FirebirdSQL logo
 SQL Language StructureCommon Language Elements 

Approximate Floating-Point Data Types

Approximate floating-point values are stored in an IEEE 754 binary format that comprises sign, exponent and mantissa.Precision is dynamic, corresponding to the physical storage format of the value, which is exactly 4 bytes for the FLOAT type and 8 bytes for DOUBLE PRECISION.

Considering the peculiarities of storing floating-point numbers in a database, these data types are not recommended for storing monetary data.For the same reasons, columns with floating-point data are not recommended for use as keys or to have uniqueness constraints applied to them.

For testing data in columns with floating-point data types, expressions should check using a range, for instance, BETWEEN, rather than searching for exact matches.

When using these data types in expressions, extreme care is advised regarding the rounding of evaluation results.

FLOAT

Data Type Declaration Format
FLOAT [(bin_prec)]
Table 1. FLOAT Type Parameters
Parameter Description

bin_prec

Precision in binary digits, default is 24

1 - 24: 32-bit single precision25 - 53: 64-bit double precision

The FLOAT data type defaults to a 32-bit single precision floating-point type with an approximate precision of 7 decimal digits after the decimal point (24 binary digits).To ensure the safety of storage, rely on 6 decimal digits of precision.

The syntax FLOAT(bin_prec) behaves as follows:

  • 1 <= _bin_prec <= 23: 32-bit single precision (synonym for FLOAT)

  • 25 <= _bin_prec <= 53: 64-bit double precision (synonym for DOUBLE PRECISION)

The behaviour of FLOAT (without explicit precision) behaves as the SQL standard type [fblangref50-datatypes-real].

Note
Compatibility Notes

  • Firebird 3.0 and earlier supported FLOAT(dec_prec) where dec_prec was the approximate precision in decimal digits, with 0 <= dec_prec <= 7 mapped to 32-bit single precision and P > 7 mapped to 64-bit double precision.This syntax was never documented.

  • For bin_prec in FLOAT(bin_prec), the values 1 <= bin_prec <= 24 are all treated as bin_prec = 24, values 25 <= bin_prec <= 53 are all handled as bin_prec = 53.

  • Most Firebird tools will report FLOAT(1) — FLOAT(24) as FLOAT, and FLOAT(25) — FLOAT(53) as DOUBLE PRECISION.

REAL

Data Type Declaration Format
REAL

The data type REAL is a synonym for FLOAT, and is provided for syntax compatibility.When used to define a column or parameter, it’s indistinguishable from using FLOAT or FLOAT(1) — FLOAT(24).

Note
Compatibility Note

  • REAL has been available as a synonym for FLOAT since Firebird 1.0 and even earlier, but was never documented.

  • Most Firebird tools will report FLOAT instead of REAL.

DOUBLE PRECISION

Data Type Declaration Format
DOUBLE PRECISION

The DOUBLE PRECISION data type is stored with an approximate precision of 15 digits.

Note
Compatibility Notes

  • Firebird also has the — previously undocumented — synonyms for DOUBLE PRECISION: LONG FLOAT and LONG FLOAT(bin_prec), with 1 <= bin_prec <= 53.

    These non-standard type names are deprecated and may be removed in a future Firebird version.

  • Firebird 3.0 and earlier supported LONG FLOAT(dec_prec) where dec_prec was the approximate precision in decimal digits, where any value for dec_prec mapped to 64-bit double precision.

Decimal Floating-Point Types

Decimal floating-point values are stored in an IEEE 754 decimal format that comprises sign, exponent and coefficient.Contrary to the approximate floating-point data types, precision is either 16 or 34 decimal digits.

DECFLOAT

Data Type Declaration Format
DECFLOAT [(dec_prec)]
Table 1. DECFLOAT Type Parameters
Parameter Description

dec_prec

Precision in decimal digits, either 16 or 34.Default is 34.

DECFLOAT is a SQL standard-compliant numeric type that stores floating-point number precisely (decimal floating-point type), unlike FLOAT or DOUBLE PRECISION that provide a binary approximation of the purported precision.

The type is stored and transmitted as IEEE 754 standard types Decimal64 (DECFLOAT(16)) or Decimal128 (DECFLOAT(34)).

All intermediate calculations are performed with 34-digit values.

16-digit and 34-digit

The “16” and “34” refer to the maximum precision in Base-10 digits.See https://en/wikipedia.org/wiki/iEEE_754#Basic_and_interchange_formats for a comprehensive table.

Table 2. Range of Values
Type Maximum precision Minimum Exponent Maximum Exponent Smallest value Largest value

DECFLOAT(16)

16

-383

+384

1E-398

9.9..9E+384

DECFLOAT(34)

34

-6143

+6144

1E-6176

9.9..9E+6144

Observe that although the smallest exponent for DECFLOAT(16) is -383, the smallest value has an exponent of -398, but 15 fewer digits.And similar for DECFLOAT(34), smallest exponent is -6143, but the smallest value has an exponent of -6176, but 33 fewer digits.The reason is that precision was “sacrificed” to be able to store a smaller value.

This is a result of how the value is stored: as a decimal value of 16 or 34 digits and an exponent.For example, 1.234567890123456e-383 is stored as coefficient 1234567890123456 and exponent -398, while 1E-398 is stored as coefficient 1, exponent -398.

Behaviour of DECFLOAT Operations

The behaviour of DECFLOAT operations in a session, specifically rounding and error behaviour, can be configured using the SET DECFLOAT management statement, and the isc_dpb_decfloat_round and isc_dpb_decfloat_traps DPB items.

Length of DECFLOAT Literals

It is possible to express DECFLOAT(34) values in approximate numeric literals, but only for values with a mantissa of 20 or more digits, or an absolute exponent larger than 308.Scientific notation literals with fewer digits or a smaller absolute exponent are DOUBLE PRECISION literals.Exact numeric literals with 40 or more digits — actually 39 digits, when larger than the maximum INT128 value — are also handled as DECFLOAT(34).

