHAR2 (20)
  SPEED NOT NULL NUMBER (7.2)
  SIGNALQUALITY NUMBER (7.2)
  BLANKCOUNT NUMBER (38)
  ISFILEVALID VARCHAR2 (20)
  MESSAGESENT CLOB
  
  ===============================================================
  
  
  and I get the following error in my log file.
  
  =================================================================
  Table SCFILTER_BACKEND_DUMP, loaded from every logical record.
  Insert the option in effect for this table: TRUNCATE
  TRAILING NULLCOLS option in effect
  
  Column Position Len term Encl. Datatype name
  ------------------------------ ---------- ----- ---- ---- ---------------------
  SERIALNO FIRST 20 WHT CHARACTER
  (FIELD FILL)
  USER NAME THEN A CHARACTER 20 WHT
  TXNID WHT 45 CHARACTER FOLLOWING
  CHARACTER 20 WHT SONGID NEXT
  VCODE NEXT WHT 20 CHARACTERS
  NEXT CHARACTER WHT 100
  ALBUM NEXT WHT 100 CHARACTERS
  45 WHT CHARACTER DEPENDING ON STATUS
  CHARACTER OF 4000 WHT FOLLOWING MESSAGESENT
  DATETIME NEXT * WHT DATETIME yyyy-mm-dd hh24:mi:ss
  CIRCLE AND THEN A CHARACTER 20 WHT
  NAME OF THE FILE AND THEN CHARACTER WHT 100
  FILELENGTH NEXT * CHARACTER WHT
  CHARACTER NEXT 20 WHT SERVER
  RESPONSETIME NEXT * CHARACTER WHT
  ERROR FOLLOWS WHT 100 CHARACTERS
  LANGUAGE NEXT WHT 20 CHARACTERS
  THEN THE SPEED * CHARACTER WHT
  SIGNALQUALITY NEXT * CHARACTER WHT
  BLANKCOUNT INTEGER THEN 40
  ISFILEVALID WHT 20 CHARACTER FOLLOWING
  
  value used for the parameter LINES increased from 64 to 42
  Sheet 1: Rejected - error on the SCFILTER_BACKEND_DUMP table.
  ORA-01460: dead letter or unreasonable conversion requested
  
  
  Table SCFILTER_BACKEND_DUMP:
  0 rows successfully loaded.
  1 row not loaded due to data errors.
  0 rows not loading because all WHEN clauses were failed.
  0 rows not populated because all fields are null.
  
  
  The space allocated to bind table: 251580 bytes (42 lines)
  ===============================================================
  
  
  I'm not able to get the reason why this method is a failure for decimal data loading, while its mentioned in textbooks.
  What is the problem with the description of the control file. In fact WHOLE EXTERNAL works great for loading If integer data only.
  
  Thank you and best regards,
  Sonam

  I guess that your error is not because of the decimal specification, but rather because INTEGER (40) -it is not allowed. Integer means binary integer and you can only specify a length 1,2,4,8 (fullword). What you probably want to do - load a representation of an integer (as in "25") tank, so you need to specify

  blankcount  INTEGER EXTERNAL(40)
  

  Best regards

  Maxim

• How to handle the data type DECIMAL MySQL converted to DOUBLE in OBIEE 10 gR 3?

  Hi all
  
  I have a problem in OBIEE 10g to I use the MySQL database as datasource direct to the server of the OBI and I have a column in a MySQL table which was designed as the DECIMAL data type (18.0).
  This column is used to measure a length, and values are included in seconds (without decimals), in the form as:
  "1200"
  in the MySQL database.
  
  After the import of the MySQL table associated with the repository of the OBI, the OBIEE converts the DECIMAL data type column to the DOUBLE PRECISION data type automatically.
  
  My goal is to reach the values in this form, certainly in the sense of the term (not at the time!):
  hh mm ss
  
  In order to accomplish, I tried to change the format of column with the ROUND function:
  ROUND (db.table.column / * 3600 * 2)
  
  The problem is that it doesn't work well if values can be divided with no rest for example
  original value: 7200
  new value (ROUND): 2
  
  In the case of my example:
  original value: 1200
  new value (ROUND): 0
  It's false, because 1200 seconds are not equal to 0 hours, I want to see two hours of _0.33333_ OR I much prefer more _0 hour 20 m_
  
  Anyone have any ideas?
  I tried to make a view of database and importing that type of CHAR data in the database, but is not the solution too, because if I filter BI app or try to make a SUM the BI would drop errors...
  
