(inlong-website) branch master updated: [INLONG-1173][Doc] Add function description of Transform SDK (#1174)

[luchunliang]

This is an automated email from the ASF dual-hosted git repository.

luchunliang pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/inlong-website.git


The following commit(s) were added to refs/heads/master by this push:
     new a4796cec56f [INLONG-1173][Doc] Add function description of Transform 
SDK (#1174)
a4796cec56f is described below

commit a4796cec56f8d57ddfb437dbc9a82770fbceb556
Author: ChunLiang Lu <luchunli...@apache.org>
AuthorDate: Mon Nov 17 19:43:55 2025 +0800

    [INLONG-1173][Doc] Add function description of Transform SDK (#1174)
    
    * [INLONG-1173][Doc] Add function description of Transform SDK
    
    * [INLONG-1173][Doc] Add function description of Transform SDK
    
    * function description change to function specification.
---
 docs/modules/transform/function_specification.md   | 217 +++++++++++++++++++
 .../modules/transform/function_specification.md    | 229 +++++++++++++++++++++
 2 files changed, 446 insertions(+)

diff --git a/docs/modules/transform/function_specification.md 
b/docs/modules/transform/function_specification.md
new file mode 100644
index 00000000000..5229e683c92
--- /dev/null
+++ b/docs/modules/transform/function_specification.md
@@ -0,0 +1,217 @@
+---
+title: Function Specification
+sidebar_position: 3
+---
+
+# Arithmetic
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| abs(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
absolute value of 'numeric'.<br> | abs(2) = 2<br>abs(-4.25) = 4.25<br> |
+| acos(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc cosine of 'numeric'.<br> | acos(1) = 0.0<br>acos(0) = 
1.5707963267948966<br>acos(-1) = 3.141592653589793<br> |
+| acosd(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc cosine of 'numeric' in units of degrees.<br> | acosd(1) = 
0.0<br>acosd(0) = 90.0<br>acosd(-1) = 180.0<br> |
+| asin(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc sine of 'numeric' in units of degrees.<br> | asin(0.5) = 
0.5235987755982989<br>asin(0) = 0.0<br>asin(-0.5) = -0.5235987755982989<br> |
+| asind(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc sine of 'numeric' in units of degrees.<br> | asind(0.5) = 
30.000000000000004<br>asind(0) = 0.0<br>asind(-0.5) = -30.000000000000004<br> |
+| atan(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc tangent of 'numeric'.<br> | atan(1) = 0.7853981633974483<br>atan(0) = 
0.0<br>atan(-1) = -0.7853981633974483<br> |
+| atan2(Numeric numeric)<br> | Return "" if 'numericx' or 'numericy' is 
NULL;<br>Return the arc tangent of a coordinate ('numericx', 'numericy').<br> | 
atan2(1, 1) = 0.7853981633974483<br>atan2(1, 0) = 
1.5707963267948966<br>atan2(0, -1) = 3.141592653589793<br> |
+| atan2d(Numeric numericx, Numeric numericy)<br> | - Return "" if 'numericx' 
or 'numericy' is NULL;<br>- Return inverse tangent of 'numericy'/'numericx', 
result in degrees.<br> | atan2d(1, 1) = 45.0<br>atan2d(1, 0) = 
90.0<br>atan2d(0, -1) = 180.0<br> |
+| atand(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the arc tangent of 'numeric' in units of degrees.<br> | atand(1) = 
45.0<br>atand(0) = 0.0<br>atand(-1) = -45.0<br> |
+| bin(Integer integer)<br> | - Return "" if 'integer' is NULL;<br>- Return a 
string representation of 'integer' in binary format.<br> | bin(4) = 100<br> |
+| cbrt(Numeric numeric)<br> | - Return NULL if 'numeric' is NULL;<br>- Return 
cube root.<br> | cbrt(5) = 1.7099759466766968<br>cbrt(27) = 3.0<br> |
+| ceil(Numeric numeric)<br> | - Return NULL if 'numeric' is NULL;<br>- Return 
the smallest number that is greater than or equal to 'numeric'.<br> | 
ceil(-5.67) = -5.0<br>ceil(1.23) = 2.0<br> |
+| chr(Integer integer)<br> | - Return NULL if 'integer' is NULL;<br>- Return 
the ASCII character having the binary equivalent to 'integer'.<br> | chr(97) = 
"a"<br>chr(353) = "a"<br> |
+| cos(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
cosine of 'numeric'.<br> | cos(0) = 1.0<br> |
+| cosd(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the cosine of 'numeric' in units of degrees.<br> | cosd(15) = 
0.9659258262890683<br> |
+| cosh(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the hyperbolic cosine of 'numeric'.<br> | cosh(1) = 
1.543080634815244<br>cosh(0) = 1.0<br>cosh(-1) = 1.543080634815244<br> |
+| cot(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
cotangent of the 'numeric' (in radians).<br> | cot(1) = 
0.6420926159343306<br>cot(0.5) = 1.830487721712452<br>cot(-1) = 
-0.6420926159343306<br> |
+| cotd(Numeric numeric)<br> | - Return "" if numeric is NULL;<br>- Return the 
cotangent of numeric in units of degrees.<br> | cotd(0) = ""<br>cotd(45) = 
1.0000000000000002<br>cotd(-1) = -57.28996163075943<br> |
+| e()<br> | - Return the mathematical constant.<br> | e() = 
2.718281828459045<br> |
+| erf(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
error function value of the input value.<br> | erf(3) = 0.9999779095030014<br> |
+| erfc(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
complementary error (1 - erf('numeric'), without loss of precision for large 
inputs).<br> | erfc(1) = 0.15729920705028488<br> |
+| exp(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return e 
raised to the power of 'numeric'.<br> | exp(2) = 7.38905609893065<br> |
+| factorial(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- 
Return the factorial of a non-negative.<br> | factorial(5) = 
120<br>factorial(0) = 1<br> |
+| fibonacci(Numeric numeric)<br> | - Return NULL if 'numeric' is NULL;<br>- 
Returns the nth Fibonacci number<br> | fibonacci(0) = 0<br>fibonacci(1) = 
1<br>fibonacci(2) = 1<br>fibonacci(3) = 2<br>fibonacci(4) = 3<br> |
+| floor(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the largest number that is less than or equal to 'numeric'.<br> | floor(1.23) = 
1.0<br>floor(-5.67) = -6.0<br> |
+| gcd(Numeric numeric1,Numeric numeric2)<br> | - Return "" if any parameter is 
NULL;<br>- Return 0 if both inputs are zero;<br>- Return greatest common 
divisor (the largest positive number that divides both inputs with no 
remainder).<br> | gcd(3.141,3.846) = 0.003<br> |
+| greatest(Numeric value1 [,Numeric value2,...])<br> | - Return "" if any 
parameter is NULL;<br>- Return the greatest value of the list of arguments.<br> 
| greatest(3.141592653589793, 3, greatest(4, 1)) = 4<br> |
+| hex(String dataStr)<br> | - Return "" if dataStr is NULL;<br>- Return the 
string obtained by converting the dataStr to hexadecimal if 'dataStr' can be 
parsed into numeric;<br>- Return the string obtained by converting the ASCII 
code corresponding to each character to hexadecimal otherwise.<br> | hex(1007) 
= "3EF"<br>hex('abc') = "616263"<br> |
+| lcm(Numeric numeric1,Numeric numeric2)<br> | - Return "" if any parameter is 
NULL;<br>- Return 0 if either input is zero;<br>- Return least common multiple 
(the smallest strictly positive number that is an integral multiple of both 
inputs).<br> | lcm(6,3) = 6<br>lcm(3.141,3.846) = 4026.762<br> |
+| least(Numeric value1 [, Numeric value2, ...])<br> | - Return "" if any 
parameter is NULL;<br>- Return the least value of the list of arguments.<br> | 
least(3.14, least(7, 2, 1)) = 1<br> |
+| ln(Numeric numeric)<br> | - Return "" if numeric is NULL;<br>- Return the 
natural logarithm (base e) of numeric.<br> | ln(10) = 2.302585092994046<br> |
+| log(Numeric numeric1 [, Numeric numeric2])<br> | - Return the natural 
logarithm of 'numeric1' when called with one argument;<br>- Return the 
logarithm of 'numeric2' to the base 'numeric1' when called with two 
arguments.<br> | log(1) = 0.0<br>log(2,8) = 3.0<br> |
+| log10(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the base 10 logarithm of 'numeric'.<br> | log10(1000) = 3.0<br> |
+| log2(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the base 2 logarithm of 'numeric'.<br> | log2(32) = 5<br> |
+| min_scale(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- 
Return minimum scale (number of fractional decimal digits) needed to represent 
the supplied value precisely.<br> | min_scale(3.1415000000000) = 4<br> |
+| mod(Numeric numeric1, Numeric numeric2)<br> | - Return "" if any parameter 
is null;<br>- Return the remainder of 'numeric1' divided by 'numeric2'.<br> | 
mod(3,2) = 1<br>mod(-3.1415926,100) = -3.1415926<br> |
+| num_nonnulls([Expr expr1, Expr expr2, ...])<br> | - Return the number of 
non-null arguments.<br> | num_nonnulls(5, 3, null, null) = 2<br> |
+| num_nulls([Expr expr1, Expr expr2, ...])<br> | - Return the number of null 
arguments.<br> | num_nulls(5, null, null, null) = 3<br> |
+| pi()<br> | - Return the mathematical constant PI.<br> | pi() = 
3.141592653589793<br> |
+| power(Numeric numeric1, Numeric numeric2)<br> | - Return "" if any parameter 
is NULL.<br>- Return 'numeric1'.power('numeric2').<br> | power(4,0.5) = 2.0<br> 
|
+| radians(Numeric x)<br> | - Return "" if 'x' is NULL;<br>- Return radians of 
'x', Convert degrees to radians.<br> | radians(18.97) = 0.33108895910332425<br> 
|
+| radix_convert(Numeric numeric)<br> | - Return "" if any of its arguments are 
NULL;<br>- Return the result of converting 'numeric' from 'from_base' to 
'to_base'.<br>Note: abs(base) between [2,36].'from_base' is a negative number, 
'numeric' is regarded as a signed number. Otherwise, 'numeric' is treated as 
unsigned. This function works with 64-bit precision.<br> | 
radix_convert('6E',18,8) = 172<br> |
+| rand(Integer seed)<br> | - Return a pseudo-random double precision value in 
the range [0.0, 1.0) if seed is NULL;<br>- Return a pseudo-random double 
precision value in the range [0.0, 1.0) with an initial 'seed' of Integer.<br> 
| rand(1)<br>rand()<br> |
+| rand_integer(Integer INT1, [Integer INT2])<br> | - Return a pseudorandom 
integer value in the range [0, 'INT1') if 'INT2' is NULL;<br>- Return a 
pseudorandom integer value in the range [0, 'INT1') with an initial seed 
'INT2'.<br>Note: Two RAND_INTEGER functions will return idential sequences of 
numbers if they have the same initial seed and bound.<br> | 
rand_integer(10)<br>rand_integer(88, 89)<br> |
+| round(String str)<br> | - Return "" if 'x' is NULL;<br>- Return the nearest 
integer to 'x', with optional parameter 'y' indicating the number of decimal 
places to be rounded.<br> | round(3.5) = 4<br>round(3.14159265358979323846,10) 
= 3.1415926536<br> |
+| scale(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