Alternatively, use a string literal and explicitly cast to the desired DECFLOAT type.

The length of DECFLOAT literals cannot exceed 1024 characters.Scientific notation is required for greater values.For example, 0.0<1020 zeroes>11 cannot be used as a literal, but the equivalent in scientific notation, 1.1E-1022, is valid.Similarly, 10<1022 zeroes>0 can be presented as 1.0E1024.Literals with more than 34 significant digits are rounded using the DECFLOAT rounding mode of the session.

DECFLOAT and Functions
Use with Standard Functions

A number of standard scalar functions can be used with expressions and values of the DECFLOAT type.They are:

ABS

CEILING

EXP

FLOOR

LN

LOG

LOG10

POWER

SIGN

SQRT

The aggregate functions SUM, AVG, MAX and MIN work with DECFLOAT data, as do all the statistical aggregates (including but not limited to STDDEV or CORR).

Special Functions for DECFLOAT

Firebird supports four functions, designed to support DECFLOAT data specifically:

COMPARE_DECFLOAT

compares two DECFLOAT values to be equal, different or unordered

NORMALIZE_DECFLOAT

takes a single DECFLOAT argument and returns it in its simplest form

QUANTIZE

takes two DECFLOAT arguments and returns the first argument scaled using the second value as a pattern

TOTALORDER

performs an exact comparison on two DECFLOAT values

Detailed descriptions are available in the Special Functions for DECFLOAT section of the Built-in Scalar Functions chapter.

Fixed-Point Data Types

Fixed-point data types ensure the predictability of multiplication and division operations, making them the choice for storing monetary values.Firebird implements two fixed-point data types: NUMERIC and DECIMAL.

According to the SQL standard, both types limit the stored number to the declared scale (the number of digits after the decimal point).The standard defines different treatment of the precision for each type: precision for NUMERIC columns is exactly “as declared”, while DECIMAL columns accepts numbers whose precision is at least equal to what was declared.

Note

The behaviour of both NUMERIC and DECIMAL in Firebird is like the SQL-standard DECIMAL;the precision is at least equal to what was declared.

For instance, NUMERIC(4, 2) defines a number consisting altogether of four digits, including two digits after the decimal point;that is, it can have up to two digits before the point[1]] and no more than two digits after the point.If the number 3.1415 is written to a column with this data type definition, the value of 3.14 will be saved in the NUMERIC(4, 2) column.

The form of declaration for fixed-point data, for instance, NUMERIC(p, s), is common to both types.The s argument in this template is scale.Understanding the mechanism for storing and retrieving fixed-point data should help to visualise why: for storage, the number is multiplied by 10s (10 to the power of s), converting it to an integer;when read, the integer is converted back by multiplying by 10-s (or, dividing by 10s).

The method of storing fixed-point data in the database depends on several factors: declared precision, database dialect, declaration type.

Table 1. Method of Physical Storage for Fixed-Point Numbers
Precision Data type Dialect 1 Dialect 3

1 - 4

NUMERIC

SMALLINT

SMALLINT

1 - 4

DECIMAL

INTEGER

INTEGER

5 - 9

NUMERIC or DECIMAL

INTEGER

INTEGER

10 - 18

NUMERIC or DECIMAL

DOUBLE PRECISION

BIGINT

19 - 38

NUMERIC or DECIMAL

INT128

INT128
Note

Numerics with precision less than 19 digits use SMALLINT, INTEGER, BIGINT or DOUBLE PRECISION as the base data type, depending on the number of digits and SQL dialect.When precision is between 19 and 38 digits an INT128 is used as the base data type, and the actual precision is always extended to the full 38 digits.

For complex calculations, those digits are cast internally to DECFLOAT(34).The result of various mathematical operations, such as LOG(), EXP() and so on, and aggregate functions using a high precision numeric argument, will be DECFLOAT(34).
1. in practice, the actual range is determined by the backing type, for NUMERIC(4, s) that is SMALLINT, which means it can store [-327.68, 327.67

NUMERIC

Data Type Declaration Format
NUMERIC [(precision [, scale])]
Table 1. NUMERIC Type Parameters
Parameter Description

precision

Precision, between 1 and 38.Defaults to 9.

scale

Scale, between 0 and precision.Defaults to 0.
Storage Examples

Further to the explanation above, Firebird will store NUMERIC data according the declared precision and scale.Some more examples are:

NUMERIC(4) stored as      SMALLINT (exact data)
NUMERIC(4,2)              SMALLINT (data * 102)
NUMERIC(10,4) (Dialect 1) DOUBLE PRECISION
              (Dialect 3) BIGINT (data * 104)
NUMERIC(38,6)             INT128 (data * 106)
Caution

Always keep in mind that the storage format depends on the precision.For instance, you define the column type as NUMERIC(2,2) presuming that its range of values will be -0.99…​0.99.However, the actual range of values for the column will be -327.68…​327.67, which is due to storing the NUMERIC(2,2) data type in the SMALLINT format.In storage, the NUMERIC(4,2), NUMERIC(3,2) and NUMERIC(2,2) data types are the same.This means that if you need to store data in a column with the NUMERIC(2,2) data type and limit the range to -0.99…​0.99, you will have to create a CHECK constraint for it.

DECIMAL

Data Type Declaration Format
{ DECIMAL | DEC } [(precision [, scale])]
Table 1. DECIMAL Type Parameters
Parameter Description

precision

Precision, between 1 and 38.Defaults to 9.

scale

Scale, between 0 and precision.Defaults to 0.
Storage Examples

The storage format in the database for DECIMAL is similar to NUMERIC, with some differences that are easier to observe with the help of some more examples:

DECIMAL(4) stored as      INTEGER (exact data)
DECIMAL(4,2)              INTEGER (data * 102)
DECIMAL(10,4) (Dialect 1) DOUBLE PRECISION
              (Dialect 3) BIGINT (data * 104)
DECIMAL(38,6)             INT128 (data * 106)

Data Types for Dates and Times

The DATE, TIME and TIMESTAMP data types are used to work with data containing dates and times.