  Kind regards
  Laszlo

  Hello
  SE refer to this link, it is exactly will meet your need.

  OBIEE 11 g - change seconds in the format hh: mm: DD
  OBIEE functions

  Share updates... :)

  check if useful/correct...

• Dynamically find the Integer and the Data Type of a Decimal column

  Table: test (INTE number, DECISION number (7.2)) of the
  
  Try to dynamically determine the type of column.
  
  Use select user_tab_columns data_type. It returns only the NUMBER that does not indicate whether the column is an integer or a decimal data type.
  
  Is there a way to dynamically determine whether a column is around or type decimal?
  
  Thank you

  create table test (INTE, DECI number number (7.2));

  Select column_name, data_type, data_precision from dba_tab_columns where table_name = 'TEST '.

  COLUMN_NAME DATA_TYPE DATA_PRECISION
  ------------------------------ ---------------------------------------------------------------------------------------------------------- --------------
  NUMBER OF INTE
  DECI, NUMBER 7

• convert the money data type in a decimal format 2.

  What is the best way to convert the money data type in a decimal 2 format in ms sql 2005 for use in my applications.

  This?

  CAST (tr.depositReceivedAmount AS decimal (10, 2))

  I respectfully disagree with the idea that you must change the query column a type of 'money' data to something else.

  In most database servers, "«argent» Silver" is a data type that is designed to provide a very homogeneous behaviour with regard to the arithmetic accuracy.  In Microsoft Access, the representation is a the whole scale.  MS SQL Server, it is obviously similar.  Ditto Oracle and others.

  You like that the money data type in the database for this clarification, because "hell has no fury like an accountant looking for a bad penny."   The column of database storage formats are designed to satisfy the accountants, and it's a good thing.

  Meanwhile, you also want to take care as to exactly how you are dealing with values.  There are several points where the district could take place.  You don't have available the strongest possible manipulation of the floating of the data types in ColdFusion.  You also are somewhat at the mercy of any interface software can be between you and that SQL server, you can use.  "It is correct to all about values , but not several times."

  I suggest to round the value just before the display and user input must be two decimals.

  Then, you may need to do some things at the end of SQL server.  For example, when you update a value in the table, you may need to use logical server to explicitly truncate the value to two decimal places, as well as an update of "$ 34,56" explicitly updates the column "$34,5600.."  (This kind of thing should happen in the context of SQL Server).  You know that the user entry has exactly two significant digits, but maybe (maybe not...)! SQL server may not know this.  You want to ensure that the stored server internally represents exactly two digits, when the origin of a user input value.

  Do not err on the side of "convenience" or "which seems." good on the screen  (If you do, get ready get telephone communications of the accountants, always at inappropriate times of the night.)

• Convert or map Typedonnees decimal Transact-SQL for Oracle Number data type?

  MSSQL 2005
  Oracle 10.2 g
  
  In a MSSQL table, I have a column with the data type set on (decimal (1.0), null) with the values of line-1. (695 lines in total)
  
  In the Oracle table, the proposed mapped column is a number data type. When I import data, I received 695 errors with the message "invalid value for the field. How to properly convert or map Decimal (MSSQL) Transact-SQL for Oracle Number data type for a negative value?
  
  Thank you.

  How do you load data into Oracle? What tool or programming language you are using? Can you post something cause what you stated in your post should work, but there may be some ODBC, or other type of conversion factors to be taken into account.

   > create table t1 (field1  number(1,0));
  
  Table created.
  
   > insert into t1 values (-1);
  
  1 row created.
  
  UT1 > select * from t1;
  
      FIELD1
  ----------
          -1
  

  HTH - Mark D Powell.

• To change the data type of control/led

  Is it possible to change the underlying data type of the indicator or a control?  I a VI that I want to use that accepts any type of data (essentially data varriant) and display it accordingly.  I have different controls selected to display data of different types in the sense of data vs decimal boolean data vs integer, but certain types of data will be of type uint32 vs int16 vs uint64 etc..  Is it possible to programmatically change the data type, so I need to have all the combinations of indicators?