scale of the argument (the number of decimal digits in the fractional 
part).<br> | scale(3.1415000000000) = 13<br> |
+| sign(Numeric x)<br> | - Return "" if 'x' is NULL;<br>- Return -1 if 'x' is a 
negative number;<br>- Return 0 if 'x' is equal to 0;<br>- Return 1 if 'x' is a 
positive number.<br> | sign(-3.5) = -1<br> |
+| sin(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
sine of 'numeric'.<br> | sin(0) = 0.0<br> |
+| sind(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the sine of 'numeric' in units of degrees.<br> | sind(15) = 
0.25881904510252074<br> |
+| sinh(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the hyperbolic sine of 'numeric'.<br> | sinh(1) = 1.1752011936438014<br> |
+| sqrt(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the square root of 'numeric'.<br> | sqrt(9) = 3.0<br> |
+| tan(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return the 
tangent of 'numeric'.<br> | tan(1) = 1.5574077246549023<br> |
+| tand(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the tangent of 'numeric' in units of degrees.<br> | tand(15) = 
0.2679491924311227<br> |
+| tanh(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- Return 
the hyperbolic tangent of 'numeric'.<br> | tanh(1) = 0.7615941559557649<br> |
+| trim_scale(Numeric numeric)<br> | - Return "" if 'numeric' is NULL;<br>- 
Return the result of reducing the proportion of values (decimal places) by 
removing trailing zeros.<br> | trim_scale(3.1415000000000) = 3.1415<br> |
+| truncate(Numeric numeric, Integer cnt)<br>trunc(Numeric numeric, Integer 
cnt)<br> | - Return "" if any parameter is NULL;<br>- Return the result of 
truncating 'numeric' values to 'cnt' decimal places.<br> | truncate(42.324, 2) 
= 42.32<br>trunc(42.324, -1) = 40<br> |
+# Collection
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| array(String value1 [,String value2, ....])<br> | - Return an array created 
from a list of values ('value1', 'value2', ....).<br> | array('he',7,'xxd') = 
[he, 7, xxd]<br>array(array('he',5),'xxd') = return [[he, 5], 
xxd]<br>array(array('he',5),array('','')) = return [[he, 5], [, ]]<br> |
+| array_append(Array array, Object element)<br> | - Return "" if either 
argument is NULL;<br>- Return the result of appends an element to the end of 
the 'array'.<br> | array_append(array('he',7,'xxd'), 'cloud') = [he, 7, xxd, 
cloud]<br> |
+| array_concat([Array array1, Array array2, ...])<br> | - Return "" if any 
input array is NULL;<br>- Return an array that is the result of concatenating 
at least one array.<br> | array_concat(array('he',7),array('xxd', 'cloud')) = 
[he, 7, xxd, cloud]<br> |
+| array_contains(Array array, Object needle)<br> | - Return "" if 'array' is 
NULL;<br>- Return whether the given element exists in 'array'.<br>Note: 
Checking for null elements in the array is supported.<br> | 
array_contains(array('he',7,'xxd'), 'cloud') = 
false<br>array_contains(array('he',-1,''),'') = true<br> |
+| array_distinct(Array array)<br> | - Return "" if 'array' is NULL;<br>- 
Return an array with unique elements.<br> | array_distinct(array('he',-1,'he')) 
= [he, -1]<br> |
+| array_except(Array array1, Array array2)<br> | - Return "" if any parameter 
is null;<br>- Return an ARRAY that contains the elements from 'array1' that are 
not in 'array2', without duplicates;<br>- Return an empty ARRAY if no elements 
remain after excluding the elements in 'array2' from 'array1'.<br> | 
array_except(array('he',7,'xxd'),array('he')) = [7, 
xxd]<br>array_except(array('he',7,'xxd'),array('cloud')) = [he, 7, xxd]<br> |
+| array_intersect(Array array1, Array array2)<br> | - Return "" if any 
parameter is null;<br>- Return an ARRAY that contains the elements from 
'array1' that are not in 'array2', without duplicates;<br>- Return an empty 
ARRAY if no elements remain after excluding the elements in 'array2' from 
'array1'.<br> | array_intersect(array('he',7,'xxd'),array('he')) = 
[he]<br>array_intersect(array('he',7,'xxd'),array('cloud')) = []<br> |
+| array_join(Array array, String delimiter[, String nullReplacement])<br> | - 
Return "" if any parameter is null;<br>- Return a string indicating that the 
elements in the given 'array' are concatenated using a 'delimiter', and empty 
array elements are filled with 'nullReplacement'.<br>- Note: If nullReplacement 
is not specified, null elements in the array will be omitted from the resulting 
string.<br> | array_join(array('he',7,'xxd'),'~') = 
he~7~xxd<br>array_join(array('he',3,''),'~','oo [...]
+| array_max(Array array)<br> | - Return "" if any parameter is null;<br>- 
Return the maximum value from the 'array'.<br> | array_max(array(4,3,56)) = 
56<br> |
+| array_min(Array array)<br> | - Return "" if any parameter is null;<br>- 
Return the minimum value from the 'array'.<br> | array_min(array(4,3,56)) = 
3<br> |
+| array_position(Array array,Object element)<br> | - Return "" if any 
parameter is null;<br>- Return the position of the first occurrence of 
'element' in the given 'array' as int (starts from 1);<br>- Return 0 if the 
given value could not be found in the 'array'.<br> | 
array_position(array('he',7,'xxd'),'he') = 
1<br>array_position(array('he',7,''),'_') = 0<br> |
+| array_prepend(Array array,Object element)<br> | - Return "" if 'array' is 
null;<br>- Return the result of appending an element to the beginning of the 
array.<br> | array_prepend(array(4,3),3) = [3, 4, 3]<br> |
+| array_remove(Array array,Object element)<br> | - Return "" if 'array' is 
null;<br>- Return the result of removing all elements that equal to 'element' 
from 'array'.<br> | array_remove(array('he',7,'xxd'),'he') = [7, xxd]<br> |
+| array_reverse(Array array)<br> | - Return "" if 'array' is null;<br>- Return 
an array in reverse order.<br> | array_reverse(array('he',7,'xxd')) = [xxd, 7, 
he]<br> |
+| array_slice(Array array, Integer start_offset[, Integer end_offset])<br> | - 
Return "" if 'array' or 'start_offset' is null;<br>- Return a subarray of the 
input 'array' between 'start_offset' and 'end_offset' inclusive;<br>- Return an 
empty array if 'start_offset' is after 'end_offset' or both are out of 'array' 
bounds.<br>Note: If 'end_offset' is omitted then this offset is treated as the 
length of the 'array'. Positive values are counted from the beginning of the 
array while negative [...]
+| array_sort(Array array[,Boolean ascending_order[,Boolean null_first]])<br> | 
- Return "" if 'array' is null;<br>- Return the array in sorted order.<br>Note: 
The function sorts an array, defaulting to ascending order with NULLs at the 
start when only the array is input. Specifying 'ascending_order' as true orders 
the array in ascending with NULLs first, and setting it to false orders it in 
descending with NULLs last. Independently, 'null_first' as true moves NULLs to 
the beginning, and  [...]
+| array_union(Array array1[, Array array2, ...])<br> | - Return "" if any 
input array is NULL;<br>- Return an array that is the result of concatenating 
at least one array.<br> | array_concat(array('he',7),array('xxd', 'cloud')) = 
[he, 7, xxd, cloud]<br> |
+| cardinality(Object input)<br> | - Return "" if the input is NULL;<br>- 
Return the number of elements in array if the input is array;<br>- Return the 
number of entries in map if the input is map.<br> | 
cardinality(array('he',7,'xxd')) = 3<br>cardinality(map('he',7,'xxd',3)) = 
2<br> |
+| element(Array array)<br> | - Return "" if 'array' is empty or NULL;<br>- 
Return the sole element of 'array' (whose cardinality should be one).<br>Note: 
Throws an exception if array has more than one element.<br> | 
element(array('he')) = he<br> |
+| find_in_set(String str,String strList)<br> | - Return "" if either argument 
is NULL;<br>- Return 0 if 'str' is not in 'strList' or if 'strList' is the 
empty string;<br>- Return a value in the range of 1 to N if the string 'str' is 
in the string list 'strList' consisting of N substrings.<br>Note: strList is a 
string composed of substrings separated by ',' characters. This function does 
not work properly if the first argument contains a comma (,) character.<br> | 
FIND_IN_SET('b','a,b,b,c [...]
+| map([String value1, String value2, ...])<br> | - Return "" if the number of 
parameters is not even;<br>- Return a map created from a list of key-value 
pairs ((value1, value2), ... ).<br> | Map('he',7,'xxd') = 
null<br>Map('he',1,'xxd','cloud') = {he=1, 
xxd=cloud}<br>Map('xxd','cloud',map(1,2),map(3,'apple')) = {xxd=cloud, 
{1=2}={3=apple}}<br> |
+| map_entries(Map map)<br> | - Return "" if 'map' is NULL;<br>- Return an 
array of all entries in the given 'map'.<br> | 
map_entries(Map('he',1,'xxd','cloud')) = [he=1, 
xxd=cloud]<br>map_entries(Map(1,2,'cloud','xxd')) = [xxd=cloud, 1=2]<br> |
+| map_from_arrays(Array array_of_keys, Array array_of_values)<br> | - Return 
"" if any parameter is NULL;<br>- Return a map created from an arrays of keys 
and values.<br> | map_from_arrays(array('he', 'xxd'),array(1, 3)) = {he=1, 
xxd=3}<br>map_from_arrays(array('xxd', array('cloud')),array(1, array(2))) = 
{1=xxd, [2]=[cloud]}<br> |
+| map_keys(Map map)<br> | - Return "" if 'map' is NULL;<br>- Return the keys 
of the 'map' as array. No order guaranteed.<br> | 
map_keys(Map('he',1,'xxd','cloud')) = [he, 
xxd]<br>map_keys(Map('xxd','cloud',map(1,2),map(3,'apple'))) = [xxd, {1=2}]<br> 
|
+| map_union([Map map1, Map map2, ...])<br> | - Return "" if any of maps is 
null;<br>- Return a map created by merging at least one map.These maps should 
have a common map type, the following map will overwrite the previous one.<br> 
| map_union(map('he', 1),map('xxd', 3)) = {he=1, xxd=3}<br>map_union(map('he', 
1),map('he', 3)) = {he=3}<br> |
+| map_values(Map map)<br> | - Return "" if 'map' is null;<br>- Return the 
values of the 'map' as array. No order guaranteed.<br> | 
map_values(Map('he',1,'xxd','cloud')) = [1, 
cloud]<br>map_values(Map('xxd','cloud',map(1,2),map(3,'apple'))) = [cloud, 
{3=apple}]<br> |
+# Compression
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| compress(String string_to_compress [, String compress_type])<br> | - Return 
"" if 'string_to_compress' is NULL;<br>- Return "" if 'string_to_compress' is 
"";<br>- Return the result as a binary string.<br>Note: This function supports 
three compression algorithms: deflater, gzip, and zip. 'compress_type' defaults 
to defer. In addition, in order to output the compressed results in the form of 
strings, this method uses the ISO_8859_1 character set.<br> | 
length(compress(replicate(string1,1 [...]