Time zone support is available using the types TIME WITH TIME ZONE and TIMESTAMP WITH TIME ZONE.In this language reference, we’ll use TIME and TIMESTAMP to refer both to the specific types without time zone — TIME [WITHOUT TIME ZONE] and TIMESTAMP [WITHOUT TIME ZONE] — and aspects of both the without time zone and with time zone types, which one we mean is usually clear from the context.

Important

The data types TIME WITHOUT TIME ZONE, TIMESTAMP WITHOUT TIME ZONE and DATE are defined to use the session time zone when converting from or to a TIME WITH TIME ZONE or TIMESTAMP WITH TIME ZONE.TIME and TIMESTAMP are synonymous to their respective WITHOUT TIME ZONE data types.

Dialect 3 supports all the five types, while Dialect 1 has only DATE.The DATE type in Dialect 3 is “date-only”, whereas the Dialect 1 DATE type stores both date and time-of-day, equivalent to TIMESTAMP in Dialect 3.Dialect 1 has no “date-only” type.

Note

Dialect 1 DATE data can be defined alternatively as TIMESTAMP and this is recommended for new definitions in Dialect 1 databases.
Fractions of Seconds

If fractions of seconds are stored in date and time data types, Firebird stores them to ten-thousandths of a second (100 microseconds or deci-milliseconds).If a lower granularity is preferred, the fraction can be specified explicitly as thousandths, hundredths or tenths of a second, or second, in Dialect 3 databases of ODS 11 or higher.

Note
Some Useful Knowledge about Sub-second Precision

The time-part of a TIME or TIMESTAMP is a 32-bit integer, with room for deci-milliseconds (or 100 microseconds) precision and time values are stored as the number of deci-milliseconds elapsed since midnight.The actual precision of values stored in or read from time(stamp) functions and variables is:

  • CURRENT_TIME and LOCALTIME default to seconds precision and can be specified up to milliseconds precision with CURRENT_TIME({0|1|2|3}) or LOCALTIME({0|1|2|3})

  • CURRENT_TIMESTAMP and LOCALTIMESTAMP default to milliseconds precision.Precision from seconds to milliseconds can be specified with CURRENT_TIMESTAMP({0|1|2|3}) or LOCALTIMESTAMP({0|1|2|3})

  • Literal 'NOW' defaults to milliseconds precision

  • Functions DATEADD() and DATEDIFF() support up to deci-milliseconds precision with MILLISECOND

  • The EXTRACT() function returns up to deci-milliseconds precision with the SECOND and MILLISECOND arguments

  • the ‘+’ and ‘-’ operators work with deci-milliseconds precision.

Deci-milliseconds precision is not supported by all drivers and access components.The best assumption to make from all this is that, although Firebird stores TIME and the TIMESTAMP time-part values as the number of deci-milliseconds (10-4 seconds) elapsed since midnight, the actual precision could vary from seconds to milliseconds.
Storage of Time Zone Types

The time zone types are stored as values at UTC (offset 0), using the structure of TIME or TIMESTAMP + two extra bytes for time zone information (either an offset in minutes, or the id of a named time zone).Storing as UTC allows Firebird to index and compare two values in different time zones.

Storing at UTC has some caveats:

  • When you use named zones, and the time zone rules for that zone change, the UTC time stays the same, but the local time in the named zone may change.

  • For TIME WITH TIME ZONE, calculating a time zone offset for a named zone to get the local time in the zone applies the rules valid at the 1st of January 2020 to ensure a stable value.This may result in unexpected or confusing results.

  • When the rules of a named time zone changes, a value in the affected date range may no longer match the intended value if the actual offset in that named zone changes.

DATE

Syntax
DATE

The DATE data type in Dialect 3 stores only date without time.The available range for storing data is from January 01, 1 to December 31, 9999.

In Dialect 1, DATE is an alias for [fblangref50-datatypes-timestamp].Dialect 1 has no “date-only” type.

Tip

In Dialect 1, date literals without a time part, as well as casts of date mnemonics 'TODAY', 'YESTERDAY' and 'TOMORROW' automatically get a zero time part.

If you need to store a Dialect 1 timestamp literal with an explicit zero time-part, the engine will accept a literal like '2016-12-25 00:00:00.0000'.However, '2016-12-25' would have the same effect, with fewer keystrokes!
Storage of Dates

Internally, Firebird stores dates in a 32-bit integer as a Modified Julian Date, or the number of days since 1858-11-17.An additional restriction is imposed, limiting valid dates to the range from 0001-01-01 AD to 9999-12-31 AD.

TIME

Syntax
TIME [{ WITHOUT | WITH } TIME ZONE]

For a bare TIME, WITHOUT TIME ZONE is assumed.

The TIME data type is available in Dialect 3 only.It stores the time of day within the range from 00:00:00.0000 to 23:59:59.9999.

If you need to get the time-part from DATE in Dialect 1, you can use the EXTRACT function.

Examples Using EXTRACT()
EXTRACT (HOUR FROM DATE_FIELD)
EXTRACT (MINUTE FROM DATE_FIELD)
EXTRACT (SECOND FROM DATE_FIELD)

See also the EXTRACT() function in the chapter entitled Built-in Scalar Functions.

TIME [WITHOUT TIME ZONE]

The TIME (or synonym TIME WITHOUT TIME ZONE) represents a time without time zone information.

TIME WITH TIME ZONE

The TIME WITH TIME ZONE represents a time with time zone information (either an offset or a named zone).

Firebird uses the ICU implementation of the IANA Time Zone Database for named zones.