  The short answer is no. The long answer is that LabVIEW has not 'generic' controls that allow you to display any type of data (apart from the variant of course). There are several ways to address this problem, but all require you as the programmer to manage explicitly each type of data you could possibly meet. Here are some possibilities (in no particular order), please let me know if you would like to help a particular implementation of these.

  1. Force incoming data to variant, use the info option to determine the data type of the original data in a case structure (open G can do this VERY easy), have an indicator on the front panel for each type of data you want and hide those unused
  2. Use a sub-group of experts on the front panel and sub - vi call of to view specific types (that lends itself very well to encapsulate your data in a class and you leave OOP allows to expand into different types of data)
  3. Use a tab with the variant (similar to option 1) control put each kind of data in its own tab, enable only the current
  4. Do a Control-X to accept a Variant option 1 under the hood
  5. Just as a string of format and use a string indicator

  That's all I can think of right now, once again, let me know if you need more information/help on the options.

• Money data type with 4 decimals acurracy

  Hello

  I want to insert values of money on an Oracle database and the need to an accuracy of four decimal places. Is NUMBER (19.4) the correct type?

  In light of this PL code:

  create table foo (money_col number(19,4));
  
  DECLARE
  i PLS_INTEGER;
  BEGIN
  FOR i IN 1..10000 LOOP
      INSERT INTO FOO VALUES(i/10000);
  END LOOP;    
  COMMIT;
  END;
  /
  

  It always inserts 10,000 different numbers?

  Thanks in advance,

  Jose Luis

  > So, what is the precision of digits decimal point that Oracle can guarantee in floating-point numbers?

  This section on the built-in data types in Oracle.  Special attention to the NUMBER types.

  http://docs.Oracle.com/CD/E11882_01/server.112/e41084/sql_elements001.htm#SQLRF0021

• the number (use of space) data type

  Hi guys,.

  Just a query.

  I have a table with a data type of existing column - number (2).

  If I change it to number (8), it will take an extra space if the inserted values are from 2-digit?

  Any URL to reference?

  Thank you

  I have a table with a data type of existing column - number (2).

  If I change it to number (8), it will take an extra space if the inserted values are from 2-digit?

  Any URL to reference?

  No - the amount of space takes a NUMERIC value is based ONLY on the actual value you store.

  You can store 20 numbers that don't take no more space than would a number to two digits.

  Do not believe? Then just store values in a column NUMBER then question the values and the values of DUMP (myColumn) out and see how the actual duration and the internal representation. The value '12300000000000' will not take any space more than the value "123" since the number of SIGNIFICANT digits is the same.

  See my response in this thread of the year last for details and sample data.

  See the programmer's Guide of the OIC for details on how MANY values are stored internally.,.

  http://docs.Oracle.com/CD/E11882_01/AppDev.112/e10646/oci03typ.htm#sthref426

  Oracle database stores the values of the NUMBER data type in a variable length format. The first byte is the exponent and is followed by 1 to 20 mantissa bytes. The high bit of the exponent byte is the sign bit; It is defined for positive numbers, and it is cleared for negative numbers. The lower 7 bits represent the exponent, which is a number of base-100 with an offset of 65.

  To calculate the decimal exponent, add 65 to the exponent of base-100 and add another 128 if the number is positive. If the number is negative, do you the same thing, but eventually the bits are inverted. For example,-5 has an exponent of base-100 = 62 (0x3e). The decimal exponent is therefore (~ 0x3e)-128-65 = 0xc1-128-65 = 193-128-65 = 0.

  Each byte of the mantissa is a base-100 digit, in the 1,100 range. For positive numbers, the digit has 1 added to it. Thus, the number of mantissa for the value 5 is 6. For negative numbers, instead of adding 1, the digit is subtracted from 101. Thus, the digit of the mantissa of the number - 5 is 96 (101-5). Negative numbers have a byte containing 102 appended to the data bytes. However, negative numbers with 20 mantissa bytes do not have byte 102 end. Because the digits of the mantissa is stored in base 100, each byte can represent 2 decimal digits. The mantissa is normalized; zeros are not stored.