+| uncompress(String string_to_uncompress, String compress_type)<br> | - Return 
"" if 'string_to_uncompress' is NULL;<br>- Return "" if 'string_to_uncompress' 
is "";<br>- Return the result as a string.<br>Note: This function supports 
three compression algorithms: deflater, gzip and zip. 'compress_type' defaults 
to defer.<br> | uncompress(compress('inlong')) = "inlong"<br> |
+# Condition
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| coalesce(String value1 [, String value2, ...])<br> | - Return "" If all 
arguments are NULL or "";<br>- Return the first argument that is not NULL or 
"".<br>Note: The return type is the least restrictive, common type of all of 
its arguments. The return type is nullable if all arguments are nullable as 
well.<br> | coalesce('', 'SQL', 'hh') = "SQL"<br> |
+| if(Expr expr1,Expr expr2,Expr expr3)<br> | - Return 'expr2' if 'expr1' 
returns true;<br>- Return 'expr3' otherwise.<br> | if(1 = 1,true,false) = 
true<br> |
+| ifnull(Expr expr1,Expr expr2)<br>if_null(Expr expr1,Expr expr2)<br> | - 
Return 'expr1' if 'expr1' returns not NULL;<br>- Return 'expr2' otherwise.<br> 
| ifnull(null, 3) = 3<br>ifnull(6,'YES') = 6<br> |
+| isnull(Expr expr)<br> | - Return true if 'expr' is NULL;<br>- Return false 
otherwise.<br> | isnull(5 + 3) = false<br>isnull(5 / 0) = true<br> |
+| nullif(Expr expr1, Expr expr2)<br>null_if(Expr expr1, Expr expr2)<br> | - 
Return "" if 'expr1' = 'expr2' is true;<br>- Return 'expr1' otherwise.<br> | 
nullif(5, 3) = 5<br> |
+# Encryption
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| dayu_decrypt<br> |  |  |
+| decode(Binary binary,String charsetStr)<br> | - Return "" if any parameter 
is NULL;<br>- Return the result of encoding 'binary' using the character set 
specified by 'charsetStr'.<br>Note: 'charsetStr' is one of ('US-ASCII', 
'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16').<br> | 
decode(encode('Hello','UTF-8'),'UTF-8') = "Hello"<br> |
+| encode(String strInfo,String charsetStr)<br> | - Return "" if any parameter 
is NULL;<br>- Return the result of encoding 'strInfo' using the character set 
specified by 'charsetStr'.<br>- Note: 'charsetStr' is one of ('US-ASCII', 
'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16').<br> | 
decode(encode('Hello','UTF-8'),'UTF-8') = "Hello"<br> |
+| from_base64(String base64Str)<br> | - Return "" if 'base64Str' is NULL;<br>- 
Return the base64-decoded result from 'base64Str'.<br> | 
from_base64('QXBhY2hlIEluTG9uZw==') = "Apache InLong"<br> |
+| md5(String string)<br> | - Return "" if the 'string' is NULL;<br>- Return 
the MD5 hash value of 'string' in the form of a 32-bit hexadecimal digit 
string.<br> | md5("") = "d41d8cd98f00b204e9800998ecf8427e"<br>md5("1") = 
"c4ca4238a0b923820dcc509a6f75849b"<br> |
+| sha(String str)<br> | - Return "" if 'str' is NULL;<br>- Return a string of 
40 hexadecimal digits (the SHA-1 160 bit).<br> | sha("5") = 
"ac3478d69a3c81fa62e60f5c3696165a4e5e6ac4"<br> |
+| sha2(String str, Integer hash_length)<br> | - Return "" if either argument 
is NULL or the 'hash_length' is not one of (224,256,384,512);<br>- Return scale 
of the argument (the number of decimal digits in the fractional part).<br> | 
sha2("5",224) = "b51d18b551043c1f145f22dbde6f8531faeaf68c54ed9dd79ce24d17"<br> |
+| to_base64(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the 
base64-encoded result from 'str'.<br> | to_base64('app-fun') = 
"YXBwLWZ1bg=="<br> |
+# Json
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| json_array_append(String json_doc, String path1, String val1[,String path2, 
String val2, ...])<br> | - Return "" if any argument is NULL;<br>- Return the 
result of appends values to the end of the indicated arrays within a JSON 
document.<br> | json_array_append(["a", ["b", "c"], "d"],$[0],2,$[1],3) = 
[["a","2"],["b","c","3"],"d"]<br> |
+| json_array_insert(String json_doc, String path1, String val1[, String path2, 
String val2, ...] )<br> | - Return "" if any argument is NULL;<br>- Return the 
'json_doc' inserted into the array.<br>Note: If multiple groups of parameters 
are passed in, the parameter subscripts of the latter groups need to be based 
on the document subscripts after the previous group of parameters are 
updated.<br> | json_array_append(["a", {"b": [1, 2]}, [3, 4]], $[1], x) = 
["a","x",{"b":[1,2]},[3,4]]<br> |
+| json_arrays([String value1, String value2, ...])<br> | - Return a JSON array 
string constructed from a list of values.<br> | JSON_ARRAYS() = 
[]<br>JSON_ARRAYS(1, '2') = [1,"2"]<br>JSON_ARRAYS(JSON_ARRAY(1)) = [[1]]<br> |
+| json_exists(String json_doc, String path)<br> | - Return true if 'json_doc' 
satisfies a given 'path' search criterion;<br>- Return false otherwise.<br> | 
JSON_EXISTS('{"a": true}', '$.a') = true<br>JSON_EXISTS('{"a": true}', '$.b') = 
false<br>JSON_EXISTS('{"a": [{ "b": 1 }]}', '$.a[0].b') = true<br> |
+| json_insert(String json_doc, String path1, String val1[, String path2, 
String val2, ...] )<br> | - Return "" if any argument is NULL or the 'json_doc' 
argument is not a valid JSON document or any path argument is not a valid path 
expression or contains a * or ** wildcard.;<br>- Return the result of inserting 
data into 'json_doc'.<br> | json_insert({"a": {"b": [1, 2]}, "c": [3, 4]}, 
$.c[1][1], "2", "$.c[1][1][5]", "1") = 
{"a":{"b":[1,2]},"c":[3,[4,["2","1"]]]}<br> |
+| json_query(String json_doc, String path)<br> | - Return "" if any parameter 
is NULL;<br>- Return the string parsed from the 'path' in 'json_doc'.<br> | 
json_query({\"people\": [{\"name\": \"Alice\"}, {\"name\": \"Bob\"}]}, 
$.people) = [{"name":"Alice"},{"name":"Bob"}]<br>json_query({\"list\": [null, 
{\"name\": \"John\"}]}, $.list[1].name) = John<br> |
+| json_quote(String data)<br>json_string(String data)<br> | - Return "" if 
data is NULL;<br>- Return a valid JSON string converted from a string 
(JSON_QUOTE) or any type of data (JSON_STRING).<br>Note: JSON_QUOTE will escape 
interior quote and special characters (’"’, ‘', ‘/’, ‘b’, ‘f’, ’n’, ‘r’, 
’t’)<br> | json_quote('Column1\tColumn2) = 
\"Column1\\tColumn2\"<br>json_string(true) = "true"<br> |
+| json_remove(String json_doc, String path1[, String path2, ...])<br> | - 
Return "" if any argument is NULL or the 'json_doc' argument is not a valid 
JSON document or any path argument is not a valid path expression or is $ or 
contains a * or ** wildcard;<br>- Return the result of removing data from 
'json_doc'.<br> | 
json_remove("{\"name\":\"Charlie\",\"hobbies\":[[\"swimming1\",\"swimming2\"], 
\"reading\",\"coding\"]}","$.age") = 
{"hobbies":[["swimming2"],"coding"],"name":"Charlie"}<br> |
+| json_replace(String json_doc, String path1, String val1[, String path2, 
String val2, ...] )<br> | - Return "" if any argument is NULL or the 'json_doc' 
argument is not a valid JSON document or any path argument is not a valid path 
expression or contains a * or ** wildcard;<br>- Return the result of replacing 
existing values in 'json_doc'.<br> | json_replace("{ \"a\": 1, \"b\": [2, 3]}", 
"$.a", 10, "$.c", "[true, false]") = {"a": 10, "b": [2, 3]}<br> |
+| json_set(String json_doc, String path1, String val1[, String path2, String 
val2, ...] )<br> | - Return "" if any argument is NULL or the 'json_doc' 
argument is not a valid JSON document or any path argument is not a valid path 
expression or contains a * or ** wildcard;<br>- Return the result of inserting 
or updating data in 'json_doc'.<br> | 
json_set({\"name\":\"Alice\"},"$.name","inlong") = {"name":"inlong"}<br> |
+| json_unquote(String data)<br> | - Return "" if data is NULL;<br>- Return the 
'data' unmodified if the value does not start and end with double quotes or if 
it starts and ends with double quotes but is not a valid JSON string 
literal.<br>Note: JSON_UNQUOTE will unescapes escaped special characters ('"', 
'', '/', 'b', 'f', 'n', 'r', 't')<br> | json_unquote('Hello, World!') = "Hello, 
World!"<br>json_unquote('Complex string with / and \\') = "Complex string with 
/ and \\"<br> |
+| json_value(String json_doc,String path)<br> | - Return "" if any parameter 
is NULL;<br>- Return the scalar extracted from JSON string ('json_doc') based 
on 'path'.<br> | json_value({"a": 1}, $.a) = 1<br>json_value({\"person\": 
{\"name\": \"Alice\" ,\"age\": 30}}, $.person.name) = Alice<br> |
+# String
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| ascii(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the 
numeric value of the first character of 'str'.<br> | ascii('abc') = 
97<br>ascii('A') = 65<br>ascii(null) = ""<br> |
+| bit_length(String str,[String charset])<br> | - Return "" if the 'str' is 
NULL;<br>- Return number of bits in 'str'.<br>Note: Charset is aligned with the 
JVM by default.<br> | bit_length("hello world") = 88<br>bit_length("hello 
你好","utf-8") = 96<br> |
+| char_length(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the 
character length of 'str'.<br> | char_length('hello world') = 11<br>case2: 
char_length('应龙') = 2<br> |
+| concat(String string1 [, String string2, ...])<br> | - Return NULL If any 
parameter is NULL;<br>- Return the string of the connection ('string1', 
'string2',...).<br> | CONCAT("AA", "BB", "CC") = "AABBCC"<br> |
+| concat_ws(String string1 [, String string2, ...])<br> | - Return NULL If 
'STRING1' is NULL;<br>- Return a string that concatenates ('STRING2', 
'STRING3', ...) with a separator STRING1.<br> | concat_ws('-', 'apple', 
'banana', 'cloud') = "apple-banana-cloud"<br>concat_ws('-', 'apple', '', 
'cloud') = "apple--cloud"<br>concat_ws('-', 'apple', null, 'cloud') = 
"apple-cloud"<br> |
+| contains(String leftStr , String rightStr)<br> | - Return "" if 'leftStr' or 
rightStr is NULL;<br>- Return True if 'rightStr' is found inside 
'leftStr';<br>- Return False otherwise.<br> | contains('Transform SQL', 'SQL') 
= true<br> |
+| elt(Integer index, Expr expr1[, Expr expr2, ...])<br> | - Return "" if 
'index' is NULL or out of range;<br>- Return the index-th('index' starts from 
1) expression.<br> | elt(2, 'a', 'b', 'c') = "b"<br> |
+| endswith(String s1, String s2)<br> | - Return "" if either argument is 
NULL;<br>- Return whether 's2' ends with 's2'.<br> | endswith('Apache InLong', 
'Long') = true<br> |
+| format(Numeric X,Integer D)<br> | - Return "" if 'X' or 'D' is NULL;<br>- 
Return the result of formatting the number 'X' to "#,###,###.##" format, 
rounded to 'D' decimal places.<br> | FORMAT(12332.123456, 4) = 
"12,332.1235"<br>FORMAT(12332.2,0) = "12,332"<br> |
+| initcap(String s1, String s2)<br>init_cap(String s1, String s2)<br> | - 
Return "" if 'str' is NULL;<br>- Return a new form of 'str' with the first 
character of each word converted to uppercase and the rest characters to 
lowercase.<br> | initcap('hello world') = "Hello world"<br> |
+| insert(String str,Integer pos,Integer len,String newStr)<br> | - Return "" 
If any parameter is NULL;<br>- Return the result of replacing the substring of 
length len with 'newStr' starting from the given position 'pos' in 
'str'.<br>Note: If the position is out of the string's bounds, the original 
string is returned.If the length exceeds the remaining length of the string 
from the given position, the replacement continues to the end of the string. If 
any argument is null, the function re [...]