Examples Using EXTRACT()
EXTRACT (TIMEZONE_HOUR FROM TIME_TZ_FIELD)
EXTRACT (TIMEZONE_MINUTE FROM TIME_TZ_FIELD)

TIMESTAMP

Syntax
TIMESTAMP [{ WITHOUT | WITH } TIME ZONE]

For a bare TIMESTAMP, WITHOUT TIME ZONE is assumed.

The TIMESTAMP data type is available in Dialect 3 and Dialect 1.It comprises two 32-bit integers — a date-part and a time-part — to form a structure that stores both date and time-of-day.In Dialect 1, [fblangref50-datatypes-date] is an alias for TIMESTAMP.

The EXTRACT function works equally well with TIMESTAMP as with the Dialect 1 DATE type.

TIMESTAMP [WITHOUT TIME ZONE]

The TIMESTAMP (or synonym TIMESTAMP WITHOUT TIME ZONE) represents a time and date without time zone information.

TIMESTAMP WITH TIME ZONE

The TIMESTAMP WITH TIME ZONE represents a time with time zone information (either an offset or a named zone).

Session Time Zone

As the name implies, the session time zone can be different for each database attachment.It can be set explicitly in the DPB with the item isc_dpb_session_time_zone;otherwise, by default, it uses the same time zone as the operating system of the Firebird server process.This default can be overridden in firebird.conf, setting DefaultTimeZone.

Note

Drivers may apply different defaults, for example specifying the client time zone as the default session time zone.Check your driver documentation for details.

Subsequently, the time zone can be changed to a given time zone using a SET TIME ZONE statement or reset to its original value with SET TIME ZONE LOCAL or ALTER SESSION RESET.

Time Zone Format

A time zone is specified as a string, either a time zone region (for example, 'America/Sao_Paulo') or a displacement from GMT in hours:minutes (for example, '-03:00').

A time/timestamp with time zone is considered equal to another time/timestamp with time zone if their conversions to UTC are equivalent.For example, time '10:00 -02:00' and time '09:00 -03:00' are equivalent, since both are the same as time '12:00 GMT'.

Important

The same equivalence applies in UNIQUE constraints and for sorting purposes.

Operations Using Date and Time Values

The method of storing date and time values makes it possible to involve them as operands in some arithmetic operations.In storage, a date value or date-part of a timestamp is represented as the number of days elapsed since “date zero” — November 17, 1858 — whilst a time value or the time-part of a timestamp is represented as the number of deci-milliseconds (100 microseconds) since midnight.

An example is to subtract an earlier date, time or timestamp from a later one, resulting in an interval of time, in days and fractions of days.

Table 1. Arithmetic Operations for Date and Time Data Types
Operand 1 Operation Operand 2 Result

DATE

+

TIME

TIMESTAMP

DATE

+

TIME WITH TIME ZONE

TIMESTAMP WITH TIME ZONE

DATE

+

Numeric value n

DATE increased by n whole days.Fractional values are rounded to the nearest integer

TIME

+

DATE

TIMESTAMP

TIME WITH TIME ZONE

+

DATE

TIMESTAMP WITH TIME ZONE

TIME

+

Numeric value n

TIME increased by n seconds.The fractional part is taken into account

TIME WITH TIME ZONE

+

Numeric value n

TIME WITH TIME ZONE increased by n seconds.The fractional part is taken into account

TIMESTAMP

+

Numeric value n

TIMESTAMP, where the date will advance by the number of days and part of a day represented by number n — so “+ 2.75” will push the date forward by 2 days and 18 hours

TIMESTAMP WITH TIME ZONE

+

Numeric value n

TIMESTAMP WITH TIME ZONE, where the date will advance by the number of days and part of a day represented by number n — so “+ 2.75” will push the date forward by 2 days and 18 hours

DATE

-

DATE

Number of days elapsed, within the range DECIMAL(9, 0)

DATE

-

Numeric value n

DATE reduced by n whole days.Fractional values are rounded to the nearest integer

TIME

-

TIME

Number of seconds elapsed, within the range DECIMAL(9, 4)

TIME

-

TIME WITH TIME ZONE

The without time zone value is converted to WITH TIME ZONE in the current session time zone.Number of seconds elapsed between the UTC values, within the range DECIMAL(9, 4).Also applies when swapping types.

TIME WITH TIME ZONE

-

TIME WITH TIME ZONE

Number of seconds elapsed between the UTC values, within the range DECIMAL(9, 4)

TIME

-

Numeric value n

TIME reduced by n seconds.The fractional part is taken into account

TIMESTAMP

-

TIMESTAMP

Number of days and part-day, within the range DECIMAL(18, 9)

TIMESTAMP

-

TIMESTAMP WITH TIME ZONE

The without time zone value is converted to WITH TIME ZONE in the current session time zone.Number of days and part-day between UTC values, within the range DECIMAL(18, 9).Also applies when swapping types.

TIMESTAMP WITH TIME ZONE

-

TIMESTAMP WITH TIME ZONE

Number of days and part-day between UTC values, within the range DECIMAL(18, 9)

TIMESTAMP

-

Numeric value n

TIMESTAMP where the date will decrease by the number of days and part of a day represented by number n — so “- 2.25” will decrease the date by 2 days and 6 hours
Note
Notes

In Dialect 1, the DATE type is considered an alias of TIMESTAMP.
See also

DATEADD, DATEDIFF

Supplemental Time Zone Features

Firebird provides a number of features to discover time zone information.

Virtual table RDB$TIME_ZONES

A virtual table listing time zones supported in the engine.

See also RDB$TIME_ZONES in System Tables.

Package RDB$TIME_ZONE_UTIL

A package of time zone utility functions and procedures.

See also RDB$TIME_ZONE_UTIL in System Packages.

Updating the Time Zone Database

Time zones are often changed: of course, when it happens, it is desirable to update the time zone database as soon as possible.