  Up to 20 data bytes can represent the mantissa. However, only 19 are guaranteed to be precise. 19 data bytes, each representing a base-100 digit, give a maximum precision of 38 digits for an Oracle NUMBER .

• data type for astromical algorithms

  Hello everyone.

  IM using Oracle 12 g of data.

  I have to face a new application where I have to develop a large number of astronomical algorithms, normally calculated them on the speed of the planets, astronomical positions and so forth.

  These algorithms usually have a large number of decimal place, between 10 and 20.

  So, Im trying to find the best types of data, because the accuracy is a must. And these are my doubts...

  Read the tutorial of Oracle it say that the number data type contains all the other subclass of numbers. For example, as I understand it, the integer data type is only one type of data number with some restrictions.

  My question is:

  If I declare a variable as number, this variable will have the same accuracy as a variable declared as binary_double? For example

  Declare

  x number; -Without defining the decimal scale in this way, it can accept the range possible decimal

  x BINARY_DOUBLE:-that class will have the same accuracy as the number of x?

  What is the best way to declare these vars?  by order of accuracy, storage and speed is not so important in this application.

  Regards to all the world and thanks in advance.

  Use the NUMBER unless you have a specific reason NOT to use it.

  See the built-in data types in the doc of the SQL language

  http://docs.Oracle.com/CD/B28359_01/server.111/b28286/sql_elements001.htm#i54330

  2

  NUMBER[ (p [, s]) ]

  Number having precision p and scale s . Precision p may vary from 1 to 38. Wide s can vary from-84 to 127. Precision and scale are in decimal digits. A NUMBER value requires 1 to 22 bytes.

• Largest number of digits for the NUMBER data type?

  What length more of a NUMBER that Oracle will "support"?

  The documentation says the following:

  Limitations of the data type says:

  «Can be represented in a comprehensive precision 38 digits»

  NUMBER of Data Types says:

  "Oracle guarantees portability of numbers with precision of up to 20 digits of base-100, which equals 39 or 40 decimal digits according to the position of the decimal point."

  I realize account that if I define a column as simply NUMBER, I can insert numbers with a size up to 126 characters.  However, Oracle seems to maintain only the first 40 digits in MOST cases.  The largest number of digits, it seems to allow is 40 before it begins to be replaced by 0.

  With numbers that have more than 40 figures, Oracle will sometimes replace all numbers according to the 38th numbers with a 0 and sometimes replace 0 after the digit 39th or 40th.

  Therefore, what is the largest number of digits, can be trusted to safely store Oracle?

  This is the code I used for this testing process.

  create the table max_num (num number);

  declare

  number of l_x;

  Start

  for x in 1.200

  loop

  l_x: = x;

  insert into max_num values (rpad (1, x, 1));

  end loop;

  exception

  while others then

  dbms_output.put_line ('STOP: ' | l_x);

  dbms_output.put_line (SQLERRM);

  end;

  /

  Select num, length (replace (num, 0)) of max_num;

  What length more of a NUMBER that Oracle will "support"?

  You have already given your own answer. If 'length': the maximum number of digits is written the doc gives you the answer:

  999... (38 9's) x 10 value maximum¹²⁵

  The 38/39/40, hereinafter referred to as the doc means "significant digits". This is why rounding or truncation occurs if you provide more significant digits of 38/39/40.

  Oracle stores the numbers internally in a binary format 21 bytes using a documented structure in the doc of the OIC in the section 'NUMBER'.

  http://docs.Oracle.com/CD/E11882_01/AppDev.112/e10646/oci03typ.htm#i423684

  Oracle database stores the values of the NUMBER data type in a variable length format. The first byte is the exponent and is followed by 1 to 20 mantissa bytes. The high bit of the exponent byte is the sign bit; It is defined for positive numbers, and it is cleared for negative numbers. The lower 7 bits represent the exponent, which is a number of base-100 with an offset of 65.

  This article from doc continues to show you how to convert the internal format to the real value.

  An additional byte of 1 is used for all types of data to store the length; That's why you often see docs saying numbers can take 22 bytes.