+| is_alpha(String str)<br> | - Return "" If 'str' is NULL;<br>- Return true if 
all characters in 'str' are letter;<br>- Return false otherwise (Including 
cases where string is null and '').<br> | is_alpha('inlong') = 
true<br>is_alpha('inlong~') = false<br> |
+| is_decimal(String str)<br> | - Return "" if 'str' is NULL;<br>- Return true 
if 'str' can be parsed to a valid numeric;<br>- Return false otherwise 
(Including cases where string is null and '').<br> | is_decimal('3he') = 
false<br>is_decimal('3.5') = true<br> |
+| is_digit(String str)<br> | - Return "" If 'str' is NULL;<br>- Return true if 
all characters in 'str' are digit;<br>- Return false otherwise (Including cases 
where 'str' is null and '').<br> | is_digit('3.5') = false<br> |
+| left(String str, Integer len)<br> | - Return "" if any parameter is 
NULL;<br>- Return "" if 'len' is less than or equal to zero;<br>- Return a 
substring of len starting from the right side of the 'str'.<br> | left('hello 
world',100) = "hello world"<br>left('hello world',-15) = ""<br> |
+| length(String str, String charset)<br> | - Return "" if 'str' is NULL;<br>- 
Return the byte length of the 'str'.<br>Note: charset defaults to matching with 
JVM.<br> | length(应龙, utf-8) = 6<br>length('hello world') = 11<br> |
+| locate(String str1, String str2, Integer pos)<br>instr(String str1, String 
str2, Integer pos)<br> | - Return "" if any of arguments is NULL'<br>- Return 0 
if not found'<br>- Return the position of the first occurrence of 'str1' in 
'str2' after position 'pos'.<br> | locate('app', 'apple') = 1<br>locate('app', 
'appapp', 2) = 4<br> |
+| lower(String str)<br>lcase(String str)<br> | - Return "" if 'str' is 
NULL;<br>- Return the string obtained by converting all letters of 'str' to 
lowercase letters.<br> | lower("ApPlE") = "apple"<br> |
+| lpad(String str1, Integer len, String str2)<br> | - Return null if any of 
the three parameters is null or 'len' is less than 0;<br>- Return the substring 
of 'str1' with subscripts in the range of [0, 'len') if 'len' is less than or 
equal to the length of 'str1';<br>- Return "" if 'len' is longer than the 
length of 'str1' and 'str2' is "";<br>- Return the filled string if 'str2' is 
not "".<br> | lpad('he',7,'xxd') = "xxdxxhe"<br>lpad('he',1,'') = "h"<br> |
+| ltrim(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the string 
'str' without leading spaces.<br> | ltrim(' in long ') = "in long "<br> |
+| parse_url(String urlStr, String keyword[, String parameter])<br> | - Return 
NULL any required parameter is NULL;<br>- Return the specified part from the 
'urlStr'. Valid values for 'keyword'.<br>Note: The 'keyword' is one of ('HOST', 
'PATH', 'QUERY', 'REF', 'PROTOCOL', 'AUTHORITY', 'FILE', 'USERINFO'). Also a 
value of a particular key in QUERY can be extracted by providing the key as the 
third argument.<br> | 
parse_url('http://inlong.apache.org/p.php?k1=v1&k2=v2#Ref1', 'QUERY', 'k1') =  
[...]
+| printf(String strfmt [, Object obj, ...])<br> | - Return a formatted string 
from printf-style format strings.<br> | printf("User %s has %d points and a 
balance of %.2f.", "Bob", 1500, 99.99) = " User Bob has 1500 points and a 
balance of 99.99."<br> |
+| regex(String str, String regexp)<br>similar(String str, String 
regexp)<br>regexp_like(String str, String regexp)<br> | - Return "" if any of 
the arguments are NULL or invalid;<br>- Return TRUE if any (possibly empty) 
substring of 'str' matches the Java regular expression 'regexp'.<br> | 
regexp("The quick brown fox", "quick") = true<br>regexp("The quick brown fox", 
"cold") = false<br> |
+| regex_substr(String source_string, String regexp)<br> | - Return "" if any 
of the arguments are NULL or regexp if invalid or pattern is not found;<br>- 
Return the first substring in 'str' that matches 'regexp'.<br> | 
regex_substr("abc123def", "(\\d+)") = 123<br> |
+| regexp_count(Integer INT1, [Integer INT2])<br> | - Return "" if any of the 
arguments are NULL or 'regexp' is invalid;<br>- Return the number of times 
'str' matches the 'regexp' pattern.<br>Note: 'regexp' must be a Java regular 
expression.<br> | regexp_count("The quick brown fox quick", "quick") = 2<br> |
+| regexp_extract(String str, String regexp, [Integer extractIndex])<br> | - 
Return "" if any of the arguments are NULL or invalid;<br>- Return a string 
from 'str' which extracted with a specified regexp expression 'regexp' and a 
regexp match group index 'extractIndex'.<br>Note: 'regexp' must be a Java 
regular expression. 'extractIndex' indicates which regexp group to extract and 
starts from 1, also the default value if not specified. 0 means matching the 
entire 'regexp' expression. In ad [...]
+| regexp_extract_all(String str, String regexp, [Integer extractIndex])<br> | 
- Return "" if any of the arguments are NULL or invalid;<br>- Return an array 
consisting of all substrings from the 'str' that match the 'regexp' expression 
and correspond to the 'extractIndex' set of regular expressions.<br>Note: 
'regexp' must be a Java regular expression. 'extractIndex' indicates which 
regexp group to extract and starts from 1, also the default value if not 
specified. 0 means matching the ent [...]
+| regexp_instr(String str, String regexp)<br> | - Return "" if any of the 
arguments are NULL or invalid;<br>- Return the position of the first substring 
in 'str' that matches 'regexp'.<br> | regexp_instr("abc123def", "(\\d+)") = 
4<br> |
+| regexp_matches(String source_string, String regexp [,String flags])<br> | - 
Return "" if any of the arguments are NULL or invalid;<br>- Return the result 
of the first match of the specified regular expression 'regexp' from 
'source_string'. Note: 'flags' is one of  ('g' -> flag can be used when we want 
to match all the substrings that occur, 'i' -> flag to ignore case for 
matching, 'x' -> flag to extend syntax (ignoring whitespace and comments in 
regular expressions), 'm' and 'n' -> fla [...]
+| regexp_replace(String source_string, String regexp, String replacement)<br> 
| - Return "" if any of the arguments are NULL or invalid;<br>- Return a string 
from 'source_string' with all the substrings that match a regular expression 
'regexp' consecutively being replaced with 'replacement'.<br> | 
regexp_replace('foobarbaz', 'b..', 'X') = "fooXbaz"<br> |
+| regexp_split_to_array(String source_string, String regexp, String flags)<br> 
| - Return "" if any of the arguments are NULL or invalid;<br>- Return a string 
from 'source_string' with all the substrings that match a regular expression 
'regexp' consecutively being replaced with 'replacement'.<br>Note: 'flags' is 
one of  ('g' -> flag can be used when we want to match all the substrings that 
occur, 'i' -> flag to ignore case for matching, 'x' -> flag to extend syntax 
(ignoring whitespace a [...]
+| repeat(String str, Integer times)<br>replicate(String str, Integer 
times)<br> | - Return "" if any parameter is null;<br>- Return a new string 
that repeats the 'str' by a certain number of 'times'.<br> | repeat('apple', 2) 
= "appleapple"<br> |
+| replace(String s, String s1, String s2)<br> | - Return "" if any parameter 
is null;<br>- Return the result of replacing string 's1' with string 's2' in 
string 's'.<br> | replace('Hello World', '', 'J') = "JHJeJlJlJoJ 
JWJoJrJlJdJ"<br> |
+| reverse(String str)<br> | - Return "" if 'str' is a empty string or 
NULL;<br>- Return the 'str' with the order of the characters reversed.<br> | 
reverse('apple') = "elppa"<br> |
+| right(String str)<br> | - Return "" if either 'str' or 'len' is NULL;<br>- 
Return "" if it is less than or equal to zero;<br>- Return a substring of 'len' 
starting from the right side of the 'str'.<br> | right('hello world',100) = 
"hello world"<br> |
+| rpad(String str)<br> | - Return "" if any of the three parameters is NULL or 
'len' is less than 0;<br>- Return the substring of 's1' with subscripts in the 
range of [0, 'len') if 'len' is less than or equal to the length of 's1';<br>- 
Return "" if 's2' is "" and 'len' is longer than the length of 's1';<br>- 
Return the result string of padding string 's2' at the end of string 's1' to 
make the length of the string 'len' if 's2' is not "".<br> | rpad('he',1,'xxd') 
= "h"<br>rpad('he',7,'') [...]
+| rtrim(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the string 
'str' with trailing space characters removed.<br> | rtrim(' in long ') = " in 
long"<br> |
+| soundex(String str)<br> | - Return "" if 'str' is NULL;<br>- Return a four 
character code representing the sound of 'str'.<br> | soundex('hello world') = 
"H464"<br> |
+| space(Integer N)<br> | - Return "" if 'N' is NULL or less than or equal to 
0;<br>- Return a string consisting of 'N' space characters.<br> | space(5) = "  
   "<br> |
+| split_index(String str, String delimiter, Integer 
index)<br>splitindex(String str, String delimiter, Integer index)<br> | - 
Return "" if the index is negative or any of the arguments is NULL;<br>- Return 
"" NULL if the index is out of bounds of the split strings;<br>- Return the 
string at the given 'index' integer(zero-based) after splitting 'str' by 
'delimiter'.<br> | split_index('a,b,c', ',', 1) = "b"<br> |
+| startswith(String s1, String s2)<br> | - Return "" if either argument is 
NULL;<br>- Return whether 's2' starts with 's2'.<br> | startswith('Apache 
InLong', 'A') = true<br> |
+| str_to_map(String s1, String pairDelimiter, String keyValueDelimiter)<br> | 
- Return "" if 'str' is NULL;<br>- Return a map after splitting the 'str' into 
key/value pairs using 'pairDelimiter'(default is ',') and 
'keyValueDelimiter'(default is '=');<br>Note: Both 'pairDelimiter' and 
'keyValueDelimiter' are treated as regular expressions.So special 
characters(e.g. <([{^-=$!\|]})?*+.>) need to be properly escaped before using 
as a delimiter literally.<br> | str_to_map('key1=value1,key2=v [...]
+| strcmp(String s1, String s2)<br> | - Return "" if either argument is 
NULL;<br>- Return 0 if the strings are the same;<br>- Return -1 if 's1' is 
smaller than 's2' according to the current sort order;<br>- Return 1 
otherwise.<br> | strcmp('hello world','banana') = 1<br> |
+| substring(String s1, Integer pos, Integer len)<br>substr(String s1, Integer 
pos, Integer len)<br>mid(String s1, Integer pos, Integer len)<br> | - Return "" 
if 'str' is NULL;<br>- Return a substring of 'str' starting from position 'pos' 
with length 'len' (to the end by default).<br>Note: This function also supports 
"substring(str FROM pos [ FOR len ])".<br> | substring('apple', 1, 3) = 
"app"<br> |
+| substring_index(String str, String delim, Integer count)<br> | - Return "" 
if any parameter is NULL;<br>- Return everything to the left of the last count 
occurrences of 'delim' (counting from the left), if 'count' is positive;<br>- 
Return everything to the right of the last count occurrences of 'delim' 
(counting from the right) if 'count' is negative.<br> | SUBSTRING_INDEX('AA. 