Firebird stores WITH TIME ZONE values translated to UTC time.Suppose a value is created with one time zone database, and a later update of that database changes the information in the range of our stored value.When that value is read, it will be returned as different to the value that was stored initially.

Firebird uses the IANA time zone database through the ICU library.The ICU library included in the Firebird kit (Windows), or installed in a POSIX operating system, can sometimes have an outdated time zone database.

An updated database can be found on this page on the FirebirdSQL GitHub.Filename le.zip stands for little-endian and is the necessary file for most computer architectures (Intel/AMD compatible x86 or x64), while be.zip stands for big-endian architectures and is necessary mostly for RISC computer architectures.The content of the zip file must be extracted in the /tzdata sub-directory of the Firebird installation, overwriting existing *.res files belonging to the database.

Note

/tzdata is the default directory where Firebird looks for the time zone database.It can be overridden with the ICU_TIMEZONE_FILES_DIR environment variable.

Character Data Types

For working with character data, Firebird has the fixed-length CHAR and the variable-length VARCHAR data types.The maximum size of text data stored in these data types is 32,767 bytes for CHAR and 32,765 bytes for VARCHAR.The maximum number of characters that will fit within these limits depends on the character set used for the data under consideration.The collation does not affect this maximum, although it may affect the maximum size of any index that involves the column.

If no character set is explicitly specified when defining a character object, the default character set of the database — at time of defining the object — will be used.If the database does not have a default character set defined, the object gets the character set NONE.

Unicode

Most current development tools support Unicode, implemented in Firebird with the character sets UTF8 and UNICODE_FSS.UTF8 comes with collations for many languages.UNICODE_FSS is more limited and was previously used mainly by Firebird internally for storing metadata.Keep in mind that one UTF8 character occupies up to 4 bytes, thus limiting the size of CHAR fields to 8,191 characters (32,767/4).

Note

The actual “bytes per character” value depends on the range the character belongs to.Non-accented Latin letters occupy 1 byte, Cyrillic letters from the WIN1251 encoding occupy 2 bytes in UTF8, characters from other encodings may occupy up to 4 bytes.

The UTF8 character set implemented in Firebird supports the latest version of the Unicode standard, thus recommending its use for international databases.

Client Character Set

While working with strings, it is essential to keep the character set of the client connection in mind.If there is a mismatch between the character sets of the stored data and that of the client connection, the output results for string columns are automatically re-encoded, both when data are sent from the client to the server and when they are sent back from the server to the client.For example, if the database was created in the WIN1251 encoding but KOI8R or UTF8 is specified in the client’s connection parameters, the mismatch will be transparent.

Special Character Sets

Character set NONE

The character set NONE is a special character set in Firebird.It can be characterized such that each byte is a part of a string, but the string is stored in the system without any clues about what constitutes any character: character encoding, collation, case, etc. are simply unknown.It is the responsibility of the client application to deal with the data and provide the means to interpret the string of bytes in some way that is meaningful to the application and the human user.

Character set OCTETS

Data in OCTETS encoding are treated as bytes that may not be interpreted as characters.OCTETS provides a way to store binary data.The database engine has no concept of what it is meant to do with a string of bytes in OCTETS, other than store and retrieve it.Again, the client side is responsible for validating the data, presenting them in formats that are meaningful to the application and its users and handling any exceptions arising from decoding and encoding them.CHAR and VARCHAR with character set OCTETS have synonyms BINARY and VARBINARY.

Collation

Each character set has a default collation (COLLATE) that specifies the collation order (or, collation sequence, or collating sequence).Usually, it provides nothing more than ordering based on the numeric code of the characters and a basic mapping of upper- and lower-case characters.If some behaviour is needed for strings that is not provided by the default collation and a suitable alternative collation is supported for that character set, a COLLATE collation clause can be specified in the column definition.

A COLLATE collation clause can be applied in other contexts besides the column definition.For comparison operations, it can be added in the WHERE clause of a SELECT statement.If output needs to be sorted in a special alphabetic sequence, or case-insensitively, and the appropriate collation exists, then a COLLATE clause can be included with the ORDER BY clause when rows are being sorted on a character field and with the GROUP BY clause in case of grouping operations.

Case-Insensitive Searching

For a case-insensitive search, the UPPER function could be used to convert both the search argument and the searched strings to upper-case before attempting a match:

...
where upper(name) = upper(:flt_name)

For strings in a character set that has a case-insensitive collation available, you can apply the collation, to compare the search argument and the searched strings directly.For example, using the WIN1251 character set, the collation PXW_CYRL is case-insensitive for this purpose:

...
WHERE FIRST_NAME COLLATE PXW_CYRL >= :FLT_NAME
...
ORDER BY NAME COLLATE PXW_CYRL
See also

CONTAINING

UTF8 Collations

The following table shows the possible collations for the UTF8 character set.

Table 1. Collations for Character Set UTF8
Collation Characteristics

UCS_BASIC

Collation works according to the position of the character in the table (binary).

UNICODE

Collation works according to the UCA algorithm (Unicode Collation Algorithm) (alphabetical).

UTF8

The default, binary collation, identical to UCS_BASIC, which was added for SQL compatibility

UNICODE_CI

Case-insensitive collation, works without taking character case into account.

UNICODE_CI_AI

Case-insensitive, accent-insensitive collation, works alphabetically without taking character case or accents into account.
Example

An example of collation for the UTF8 character set without taking into account the case or accentuation of characters (similar to COLLATE PXW_CYRL in the earlier example).

...
ORDER BY NAME COLLATE UNICODE_CI_AI

Character Indexes

The maximum length for an index key equals one quarter of the page size, i.e. from 1,024 — for page size 4,096 — to 8,192 bytes — for page size 32,768.The maximum length of an indexed string is 9 bytes less than that quarter-page limit.