• Subtype and data types

  Hi all
  
  We can overload subprogrammes that setting differ in the number, order, and type of family data. But impossible to overload if the parameter differs only in the void type.
  
  If I see the standard package, I get
  subtype types BINARY_FLOAT is NUMBER;
  subtype BINARY_DOUBLE is NUMBER;
  
  But I am able to overload the two procedures, we have a parameter types binary_float and binary_double other. They are of subtype of the same data type. that is the number. So, how is it possible. Also, most of the subtypes have no difference. So, why are there these number of subtypes? Oracle defines subtypes as data type with some sort of constraint on the base data type, but is valid for all? for example, smallint, and decimal.
  
  type of NUMBER is NUMBER_BASE;
  subtype of FLOAT is NUMBER;
  ACTUAL subtype is FLOAT;
  subtype 'DOUBLE PRECISION' is FLOAT;
  INTEGER subtype is NUMBER (38.0).
  subtype INT is INTEGER;
  subtype SMALLINT is NUMBER (38.0).
  DECIMAL subtype is NUMBER (38.0).
  DIGITAL subtype is DECIMAL;
  DEC subtype is DECIMAL;

  >
  We can overload subprogrammes that setting differ in the number, order, and type of family data. But impossible to overload if the parameter differs only in the void type.

  If I see the standard package, I get
  subtype types BINARY_FLOAT is NUMBER;
  subtype BINARY_DOUBLE is NUMBER;

  But I am able to overload the two procedures, we have a parameter types binary_float and binary_double other. They are of subtype of the same data type. that is the number. So, how is it possible. Also, most of the subtypes have no difference. So, why are there these number of subtypes? Oracle defines subtypes as data type with some sort of constraint on the base data type, but is valid for all? for example, smallint, and decimal.

  type of NUMBER is NUMBER_BASE;
  subtype of FLOAT is NUMBER;
  ACTUAL subtype is FLOAT;
  subtype 'DOUBLE PRECISION' is FLOAT;
  INTEGER subtype is NUMBER (38.0).
  subtype INT is INTEGER;
  subtype SMALLINT is NUMBER (38.0).
  DECIMAL subtype is NUMBER (38.0).
  DIGITAL subtype is DECIMAL;
  DEC subtype is DECIMAL;
  >
  Some of those who, as a smallint, SQL ANSI data types. See table 2-6 ANSI Datatypes converted to Oracle data types in the doc of the SQL language
  http://docs.Oracle.com/CD/B28359_01/server.111/b28286/sql_elements001.htm

  Oracle allows the name of the ANSI data type to use and combines the name of the ANSI for the appropriate Oracle data type
  >
  ANSI SQL/DS and DB2 data types

  The SQL statements that create tables and clusters allows also ANSI data types and products IBM SQL/DS and DB2 data types. Oracle recognizes the ANSI or IBM data type name that differs from the Oracle database data type name. It converts the data type for the equivalent Oracle data type, stores the Oracle data type under the name of the column data type and stores the data in the column in the data type Oracle based on the conversions listed in the following tables.
  >
  But BINARY_FLOAT and BINARY_DOUBLE types are each defined as unique data types (see table 2-1 in this doc).

  Because they are unique Oracle data types, they can be overloaded.

• Oracle data type: on storage data as NUMBER of versus BINARY_DOUBLE?

  My request did a lot of scientific computing. The calculation uses double-precision variables in C and Java (both based on IEEE 754).
  
  If my goal is to store these variables in Oracle 11.2 G database between the calculations, such as the value read FROM the database is an exact match with the value that is used to write to the database, would you recommend store variables as an Oracle Data Type of NUMBER of DOUBLE BINARY?
  
  And, why?
  
  Here, I am only concerned about the accuracy (speed or storage).

  918175 wrote:
  Same double precision in Java and C have figures largely exceeded 17 (decimal) (I forgot the exact number of digits, but he has more than 30 years). I guess what I mean is accurate with respect to the first 17 digits (decimal), as the others have little or no effect in calculations.

  If you care just the 17 first decimal digits being the same after the persistence of data, it should not matter if you use NUMBER or BINARY_DOUBLE (or RAW). I prefer strongly NUMBER or BINARY_DOUBLE because then, you can use data outside your application in the future if someone wants to write a report.

  Justin