',' ',1) = "AA."<br> |
+| translate(String origin_string, String find_chars, String replace_chars)<br> 
| - Return "" if any parameter is NULL;<br>- Return the result of replacing all 
occurrences of both 'find_chars' and 'origin_string' with the characters in 
'replace_chars'.<br> | translate(apa...@inlong.com, '@', '.') = 
"apache.inlong.com"<br>translate(hello WorD, 'WD', 'wd') = "hello word"<br> |
+| trim(String str)<br>btrim(String str)<br> | - Return "" if 'str' is 
NULL;<br>- Return the result of deleting spaces before and after the 'str'.<br> 
| trim(' in long ') = "in long"<br> |
+| unhex(String str)<br> | - Return "" if 'str' is NULL;<br>- Return the result 
of interpreting each pair of characters in the argument as the character 
corresponding to its hexadecimal number.<br> | unhex("696E6C6F6E67") = 
"inlong"<br> |
+| upper(String s)<br>ucase(String s)<br> | - Return "" if 's' is NULL;<br>- 
Return the result of converting 's' to uppercase<br> | upper("ApPlE") = 
"APPLE"<br> |
+| url_decode(String str[, String charset])<br> | - Return "" if 'str' is NULL, 
or there is an issue with the decoding process(such as encountering an illegal 
escape pattern), or the encoding scheme is not supported;<br>- Return the 
result of decoding a given 'str' in 'application/x-www-form-urlencoded' format 
using the charset(default:UTF-8) encoding scheme.<br> | 
url_decode('https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+encode') = 
"https://apache.inlong.com/search?q=java url e [...]
+| url_encode(String str[, String charset])<br> | - Return "" if 'str' is NULL, 
or there is an issue with the decoding process(such as encountering an illegal 
escape pattern), or the encoding scheme is not supported;<br>- Return the 
result of translating 'str' into 'application/x-www-form-urlencoded' format 
using the charset(default:UTF-8) encoding scheme.<br> | 
url_encode('https://apache.inlong.com/search?q=java url encode') = 
"https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+enco [...]
+| uuid(String s)<br> | - Return a UUID (universally unique identifier) string 
based on RFC 4122 Type 4 (pseudo-randomly generated) UUID.<br> | uuid()<br> |
+# Temporal
+| Function | Description | Example |
+| ------ | ------ | ------ |
+| convert_tz(String leftStr , String rightStr)<br> | - Return NULL if any 
parameter is NULL;<br>- Return the result of converts a datetime 'string1' 
(with default ISO timestamp format yyyy-MM-dd HH:mm:ss’) from time zone 
'string2' to time zone 'string3'.<br>Note: The format of time zone should be  
either an abbreviation such as “PST”, a full name such as 
“America/Los_Angeles”, or a custom ID such as “GMT-08:00”.<br> | 
CONVERT_TZ('1970-01-01 00:00:00', 'UTC', 'America/Los_Angeles') = "196 [...]
+| date_add(String dateStr,String intervalExprStr)<br> | - Return "" if 
'dateStr' is NULL;<br>- Return DATE if 'dateStr' is a DATE value and your 
calculations involve only YEAR, MONTH, and DAY parts(that is, no time 
parts);<br>- Return TIME if 'dateStr' is a TIME value and the calculations 
involve only HOURS, MINUTES,and SECONDS parts (that is, no date parts);<br>- 
Return DATETIME if the first argument is a DATETIME (or TIMESTAMP) value, or if 
the first argument is a DATE and the unit val [...]
+| date_format(String timestampStr, String formatStr)<br> | - Return "" if any 
parameter is NULL;<br>- Return a string value that converts a timestamp (in 
seconds) to a date format string in the specified format.<br>Note: The format 
string is compatible with Java’s SimpleDateFormat<br> | date_format('2024-08-01 
22:56:56', 'yyyy/MM/dd HH:mm:ss') = "2024/08/01 22:56:56"<br> |
+| date_sub(String dateStr,String intervalExprStr)<br>datesub(String 
dateStr,String intervalExprStr)<br> | - Return "" if 'dateStr' is NULL;<br>- 
Return DATE if the date argument is a DATE value and your calculations involve 
only YEAR, MONTH, and DAY parts(that is, no time parts);<br>- Return TIME if 
the date argument is a TIME value and the calculations involve only HOURS, 
MINUTES,and SECONDS parts (that is, no date parts);<br>- Return DATETIME if the 
first argument is a DATETIME (or TIM [...]
+| datediff(String dateStr1, String dateStr2)<br> | - Return "" if one of the 
two parameters is null or "";<br>- Return "" if one of the two parameters has 
an incorrect date format;<br>- Return the number of days between the dates 
'dateStr1'->'dateStr2'.<br> | datediff('2018-12-10 12:30:00', '2018-12-09 
13:30:00') = 1<br>datediff('2018-12', '2018-12-12') = ""<br> |
+| day_name(String dateStr)<br>dayname(String dateStr)<br> | - Return "" if 
'dateStr' is null;<br>- Return the name of the day of the week from 
'dateStr'.<br> | dayname(2024-02-29) = THURSDAY<br> |
+| day_of_month(String dateStr)<br>dayofmonth(String dateStr)<br> | - Return "" 
if 'dateStr' is null;<br>- Return the day of a month (an integer between 1 and 
31) from 'dateStr'.<br> | dayofmonth(2024-02-29) = 29<br> |
+| day_of_week(String dateStr)<br>dayofweek(String dateStr)<br> | - Return "" 
if 'dateStr' is null;<br>- Return the day of a week (an integer between 
1(Sunday) and 7(Saturday)) from 'dateStr'.<br> | dayofweek(2024-02-29) = 5<br> |
+| day_of_year(String dateStr)<br>dayofyear(String dateStr)<br> | - Return "" 
if 'dateStr' is null;<br>- Return the day of a year (an integer between 1 and 
366) from 'dateStr'.<br> | dayofyear(2024-02-29) = 60<br> |
+| from_unix_time(Numeric numeric,String formatStr)<br>form_unixtime(Numeric 
numeric,String formatStr)<br> | - Return "" if 'numeric' is NULL;<br>- Return a 
representation of the numeric argument as a value in string.<br>Note: 
'formatStr' defaults to 'yyyy-MM-dd HH:mm:ss'. numeric is an internal timestamp 
value representing seconds since ‘1970-01-01 00:00:00’ UTC, such as produced by 
the UNIX_TIMESTAMP() function.<br> | from_unix_time(44) = "1970-01-01 
08:00:44"<br> |
+| hour(String timestamp)<br> | - Return "" if 'timestamp' is null;<br>- Return 
the hour of a day (an integer between 0 and 23) from SQL 'timestamp'.<br> | 
hour(2024-08-12 12:23:34) = 12<br> |
+| localdate([String timeZoneStr])<br>currentdate([String 
timeZoneStr])<br>current_date([String timeZoneStr])<br>curdate([String 
timeZoneStr])<br> | - Return the current date in the specified 
'timeZoneStr'.<br> | localDate("UTC") = currentDate<br>localDate() = 
currentDate<br> |
+| localtime([String timeZoneStr])<br>current_time([String timeZoneStr])<br> | 
- Return the current time in the specified time zone.<br> | localTime() = 
currentTime<br>currentTime("UTC") = currentTime<br> |
+| minute(String timestamp)<br> | - Return "" if 'timestamp' is null;<br>- 
Return the minute of an hour (an integer between 0 and 59) from SQL 
'timestamp'.<br> | minute(2024-08-12 12:23:34) = 23<br> |
+| month(String dateStr)<br> | - Return "" if 'dateStr' is null;<br>- Return 
the month of a year (an integer between 1 and 12) from 'dateStr'.<br> | 
month(2024-08-08) = 8<br> |
+| now()<br> | - Return the current SQL timestamp for the local time zone.<br> 
| now()<br> |
+| quarter(String dateStr)<br> | - Return "" if 'dateStr' is null;<br>- Return 
the quarter of a year (an integer between 1 and 4) from 'dateStr'.<br> | 
quarter(2024-08-08) = 3<br> |
+| second(String timestamp)<br> | - Return "" if 'timestamp' is null;<br>- 
Return the second of a minute (an integer between 0 and 59) from SQL 
'timestamp'.<br> | second(2024-08-12 12:23:34) = 34<br> |
+| timediff(String dateStr1, String dateStr2)<br>time_diff(String dateStr1, 
String dateStr2)<br> | - Return "" if 'dateStr1' or 'dateStr2' is NULL and the 
conversion types of 'dateStr1' and 'dateStr2' are different;<br>- Return 
'dateStr1' - 'dateStr2' expressed as a time value.<br>Note: 'dateStr1' and 
'dateStr2' are strings converted to TIME or DATETIME expressions.<br> | 
timediff('23:59:59.000001','01:01:01.000002') = "22:58:57.999998"<br> |
+| timestamp(String unit, String datetime_expr1, String datetime_expr2)<br> | - 
Return "" if 'datetime_expr1' or 'datetime_expr2' is NULL;<br>- Return the date 
or datetime expression expr as a datetime value if there is only one 
parameter;<br>- Return the result of the date or date time expression 
'datetime_expr1' plus the time expression 'datetime_expr2' if there are two 
parameters.<br> | timestamp('2003-12-31 12:00:00.600000','12:00:00') = 
"2004-01-01 00:00:00.600000"<br> |
+| timestamp_add(String unit, Integer cnt, String 
baseDateStr)<br>timestampadd(String unit, Integer cnt, String baseDateStr)<br> 
| - Return "" if any parameter is null;<br>- Return the result of adding the 
integer expression interval to the date or datetime expression 
'baseDateStr'.<br>Note: 'unit' is one of (MICROSECOND, SECOND, MINUTE, HOUR, 