Calculating Maximum Length of an Indexed String Field

The following formula calculates the maximum length of an indexed string (in characters):

max_char_length = FLOOR((page_size / 4 - 9) / N)

where N is the number of bytes per character in the character set.

The table below shows the maximum length of an indexed string (in characters), according to page size and character set, calculated using this formula.

Table 1. Maximum Index Lengths by Page Size and Character Size

Page Size

Bytes per character

1

2

3

4

6

4,096

1,015

507

338

253

169

8,192

2,039

1,019

679

509

339

16,384

4,087

2,043

1,362

1,021

681

32,768

8,183

4,091

2,727

2,045

1,363
Note

With case-insensitive collations (“_CI”), one character in the index key will occupy not 4, but 6 (six) bytes, so the maximum key length for a page of — for example — 4,096 bytes, will be 169 characters.
See also

CREATE DATABASE, Collation, SELECT, WHERE, GROUP BY, ORDER BY

Character Types in Detail

BINARY

Data Type Declaration Format
BINARY [(length)]
Table 1. BINARY Type Parameters
Parameter Description

length

Length in bytes between 1 and 32,767;defaults to 1.

BINARY is a fixed-length binary data type, and is an SQL standard-compliant alias for CHAR(length) CHARACTER SET OCTETS.Values shorter than the declared length are padded with NUL (0x00) up to the declared length.

Note

Some tools may report the type as CHAR CHARACTER SET OCTETS instead of BINARY.
See also

[fblangref50-datatypes-chartypes-char], [fblangref50-datatypes-chartypes-varbinary]

CHAR

Data Type Declaration Format
{CHAR | CHARACTER} [(length)]
  [CHARACTER SET <set>] [COLLATE <name>]
Table 1. CHAR Type Parameters
Parameter Description

length

Length in characters, defaults to 1.A valid length is from 1 to the maximum number of characters that can be accommodated within 32,767 bytes.

set

Character set name

name

Collation name

CHAR is a fixed-length character data type.Values shorter than the declared length are padded with spaces up to the declared length.The pad character does not have to be a space (0x20): it depends on the character set.For example, the pad character for the OCTETS character set is NUL (0x00).

Fixed-length character data can be used to store codes whose length is standard and has a definite “width”.An example of such a code is an EAN13 barcode — 13 characters, all filled.

Note

  • CHAR CHARACTER SET OCTETS has the alias BINARY

  • Formally, the COLLATE clause is not part of the data type declaration, and its position depends on the syntax of the statement.
See also

[fblangref50-datatypes-chartypes-binary], [fblangref50-datatypes-chartypes-varchar]

VARBINARY

Data Type Declaration Format
{VARBINARY | BINARY VARYING} (length)
Table 1. VARBINARY Type Parameters
Parameter Description

length

Length in bytes between 1 and 32,765

VARBINARY is a variable-length binary type, and is an SQL standard-compliant alias for VARCHAR(length) CHARACTER SET OCTETS.

Note

Some tools may report the type as VARCHAR CHARACTER SET OCTETS instead of VARBINARY.
See also

[fblangref50-datatypes-chartypes-varchar], [fblangref50-datatypes-chartypes-binary]

VARCHAR

Data Type Declaration Format
{VARCHAR | {CHAR | CHARACTER} VARYING} (length)
  [CHARACTER SET <set>] [COLLATE <name>]
Table 1. VARCHAR Type Parameters
Parameter Description

length

Length in characters.A valid length is from 1 to the maximum number of characters that can be accommodated within 32,765 bytes.

set

Character set name

name

Collation name

VARCHAR is a variable-length character data type, up to a maximum of 32,765 bytes.The stored structure is equal to the actual size of the data plus 2 bytes to record the length of the data.

All characters that are sent from the client application to the database are considered meaningful, including leading and trailing spaces.

Note

  • VARCHAR CHARACTER SET OCTETS has the alias VARBINARY

  • Formally, the COLLATE clause is not part of the data type declaration, and its position depends on the syntax of the statement.
See also

[fblangref50-datatypes-chartypes-varbinary], [fblangref50-datatypes-chartypes-char]

NCHAR

Data Type Declaration Format
{NCHAR | NATIONAL {CHAR | CHARACTER}} [(length)]

NCHAR is a fixed-length character data type with the ISO8859_1 character set.In all other respects it is the same as CHAR.

Note

If no length is specified, it is taken to be 1.

A similar data type is available for the variable-length string type: NATIONAL {CHAR | CHARACTER} VARYING.

See also

[fblangref50-datatypes-chartypes-char], [fblangref50-datatypes-chartypes-varchar]

Boolean Data Type

BOOLEAN

Data Type Declaration Format
BOOLEAN

The SQL-compliant BOOLEAN data type (8 bits) comprises the distinct truth values TRUE and FALSE.Unless prohibited by a NOT NULL constraint, the BOOLEAN data type also supports the truth value UNKNOWN as the null value.The specification does not make a distinction between the NULL value of this data type, and the truth value UNKNOWN that is the result of an SQL predicate, search condition, or Boolean value expression: they may be used interchangeably to mean the same thing.

As with many programming languages, the SQL BOOLEAN values can be tested with implicit truth values.For example, field1 OR field2 and NOT field1 are valid expressions.

The IS Operator

Predicates can use the operator Boolean IS [NOT] for matching.For example, field1 IS FALSE, or field1 IS NOT TRUE.

Note

  • Equivalence operators (“=”, “!=”, “<>” and so on) are valid in all comparisons.