DAY, WEEK, MONTH, QUARTER, YEAR).<br> | 
timestamp_add('MICROSECOND',3,'1970-01-01 00:00:44') = "1970-01-01 
00:00:44.000003"<br> |
+| timestamp_diff(String unit, String datetime_expr1, String 
datetime_expr2)<br>timestampdiff(String unit, String datetime_expr1, String 
datetime_expr2)<br> | - Return "" if any parameter is null;<br>- Return 
'datetime_expr2' − 'datetime_expr1', where 'datetime_expr1' and 
'datetime_expr2' are date or datetime expressions.<br>Note: 'unit' is one of 
(MICROSECOND, SECOND, MINUTE, HOUR, DAY, WEEK, MONTH, QUARTER, YEAR).<br> | 
timestampdiff('MONTH','2003-02-01','2003-05-01') = 3<br> |
+| to_date(String str [,String format])<br> | - Return "" if 'str' is 
NULL;<br>- Return the result of converting the date string 'str' to a date in 
the format 'format'(default is 'yyyy-MM-dd').<br> | to_date('20240815', 
'yyyyMMdd') = "2024-08-15"<br> |
+| to_timestamp(String str [,String format])<br> | - Return "" if 'str' is 
NULL;<br>- Return the result of converting the date and time string 'str' to 
the 'format' (by default: yyyy-MM-dd HH:mm:ss if not specified) under the 
'UTC+0' time zone to a timestamp.<br> | to_timestamp('1970/01/01 00:00:44', 
'yyyy/MM/dd HH:mm:ss') = "1970-01-01 00:00:44.0"<br> |
+| unix_timestamp([String dateStr [, String format]])<br> | - Return current 
Unix timestamp in seconds if no parameter is specified;<br>- Return the result 
of converting the date and time string 'dateStr' to the format 'format' (by 
default: yyyy-MM-dd HH:mm:ss if not specified) to Unix timestamp (in seconds) 
if there is a parameter specified<br> | unix_timestamp('1970/01/01 08:00:44', 
'yyyy/MM/dd HH:mm:ss') = "1970/01/01 08:00:44"<br> |
+| week(String dateStr)<br> | - Return "" if 'dateStr' is null;<br>- Return the 
week of a year (an integer between 1 and 53) from 'dateStr'.<br> | 
week(2024-02-29) = 9<br> |
+| year(String dateStr)<br> | - Return "" if 'dateStr' is null;<br>- Return the 
year from SQL date.<br> | year(2024-08-08) = 2024<br> |
\ No newline at end of file
diff --git 
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/modules/transform/function_specification.md
 
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/modules/transform/function_specification.md
new file mode 100644
index 00000000000..5add18a0def
--- /dev/null
+++ 
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/modules/transform/function_specification.md
@@ -0,0 +1,229 @@
+# 函数说明
+
+## Arithmetic (算术函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| abs(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的绝对值 | 
abs(2) = 2<br>abs(-4.25) = 4.25 |
+| acos(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的反余弦值 
| acos(1) = 0.0<br>acos(0) = 1.5707963267948966<br>acos(-1) = 3.141592653589793 
|
+| acosd(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的反余弦值（以度为单位） | acosd(1) = 0.0<br>acosd(0) = 90.0<br>acosd(-1) = 180.0 |
+| asin(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的反正弦值 
| asin(0.5) = 0.5235987755982989<br>asin(0) = 0.0<br>asin(-0.5) = 
-0.5235987755982989 |
+| asind(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的反正弦值（以度为单位） | asind(0.5) = 30.000000000000004<br>asind(0) = 0.0<br>asind(-0.5) 
= -30.000000000000004 |
+| atan(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的反正切值 
| atan(1) = 0.7853981633974483<br>atan(0) = 0.0<br>atan(-1) = 
-0.7853981633974483 |
+| atan2(Numeric numericx, Numeric numericy) | - 如果 'numericx' 或 'numericy' 为 
NULL，返回 ""<br>- 返回坐标 ('numericx', 'numericy') 的反正切值 | atan2(1, 1) = 
0.7853981633974483<br>atan2(1, 0) = 1.5707963267948966<br>atan2(0, -1) = 
3.141592653589793 |
+| atan2d(Numeric numericx, Numeric numericy) | - 如果 'numericx' 或 'numericy' 为 
NULL，返回 ""<br>- 返回 'numericy'/'numericx' 的反正切值（以度为单位） | atan2d(1, 1) = 
45.0<br>atan2d(1, 0) = 90.0<br>atan2d(0, -1) = 180.0 |
+| atand(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的反正切值（以度为单位） | atand(1) = 45.0<br>atand(0) = 0.0<br>atand(-1) = -45.0 |
+| bin(Integer integer) | - 如果 'integer' 为 NULL，返回 ""<br>- 返回 'integer' 
的二进制字符串表示 | bin(4) = 100 |
+| cbrt(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 NULL<br>- 返回立方根 | cbrt(5) = 
1.7099759466766968<br>cbrt(27) = 3.0 |
+| ceil(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 NULL<br>- 返回大于或等于 'numeric' 
的最小整数 | ceil(-5.67) = -5.0<br>ceil(1.23) = 2.0 |
+| chr(Integer integer) | - 如果 'integer' 为 NULL，返回 NULL<br>- 返回与 'integer' 
二进制等价的 ASCII 字符 | chr(97) = "a"<br>chr(353) = "a" |
+| cos(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的余弦值 | 
cos(0) = 1.0 |
+| cosd(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的余弦值（以度为单位） | cosd(15) = 0.9659258262890683 |
+| cosh(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的双曲余弦值 
| cosh(1) = 1.543080634815244<br>cosh(0) = 1.0<br>cosh(-1) = 1.543080634815244 |
+| cot(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的余切值（以弧度为单位） | cot(1) = 0.6420926159343306<br>cot(0.5) = 
1.830487721712452<br>cot(-1) = -0.6420926159343306 |
+| cotd(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的余切值（以度为单位） | cotd(0) = ""<br>cotd(45) = 1.0000000000000002<br>cotd(-1) = 
-57.28996163075943 |
+| e() | - 返回数学常数 e | e() = 2.718281828459045 |
+| erf(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回输入值的误差函数值 | erf(3) 
= 0.9999779095030014 |
+| erfc(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回互补误差函数值 (1 - 
erf('numeric'))，对于大输入值不会损失精度 | erfc(1) = 0.15729920705028488 |
+| exp(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 e 的 'numeric' 次幂 
| exp(2) = 7.38905609893065 |
+| factorial(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回非负数的阶乘 | 
factorial(5) = 120<br>factorial(0) = 1 |
+| fibonacci(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 NULL<br>- 返回第 n 个斐波那契数 
| fibonacci(0) = 0<br>fibonacci(1) = 1<br>fibonacci(2) = 1<br>fibonacci(3) = 
2<br>fibonacci(4) = 3 |
+| floor(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回小于或等于 'numeric' 
的最大整数 | floor(1.23) = 1.0<br>floor(-5.67) = -6.0 |
+| gcd(Numeric numeric1, Numeric numeric2) | - 如果任一参数为 NULL，返回 ""<br>- 如果两个输入都为 
0，返回 0<br>- 返回最大公约数（能整除两个输入的最大正整数） | gcd(3.141,3.846) = 0.003 |
+| greatest(Numeric value1 [,Numeric value2,...]) | - 如果任一参数为 NULL，返回 ""<br>- 
返回参数列表中的最大值 | greatest(3.141592653589793, 3, greatest(4, 1)) = 4 |
+| hex(String dataStr) | - 如果 dataStr 为 NULL，返回 ""<br>- 如果 'dataStr' 
可以解析为数字，则将其转换为十六进制字符串<br>- 否则将每个字符对应的 ASCII 码转换为十六进制字符串 | hex(1007) = 
"3EF"<br>hex('abc') = "616263" |
+| lcm(Numeric numeric1, Numeric numeric2) | - 如果任一参数为 NULL，返回 ""<br>- 如果任一输入为 
0，返回 0<br>- 返回最小公倍数（两个输入的整数倍的最小正整数） | lcm(6,3) = 6<br>lcm(3.141,3.846) = 
4026.762 |
+| least(Numeric value1 [, Numeric value2, ...]) | - 如果任一参数为 NULL，返回 ""<br>- 
返回参数列表中的最小值 | least(3.14, least(7, 2, 1)) = 1 |
+| ln(Numeric numeric) | - 如果 numeric 为 NULL，返回 ""<br>- 返回 numeric 的自然对数（以 e 
为底） | ln(10) = 2.302585092994046 |
+| log(Numeric numeric1 [, Numeric numeric2]) | - 一个参数时：返回 'numeric1' 
的自然对数<br>- 两个参数时：返回以 'numeric1' 为底 'numeric2' 的对数 | log(1) = 0.0<br>log(2,8) = 
3.0 |
+| log10(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的以 10 
为底的对数 | log10(1000) = 3.0 |
+| log2(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的以 2 
为底的对数 | log2(32) = 5 |
+| min_scale(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 
返回精确表示该值所需的最小小数位数 | min_scale(3.1415000000000) = 4 |
+| mod(Numeric numeric1, Numeric numeric2) | - 如果任一参数为 NULL，返回 ""<br>- 返回 
'numeric1' 除以 'numeric2' 的余数 | mod(3,2) = 1<br>mod(-3.1415926,100) = -3.1415926 
|
+| num_nonnulls([Expr expr1, Expr expr2, ...]) | - 返回非空参数的数量 | num_nonnulls(5, 
3, null, null) = 2 |
+| num_nulls([Expr expr1, Expr expr2, ...]) | - 返回空参数的数量 | num_nulls(5, null, 
null, null) = 3 |
+| pi() | - 返回数学常数 PI | pi() = 3.141592653589793 |
+| power(Numeric numeric1, Numeric numeric2) | - 如果任一参数为 NULL，返回 ""<br>- 返回 
'numeric1' 的 'numeric2' 次幂 | power(4,0.5) = 2.0 |
+| radians(Numeric x) | - 如果 'x' 为 NULL，返回 ""<br>- 返回 'x' 的弧度值，将度转换为弧度 | 
radians(18.97) = 0.33108895910332425 |
+| radix_convert(Numeric numeric, Integer from_base, Integer to_base) | - 
如果任一参数为 NULL，返回 ""<br>- 将 'numeric' 从 'from_base' 进制转换为 'to_base' 
进制<br>注意：abs(base) 在 [2,36] 之间。'from_base' 为负数时，'numeric' 
被视为有符号数，否则视为无符号数。此函数使用 64 位精度 | radix_convert('6E',18,8) = 172 |
+| rand([Integer seed]) | - 如果 seed 为 NULL 或无参数：返回 [0.0, 1.0) 范围内的伪随机双精度值<br>- 
有 seed 参数：使用初始种子返回伪随机双精度值 | rand(1)<br>rand() |
+| rand_integer(Integer INT1, [Integer INT2]) | - 如果 'INT2' 为 NULL：返回 [0, 
'INT1') 范围内的伪随机整数值<br>- 有 'INT2' 参数：使用初始种子返回伪随机整数值<br>注意：如果两个 RAND_INTEGER 
函数具有相同的初始种子和边界，将返回相同的数字序列 | rand_integer(10)<br>rand_integer(88, 89) |
+| round(Numeric x [, Integer y]) | - 如果 'x' 为 NULL，返回 ""<br>- 返回最接近 'x' 
的整数，可选参数 'y' 指定要保留的小数位数 | round(3.5) = 4<br>round(3.14159265358979323846,10) = 