BOOLEAN Examples

  1. Inserting and selecting

    CREATE TABLE TBOOL (ID INT, BVAL BOOLEAN);
COMMIT;

INSERT INTO TBOOL VALUES (1, TRUE);
INSERT INTO TBOOL VALUES (2, 2 = 4);
INSERT INTO TBOOL VALUES (3, NULL = 1);
COMMIT;

SELECT * FROM TBOOL;
          ID    BVAL
============ =======
           1 <true>
           2 <false>
           3 <null>

  2. Test for TRUE value

    SELECT * FROM TBOOL WHERE BVAL;
          ID    BVAL
============ =======
           1 <true>

  3. Test for FALSE value

    SELECT * FROM TBOOL WHERE BVAL IS FALSE;
          ID    BVAL
============ =======
           2 <false>

  4. Test for UNKNOWN value

    SELECT * FROM TBOOL WHERE BVAL IS UNKNOWN;
          ID    BVAL
============ =======
           3 <null>

  5. Boolean values in SELECT list

    SELECT ID, BVAL, BVAL AND ID < 2
  FROM TBOOL;
          ID    BVAL
============ ======= =======
           1 <true>  <true>
           2 <false> <false>
           3 <null>  <false>

  6. PSQL declaration with start value

    DECLARE VARIABLE VAR1 BOOLEAN = TRUE;

  7. Valid syntax, but as with a comparison with NULL, will never return any record

SELECT * FROM TBOOL WHERE BVAL = UNKNOWN;
SELECT * FROM TBOOL WHERE BVAL <> UNKNOWN;
Use of Boolean Against Other Data Types

Although BOOLEAN is not inherently convertible to any other data type, the strings 'true' and 'false' (case-insensitive) will be implicitly cast to BOOLEAN in value expressions.For example:

if (true > 'false') then ...

The value 'false' is converted to BOOLEAN.Any attempt to use the Boolean operators AND, NOT, OR and IS will fail.NOT 'False', for example, is invalid.

A BOOLEAN can be explicitly converted to and from string with CAST.UNKNOWN is not available for any form of casting.

Note
Other Notes

  • The type is represented in the API with the FB_BOOLEAN type and FB_TRUE and FB_FALSE constants.

  • The value TRUE is greater than the value FALSE.

Binary Data Types

Note

The types [fblangref50-datatypes-chartypes-binary] and [fblangref50-datatypes-chartypes-varbinary] are covered earlier in section [fblangref50-datatypes-chartypes].

BLOBs (Binary Large Objects) are complex structures used to store text and binary data of an undefined length, often very large.

Syntax
BLOB [SUB_TYPE <subtype>]
  [SEGMENT SIZE <segment size>]
  [CHARACTER SET <character set>]
  [COLLATE <collation name>]

If the SUB_TYPE and CHARACTER SET clauses are absent, then subtype BINARY (or 0) is used.If the SUB_TYPE clause is absent and the CHARACTER SET clause is present, then subtype TEXT (or 1) is used.

Shortened syntax
BLOB (<segment size>)
BLOB (<segment size>, <subtype>)
BLOB (, <subtype>)
Note

Formally, the COLLATE clause is not part of the data type declaration, and its position depends on the syntax of the statement.
Segment Size

Specifying the BLOB segment size is a throwback to times past, when applications for working with BLOB data were written in C (Embedded SQL) with the help of the gpre pre-compiler.Nowadays, it is effectively irrelevant.The segment size for BLOB data is determined by the client side and is usually larger than the data page size, in any case.

BLOB Subtypes

The optional SUB_TYPE parameter specifies the nature of data written to the column.Firebird provides two pre-defined subtypes for storing user data:

Subtype 0: BINARY

If a subtype is not specified, the specification is assumed to be for untyped data and the default SUB_TYPE BINARY (or SUB_TYPE 0) is applied.This is the subtype to specify when the data are any form of binary file or stream: images, audio, word-processor files, PDFs and so on.

Subtype 1: TEXT

Subtype 1 has an alias, TEXT, which can be used in declarations and definitions.For instance, BLOB SUB_TYPE TEXT (or BLOB SUB_TYPE 1).It is a specialized subtype used to store plain text data that is too large to fit into a string type.A CHARACTER SET may be specified, if the field is to store text with a different encoding to that specified for the database.A COLLATE clause is also supported.

Specifying CHARACTER SET without specifying a SUB_TYPE implies SUB_TYPE TEXT.

Custom Subtypes

It is also possible to add custom data subtypes, for which the range of enumeration from -1 to -32,768 is reserved.Custom subtypes enumerated with positive numbers are not allowed, as the Firebird engine uses the numbers from 2-upward for some internal subtypes in metadata.Custom subtype aliases can be inserted into the RDB$TYPES table by users with the system privilege CREATE_USER_TYPES.

BLOB Specifics

Size

The maximum size of a BLOB field depends on the page size of the database, whether the blob value is created as a stream blob or a segmented blob, and if segmented, the actual segment sizes used when populating the blob.For most built-in functions, the maximum size of a BLOB field is 4 GB, or data beyond the 4 GB limit is not addressable.For a page size of 4 KB (4096 bytes) the maximum size is slightly less than 4 GB.

Operations and Expressions

Text BLOBs of any length and any character set — including multi-byte — can be operands for practically any statement or internal functions.The following operators are fully supported:

=

(assignment)

=, <>, <, <=, >, >=

(comparison)

||

(concatenation)

BETWEEN,

IS [NOT] DISTINCT FROM,

IN,

ANY | SOME,

ALL

 

As an efficient alternative to concatenation, you can also use BLOB_APPEND() or the functions and procedures of system package RDB$BLOB_UTIL.

Partial support:

  • An error occurs with these if the search argument is larger than or equal to 32 KB:

    STARTING [WITH],

    LIKE,

    CONTAINING

  • Aggregation clauses work not on the contents of the field itself, but on the BLOB ID.Aside from that, there are some quirks:

    SELECT DISTINCT

    returns several NULL values by mistake if they are present

    ORDER BY

     — 

    GROUP BY

    concatenates the same strings if they are adjacent to each other, but does not do it if they are remote from each other
BLOB Storage

  • By default, a regular record is created for each BLOB, and it is stored on a data page that is allocated for it.If the entire BLOB fits onto this page, it is called a level 0 BLOB.The number of this special record is stored in the table record and occupies 8 bytes.