3.1415926536 |
+| scale(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回参数的小数位数 | 
scale(3.1415000000000) = 13 |
+| sign(Numeric x) | - 如果 'x' 为 NULL，返回 ""<br>- 如果 'x' 为负数，返回 -1<br>- 如果 'x' 等于 
0，返回 0<br>- 如果 'x' 为正数，返回 1 | sign(-3.5) = -1 |
+| sin(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的正弦值 | 
sin(0) = 0.0 |
+| sind(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的正弦值（以度为单位） | sind(15) = 0.25881904510252074 |
+| sinh(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的双曲正弦值 
| sinh(1) = 1.1752011936438014 |
+| sqrt(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的平方根 | 
sqrt(9) = 3.0 |
+| tan(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的正切值 | 
tan(1) = 1.5574077246549023 |
+| tand(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 
的正切值（以度为单位） | tand(15) = 0.2679491924311227 |
+| tanh(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 返回 'numeric' 的双曲正切值 
| tanh(1) = 0.7615941559557649 |
+| trim_scale(Numeric numeric) | - 如果 'numeric' 为 NULL，返回 ""<br>- 
通过去除尾随零来减少值的小数位数 | trim_scale(3.1415000000000) = 3.1415 |
+| truncate(Numeric numeric, Integer cnt)<br>trunc(Numeric numeric, Integer 
cnt) | - 如果任一参数为 NULL，返回 ""<br>- 将 'numeric' 值截断为 'cnt' 位小数 | truncate(42.324, 
2) = 42.32<br>trunc(42.324, -1) = 40 |
+
+## Collection (集合函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| array(String value1 [,String value2, ....]) | - 从值列表创建数组 | 
array('he',7,'xxd') = [he, 7, xxd]<br>array(array('he',5),'xxd') = [[he, 5], 
xxd]<br>array(array('he',5),array('','')) = [[he, 5], [, ]] |
+| array_append(Array array, Object element) | - 如果任一参数为 NULL，返回 ""<br>- 
将元素追加到数组末尾 | array_append(array('he',7,'xxd'), 'cloud') = [he, 7, xxd, cloud] |
+| array_concat([Array array1, Array array2, ...]) | - 如果任一输入数组为 NULL，返回 
""<br>- 返回连接至少一个数组的结果 | array_concat(array('he',7),array('xxd', 'cloud')) = 
[he, 7, xxd, cloud] |
+| array_contains(Array array, Object needle) | - 如果 'array' 为 NULL，返回 ""<br>- 
返回给定元素是否存在于 'array' 中<br>注意：支持检查数组中的 null 元素 | 
array_contains(array('he',7,'xxd'), 'cloud') = 
false<br>array_contains(array('he',-1,''),'') = true |
+| array_distinct(Array array) | - 如果 'array' 为 NULL，返回 ""<br>- 返回具有唯一元素的数组 | 
array_distinct(array('he',-1,'he')) = [he, -1] |
+| array_except(Array array1, Array array2) | - 如果任一参数为 NULL，返回 ""<br>- 返回包含 
'array1' 中不在 'array2' 中的元素的数组（去重）<br>- 如果排除后没有剩余元素，返回空数组 | 
array_except(array('he',7,'xxd'),array('he')) = [7, 
xxd]<br>array_except(array('he',7,'xxd'),array('cloud')) = [he, 7, xxd] |
+| array_intersect(Array array1, Array array2) | - 如果任一参数为 NULL，返回 ""<br>- 返回包含 
'array1' 和 'array2' 中共有元素的数组（去重）<br>- 如果没有共同元素，返回空数组 | 
array_intersect(array('he',7,'xxd'),array('he')) = 
[he]<br>array_intersect(array('he',7,'xxd'),array('cloud')) = [] |
+| array_join(Array array, String delimiter[, String nullReplacement]) | - 
如果任一参数为 NULL，返回 ""<br>- 使用 'delimiter' 连接数组元素，空数组元素用 'nullReplacement' 
填充<br>注意：如果未指定 nullReplacement，数组中的 null 元素将从结果字符串中省略 | 
array_join(array('he',7,'xxd'),'~') = 
he~7~xxd<br>array_join(array('he',3,''),'~','oo') = he~3~oo |
+| array_max(Array array) | - 如果任一参数为 NULL，返回 ""<br>- 返回数组中的最大值 | 
array_max(array(4,3,56)) = 56 |
+| array_min(Array array) | - 如果任一参数为 NULL，返回 ""<br>- 返回数组中的最小值 | 
array_min(array(4,3,56)) = 3 |
+| array_position(Array array, Object element) | - 如果任一参数为 NULL，返回 ""<br>- 返回 
'element' 在数组中第一次出现的位置（从 1 开始）<br>- 如果未找到，返回 0 | 
array_position(array('he',7,'xxd'),'he') = 
1<br>array_position(array('he',7,''),'_') = 0 |
+| array_prepend(Array array, Object element) | - 如果 'array' 为 NULL，返回 ""<br>- 
将元素添加到数组开头 | array_prepend(array(4,3),3) = [3, 4, 3] |
+| array_remove(Array array, Object element) | - 如果 'array' 为 NULL，返回 ""<br>- 
从数组中移除所有等于 'element' 的元素 | array_remove(array('he',7,'xxd'),'he') = [7, xxd] |
+| array_reverse(Array array) | - 如果 'array' 为 NULL，返回 ""<br>- 返回反转顺序的数组 | 
array_reverse(array('he',7,'xxd')) = [xxd, 7, he] |
+| array_slice(Array array, Integer start_offset[, Integer end_offset]) | - 如果 
'array' 或 'start_offset' 为 NULL，返回 ""<br>- 返回从 'start_offset' 到 
'end_offset'（包含）的子数组<br>- 如果 'start_offset' 在 'end_offset' 
之后或两者都超出数组边界，返回空数组<br>注意：如果省略 'end_offset'，则视为数组长度。正值从数组开头计数，负值从末尾计数 | 
array_slice(array('he',7,'xxd'),1,2) = ['he', 
7]<br>array_slice(array('he','xxd','b'),-2,-1) = [3, 'xxd'] |
+| array_sort(Array array[,Boolean ascending_order[,Boolean null_first]]) | - 
如果 'array' 为 NULL，返回 ""<br>- 返回排序后的数组<br>注意：默认升序排序，NULL 在开头。'ascending_order' 为 
true 时升序（NULL 在前），false 时降序（NULL 在后）。'null_first' 独立控制 NULL 位置 | 
array_sort(array('he',7,'xxd')) = [7, he, xxd]<br>array_sort(array(3,7,5)) = 
[3, 5, 7]<br>array_sort(array(,3,7),false,false) = [7, 3, 
]<br>array_sort(array(3,7,),true,false) = [3, 7, ] |
+| array_union(Array array1[, Array array2, ...]) | - 如果任一输入数组为 NULL，返回 ""<br>- 
返回连接至少一个数组的结果 | array_concat(array('he',7),array('xxd', 'cloud')) = [he, 7, 
xxd, cloud] |
+| cardinality(Object input) | - 如果输入为 NULL，返回 ""<br>- 输入为数组时返回元素数量<br>- 
输入为映射时返回条目数量 | cardinality(array('he',7,'xxd')) = 
3<br>cardinality(map('he',7,'xxd',3)) = 2 |
+| element(Array array) | - 如果 'array' 为空或 NULL，返回 ""<br>- 返回数组的唯一元素（基数应为 
1）<br>注意：如果数组有多个元素会抛出异常 | element(array('he')) = he |
+| find_in_set(String str, String strList) | - 如果任一参数为 NULL，返回 ""<br>- 如果 'str' 
不在 'strList' 中或 'strList' 为空字符串，返回 0<br>- 如果字符串 'str' 在由 N 个子字符串组成的字符串列表 
'strList' 中，返回 1 到 N 范围内的值<br>注意：strList 是由 ',' 
字符分隔的子字符串组成的字符串。如果第一个参数包含逗号字符，此函数可能无法正常工作 | FIND_IN_SET('b','a,b,b,c,d') = 
2<br>FIND_IN_SET('','a,,b,c,d') = 2 |
+| map([String value1, String value2, ...]) | - 如果参数数量不是偶数，返回 ""<br>- 
从键值对列表创建映射 | Map('he',7,'xxd') = null<br>Map('he',1,'xxd','cloud') = {he=1, 
xxd=cloud}<br>Map('xxd','cloud',map(1,2),map(3,'apple')) = {xxd=cloud, 
{1=2}={3=apple}} |
+| map_entries(Map map) | - 如果 'map' 为 NULL，返回 ""<br>- 返回给定映射中的所有条目数组 | 
map_entries(Map('he',1,'xxd','cloud')) = [he=1, 
xxd=cloud]<br>map_entries(Map(1,2,'cloud','xxd')) = [xxd=cloud, 1=2] |
+| map_from_arrays(Array array_of_keys, Array array_of_values) | - 如果任一参数为 
NULL，返回 ""<br>- 从键数组和值数组创建映射 | map_from_arrays(array('he', 'xxd'),array(1, 3)) 
= {he=1, xxd=3}<br>map_from_arrays(array('xxd', array('cloud')),array(1, 
array(2))) = {1=xxd, [2]=[cloud]} |
+| map_keys(Map map) | - 如果 'map' 为 NULL，返回 ""<br>- 以数组形式返回映射的键（不保证顺序） | 
map_keys(Map('he',1,'xxd','cloud')) = [he, 
xxd]<br>map_keys(Map('xxd','cloud',map(1,2),map(3,'apple'))) = [xxd, {1=2}] |
+| map_union([Map map1, Map map2, ...]) | - 如果任一映射为 NULL，返回 ""<br>- 
通过合并至少一个映射创建新映射。这些映射应具有相同的映射类型，后面的映射会覆盖前面的映射 | map_union(map('he', 
1),map('xxd', 3)) = {he=1, xxd=3}<br>map_union(map('he', 1),map('he', 3)) = 
{he=3} |
+| map_values(Map map) | - 如果 'map' 为 NULL，返回 ""<br>- 以数组形式返回映射的值（不保证顺序） | 
map_values(Map('he',1,'xxd','cloud')) = [1, 
cloud]<br>map_values(Map('xxd','cloud',map(1,2),map(3,'apple'))) = [cloud, 
{3=apple}] |
+
+## Compression (压缩函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| compress(String string_to_compress [, String compress_type]) | - 如果 
'string_to_compress' 为 NULL，返回 ""<br>- 如果 'string_to_compress' 为 ""，返回 ""<br>- 
返回二进制字符串结果<br>注意：支持三种压缩算法：deflater、gzip 和 zip。'compress_type' 默认为 
deflater。为了以字符串形式输出压缩结果，此方法使用 ISO_8859_1 字符集 | 
length(compress(replicate(string1,100)),'ISO_8859_1') = 
33<br>length(compress(''),'ISO_8859_1') = 0 |
+| uncompress(String string_to_uncompress, String compress_type) | - 如果 
'string_to_uncompress' 为 NULL，返回 ""<br>- 如果 'string_to_uncompress' 为 ""，返回 
""<br>- 返回字符串结果<br>注意：支持三种压缩算法：deflater、gzip 和 zip。'compress_type' 默认为 deflater 
| uncompress(compress('inlong')) = "inlong" |
+
+## Condition (条件函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| coalesce(String value1 [, String value2, ...]) | - 如果所有参数都为 NULL 或 ""，返回 
""<br>- 返回第一个不为 NULL 或 "" 的参数<br>注意：返回类型是其所有参数的限制最少的公共类型。如果所有参数都可为空，返回类型也可为空 | 
coalesce('', 'SQL', 'hh') = "SQL" |
+| if(Expr expr1, Expr expr2, Expr expr3) | - 如果 'expr1' 返回 true，返回 
'expr2'<br>- 否则返回 'expr3' | if(1 = 1,true,false) = true |
+| ifnull(Expr expr1, Expr expr2)<br>if_null(Expr expr1, Expr expr2) | - 如果 
'expr1' 不为 NULL，返回 'expr1'<br>- 否则返回 'expr2' | ifnull(null, 3) = 
3<br>ifnull(6,'YES') = 6 |
+| isnull(Expr expr) | - 如果 'expr' 为 NULL，返回 true<br>- 否则返回 false | isnull(5 + 
3) = false<br>isnull(5 / 0) = true |
+| nullif(Expr expr1, Expr expr2)<br>null_if(Expr expr1, Expr expr2) | - 如果 
'expr1' = 'expr2' 为 true，返回 ""<br>- 否则返回 'expr1' | nullif(5, 3) = 5 |
+
+## Encryption (加密函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| dayu_decrypt | （描述待补充） |  |
+| decode(Binary binary, String charsetStr) | - 如果任一参数为 NULL，返回 ""<br>- 使用 
'charsetStr' 指定的字符集对 'binary' 进行解码<br>注意：'charsetStr' 是以下之一 ('US-ASCII', 
'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16') | 
decode(encode('Hello','UTF-8'),'UTF-8') = "Hello" |