  • If a BLOB does not fit onto one data page, its contents are put onto separate pages allocated exclusively to it (blob pages), while the numbers of these pages are stored into the BLOB record.This is a level 1 BLOB.

  • If the array of page numbers containing the BLOB data does not fit onto a data page, the array is put on separate blob pages, while the numbers of these pages are put into the BLOB record.This is a level 2 BLOB.

  • Levels higher than 2 are not supported.

See also

FILTER, DECLARE FILTER, BLOB_APPEND(), RDB$BLOB_UTIL

Array Types

Note

Firebird does not offer much in the way of language or tools for working with the contents of arrays, and there are no plans to improve this.This limits the usefulness and accessibility of array types.Therefore, the general advice is: do not use arrays.

The support of arrays in the Firebird DBMS is a departure from the traditional relational model.Supporting arrays in the DBMS could make it easier to solve some data-processing tasks involving large sets of similar data.

Arrays in Firebird are stored in BLOB of a specialized type.Arrays can be one-dimensional and multi-dimensional and of any data type except BLOB and ARRAY.

Example
CREATE TABLE SAMPLE_ARR (
  ID INTEGER NOT NULL PRIMARY KEY,
  ARR_INT INTEGER [4]
);

This example will create a table with a field of the array type consisting of four integers.The subscripts of this array are from 1 to 4.

Specifying Explicit Boundaries for Dimensions

By default, dimensions are 1-based — subscripts are numbered from 1.To specify explicit upper and lower bounds of the subscript values, use the following syntax:

'[' <lower>:<upper> ']'

Adding More Dimensions

A new dimension is added using a comma in the syntax.In this example we create a table with a two-dimensional array, with the lower bound of subscripts in both dimensions starting from zero:

CREATE TABLE SAMPLE_ARR2 (
  ID INTEGER NOT NULL PRIMARY KEY,
  ARR_INT INTEGER [0:3, 0:3]
);

The database employee.fdb, found in the ../examples/empbuild directory of any Firebird distribution package, contains a sample stored procedure showing some simple work with arrays:

PSQL Source for SHOW_LANGS, a procedure involving an array

CREATE OR ALTER PROCEDURE SHOW_LANGS (
  CODE VARCHAR(5),
  GRADE SMALLINT,
  CTY VARCHAR(15))
RETURNS (LANGUAGES VARCHAR(15))
AS
  DECLARE VARIABLE I INTEGER;
BEGIN
  I = 1;
  WHILE (I <= 5) DO
  BEGIN
    SELECT LANGUAGE_REQ[:I]
    FROM JOB
    WHERE (JOB_CODE = :CODE)
      AND (JOB_GRADE = :GRADE)
      AND (JOB_COUNTRY = :CTY)
      AND (LANGUAGE_REQ IS NOT NULL))
    INTO :LANGUAGES;

    IF (LANGUAGES = '') THEN
    /* PRINTS 'NULL' INSTEAD OF BLANKS */
      LANGUAGES = 'NULL';
    I = I +1;
    SUSPEND;
  END
END

If the features described are enough for your tasks, you might consider using arrays in your projects.Currently, no improvements are planned to enhance support for arrays in Firebird.

Special Data Types

“Special” data types …​

SQL_NULL Data Type

The SQL_NULL type holds no data, but only a state: NULL or NOT NULL.It is not available as a data type for declaring table fields, PSQL variables or parameter descriptions.This data type exists to support the use of untyped parameters in expressions involving the IS NULL predicate.

An evaluation problem occurs when optional filters are used to write queries of the following type:

WHERE col1 = :param1 OR :param1 IS NULL

After processing, at the API level, the query will look like this:

WHERE col1 = ? OR ? IS NULL

This is a case where the developer writes an SQL query and considers :param1 as though it were a variable that they can refer to twice.However, at the API level, the query contains two separate and independent parameters.The server cannot determine the type of the second parameter since it comes in association with IS NULL.

The SQL_NULL data type solves this problem.Whenever the engine encounters an “? IS NULL” predicate in a query, it assigns the SQL_NULL type to the parameter, which will indicate that parameter is only about “nullness” and the data type or the value need not be addressed.

The following example demonstrates its use in practice.It assumes two named parameters — say, :size and :colour — which might, for example, get values from on-screen text fields or drop-down lists.Each named parameter corresponds with two positional parameters in the query.

SELECT
  SH.SIZE, SH.COLOUR, SH.PRICE
FROM SHIRTS SH
WHERE (SH.SIZE = ? OR ? IS NULL)
  AND (SH.COLOUR = ? OR ? IS NULL)

Explaining what happens here assumes the reader is familiar with the Firebird API and the passing of parameters in XSQLVAR structures — what happens under the surface will not be of interest to those who are not writing drivers or applications that communicate using the “naked” API.

The application passes the parameterized query to the server in the usual positional ?-form.Pairs of “identical” parameters cannot be merged into one, so for the two optional filters in the example, four positional parameters are needed: one for each ? in our example.

After the call to isc_dsql_describe_bind(), the SQLTYPE of the second and fourth parameters will be set to SQL_NULL.Firebird has no knowledge of their special relation with the first and third parameters: that responsibility lies entirely on the application side.

Once the values for size and colour have been set (or left unset) by the user, and the query is about to be executed, each pair of XSQLVARs must be filled as follows:

User has supplied a value

First parameter (value compare): set *sqldata to the supplied value and *sqlind to 0 (for NOT NULL)

Second parameter (NULL test): set sqldata to null (null pointer, not SQL NULL) and *sqlind to 0 (for NOT NULL)

User has left the field blank

Both parameters: set sqldata to null (null pointer, not SQL NULL) and *sqlind to -1 (indicating NULL)

In other words: The value compare parameter is always set as usual.The SQL_NULL parameter is set the same, except that sqldata remains null at all times.