+| encode(String strInfo, String charsetStr) | - 如果任一参数为 NULL，返回 ""<br>- 使用 
'charsetStr' 指定的字符集对 'strInfo' 进行编码<br>注意：'charsetStr' 是以下之一 ('US-ASCII', 
'ISO-8859-1', 'UTF-8', 'UTF-16BE', 'UTF-16LE', 'UTF-16') | 
decode(encode('Hello','UTF-8'),'UTF-8') = "Hello" |
+| from_base64(String base64Str) | - 如果 'base64Str' 为 NULL，返回 ""<br>- 从 
'base64Str' 返回 base64 解码结果 | from_base64('QXBhY2hlIEluTG9uZw==') = "Apache 
InLong" |
+| md5(String string) | - 如果 'string' 为 NULL，返回 ""<br>- 返回 'string' 的 MD5 
哈希值（32 位十六进制数字字符串形式） | md5("") = "d41d8cd98f00b204e9800998ecf8427e"<br>md5("1") 
= "c4ca4238a0b923820dcc509a6f75849b" |
+| sha(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回 40 位十六进制数字字符串（SHA-1 160 位） 
| sha("5") = "ac3478d69a3c81fa62e60f5c3696165a4e5e6ac4" |
+| sha2(String str, Integer hash_length) | - 如果任一参数为 NULL 或 'hash_length' 不是 
(224,256,384,512) 之一，返回 ""<br>- 返回参数的尺度（小数部分中的十进制位数） | sha2("5",224) = 
"b51d18b551043c1f145f22dbde6f8531faeaf68c54ed9dd79ce24d17" |
+| to_base64(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 从 'str' 返回 base64 编码结果 
| to_base64('app-fun') = "YXBwLWZ1bg==" |
+
+## Json (JSON 函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| json_array_append(String json_doc, String path1, String val1[,String path2, 
String val2, ...]) | - 如果任一参数为 NULL，返回 ""<br>- 在 JSON 文档中的指定数组末尾追加值 | 
json_array_append(["a", ["b", "c"], "d"],$[0],2,$[1],3) = 
[["a","2"],["b","c","3"],"d"] |
+| json_array_insert(String json_doc, String path1, String val1[, String path2, 
String val2, ...]) | - 如果任一参数为 NULL，返回 ""<br>- 
在数组中插入值<br>注意：如果传入多组参数，后一组参数的下标需要基于前一组参数更新后的文档下标 | json_array_append(["a", 
{"b": [1, 2]}, [3, 4]], $[1], x) = ["a","x",{"b":[1,2]},[3,4]] |
+| json_arrays([String value1, String value2, ...]) | - 从值列表构造 JSON 数组字符串 | 
JSON_ARRAYS() = []<br>JSON_ARRAYS(1, '2') = 
[1,"2"]<br>JSON_ARRAYS(JSON_ARRAY(1)) = [[1]] |
+| json_exists(String json_doc, String path) | - 如果 'json_doc' 满足给定的 'path' 
搜索条件，返回 true<br>- 否则返回 false | JSON_EXISTS('{"a": true}', '$.a') = 
true<br>JSON_EXISTS('{"a": true}', '$.b') = false<br>JSON_EXISTS('{"a": [{ "b": 
1 }]}', '$.a[0].b') = true |
+| json_insert(String json_doc, String path1, String val1[, String path2, 
String val2, ...]) | - 如果任一参数为 NULL 或 'json_doc' 不是有效的 JSON 文档或任何 path 
参数不是有效的路径表达式或包含 * 或 ** 通配符，返回 ""<br>- 返回向 'json_doc' 插入数据的结果 | 
json_insert({"a": {"b": [1, 2]}, "c": [3, 4]}, $.c[1][1], "2", "$.c[1][1][5]", 
"1") = {"a":{"b":[1,2]},"c":[3,[4,["2","1"]]]} |
+| json_query(String json_doc, String path) | - 如果任一参数为 NULL，返回 ""<br>- 返回从 
'json_doc' 中的 'path' 解析的字符串 | json_query({"people": [{"name": "Alice"}, 
{"name": "Bob"}]}, $.people) = 
[{"name":"Alice"},{"name":"Bob"}]<br>json_query({"list": [null, {"name": 
"John"}]}, $.list[1].name) = John |
+| json_quote(String data)<br>json_string(String data) | - 如果 data 为 NULL，返回 
""<br>- 从字符串 (JSON_QUOTE) 或任何类型的数据 (JSON_STRING) 转换的有效 JSON 
字符串<br>注意：JSON_QUOTE 会转义内部引号和特殊字符 (’"’, ‘', ‘/’, ‘b’, ‘f’, ’n’, ‘r’, ’t’) | 
json_quote('Column1\tColumn2) = \"Column1\\tColumn2\"<br>json_string(true) = 
"true" |
+| json_remove(String json_doc, String path1[, String path2, ...]) | - 如果任一参数为 
NULL 或 'json_doc' 不是有效的 JSON 文档或任何 path 参数不是有效的路径表达式或是 $ 或包含 * 或 ** 通配符，返回 
""<br>- 返回从 'json_doc' 移除数据的结果 | 
json_remove("{\"name\":\"Charlie\",\"hobbies\":[[\"swimming1\",\"swimming2\"], 
\"reading\",\"coding\"]}","$.age") = 
{"hobbies":[["swimming2"],"coding"],"name":"Charlie"} |
+| json_replace(String json_doc, String path1, String val1[, String path2, 
String val2, ...]) | - 如果任一参数为 NULL 或 'json_doc' 不是有效的 JSON 文档或任何 path 
参数不是有效的路径表达式或包含 * 或 ** 通配符，返回 ""<br>- 返回替换 'json_doc' 中现有值的结果 | json_replace("{ 
\"a\": 1, \"b\": [2, 3]}", "$.a", 10, "$.c", "[true, false]") = {"a": 10, "b": 
[2, 3]} |
+| json_set(String json_doc, String path1, String val1[, String path2, String 
val2, ...]) | - 如果任一参数为 NULL 或 'json_doc' 不是有效的 JSON 文档或任何 path 参数不是有效的路径表达式或包含 
* 或 ** 通配符，返回 ""<br>- 返回在 'json_doc' 中插入或更新数据的结果 | 
json_set({"name":"Alice"},"$.name","inlong") = {"name":"inlong"} |
+| json_unquote(String data) | - 如果 data 为 NULL，返回 ""<br>- 
如果值不以双引号开头和结尾，或者以双引号开头和结尾但不是有效的 JSON 字符串字面量，则返回未修改的 'data'<br>注意：JSON_UNQUOTE 
会取消转义特殊字符 ('"', '', '/', 'b', 'f', 'n', 'r', 't') | json_unquote('Hello, 
World!') = "Hello, World!"<br>json_unquote('Complex string with / and \\') = 
"Complex string with / and \\" |
+| json_value(String json_doc, String path) | - 如果任一参数为 NULL，返回 ""<br>- 基于 
'path' 从 JSON 字符串 ('json_doc') 中提取标量值 | json_value({"a": 1}, $.a) = 
1<br>json_value({"person": {"name": "Alice" ,"age": 30}}, $.person.name) = 
Alice |
+
+## String (字符串函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| ascii(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回 'str' 第一个字符的数值 | 
ascii('abc') = 97<br>ascii('A') = 65<br>ascii(null) = "" |
+| bit_length(String str[, String charset]) | - 如果 'str' 为 NULL，返回 ""<br>- 返回 
'str' 的位数<br>注意：字符集默认与 JVM 对齐 | bit_length("hello world") = 
88<br>bit_length("hello 你好","utf-8") = 96 |
+| char_length(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回 'str' 的字符长度 | 
char_length('hello world') = 11<br>char_length('应龙') = 2 |
+| concat(String string1 [, String string2, ...]) | - 如果任一参数为 NULL，返回 NULL<br>- 
返回连接 ('string1', 'string2',...) 的字符串 | CONCAT("AA", "BB", "CC") = "AABBCC" |
+| concat_ws(String string1 [, String string2, ...]) | - 如果 'STRING1' 为 NULL，返回 
NULL<br>- 使用分隔符 STRING1 连接 ('STRING2', 'STRING3', ...) | concat_ws('-', 
'apple', 'banana', 'cloud') = "apple-banana-cloud"<br>concat_ws('-', 'apple', 
'', 'cloud') = "apple--cloud"<br>concat_ws('-', 'apple', null, 'cloud') = 
"apple-cloud" |
+| contains(String leftStr, String rightStr) | - 如果 'leftStr' 或 rightStr 为 
NULL，返回 ""<br>- 如果 'rightStr' 在 'leftStr' 中找到，返回 True<br>- 否则返回 False | 
contains('Transform SQL', 'SQL') = true |
+| elt(Integer index, Expr expr1[, Expr expr2, ...]) | - 如果 'index' 为 NULL 
或超出范围，返回 ""<br>- 返回第 index 个表达式（从 1 开始） | elt(2, 'a', 'b', 'c') = "b" |
+| endswith(String s1, String s2) | - 如果任一参数为 NULL，返回 ""<br>- 返回 's1' 是否以 's2' 
结尾 | endswith('Apache InLong', 'Long') = true |
+| format(Numeric X, Integer D) | - 如果 'X' 或 'D' 为 NULL，返回 ""<br>- 将数字 'X' 格式化为 
"#,###,###.##" 格式，四舍五入到 'D' 位小数 | FORMAT(12332.123456, 4) = 
"12,332.1235"<br>FORMAT(12332.2,0) = "12,332" |
+| initcap(String str)<br>init_cap(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 
返回 'str' 的新形式，每个单词的首字母转换为大写，其余字符转换为小写 | initcap('hello world') = "Hello world" |
+| insert(String str, Integer pos, Integer len, String newStr) | - 如果任一参数为 
NULL，返回 ""<br>- 返回将 'str' 中从位置 'pos' 开始长度为 len 的子字符串替换为 'newStr' 
的结果<br>注意：如果位置超出字符串边界，返回原字符串。如果长度超过从给定位置开始的剩余长度，替换持续到字符串末尾 | INSERT('12345678', 
3, 4, 'word') = '12word78'<br>INSERT('12345678', -1, 4, 'word') = 
'12345678'<br>INSERT('12345678', 3, 100, 'word') = '12word' |
+| is_alpha(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 如果 'str' 中的所有字符都是字母，返回 
true<br>- 否则返回 false（包括字符串为 null 和 '' 的情况） | is_alpha('inlong') = 
true<br>is_alpha('inlong~') = false |
+| is_decimal(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 如果 'str' 可以解析为有效的数字，返回 
true<br>- 否则返回 false（包括字符串为 null 和 '' 的情况） | is_decimal('3he') = 
false<br>is_decimal('3.5') = true |
+| is_digit(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 如果 'str' 中的所有字符都是数字，返回 
true<br>- 否则返回 false（包括 'str' 为 null 和 '' 的情况） | is_digit('3.5') = false |
+| left(String str, Integer len) | - 如果任一参数为 NULL，返回 ""<br>- 如果 'len' 小于或等于零，返回 
""<br>- 返回 'str' 右侧开始的 len 长度的子字符串 | left('hello world',100) = "hello 
world"<br>left('hello world',-15) = "" |
+| length(String str[, String charset]) | - 如果 'str' 为 NULL，返回 ""<br>- 返回 'str' 
的字节长度<br>注意：字符集默认与 JVM 匹配 | length('应龙', 'utf-8') = 6<br>length('hello world') 
= 11 |
+| locate(String str1, String str2[, Integer pos])<br>instr(String str1, String 
str2[, Integer pos]) | - 如果任一参数为 NULL，返回 ""<br>- 如果未找到，返回 0<br>- 返回在位置 'pos' 之后 
'str1' 在 'str2' 中第一次出现的位置 | locate('app', 'apple') = 1<br>locate('app', 
'appapp', 2) = 4 |
+| lower(String str)<br>lcase(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回将 
'str' 的所有字母转换为小写的结果 | lower("ApPlE") = "apple" |
+| lpad(String str1, Integer len, String str2) | - 如果三个参数中任一为 NULL 或 'len' 小于 
0，返回 null<br>- 如果 'len' 小于或等于 'str1' 的长度，返回下标在 [0, 'len') 范围内的 'str1' 子字符串<br>- 
如果 'len' 长于 'str1' 的长度且 'str2' 为 ""，返回 ""<br>- 如果 'str2' 不为 ""，返回在字符串 'str1' 
末尾填充字符串 'str2' 使字符串长度为 'len' 的结果字符串 | lpad('he',7,'xxd') = 
"xxdxxhe"<br>lpad('he',1,'') = "h" |
+| ltrim(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回去除前导空格的 'str' 字符串 | 
ltrim(' in long ') = "in long " |
+| parse_url(String urlStr, String keyword[, String parameter]) | - 如果任何必需参数为 
NULL，返回 NULL<br>- 从 'urlStr' 返回指定部分。'keyword' 的有效值<br>注意：'keyword' 是以下之一 
('HOST', 'PATH', 'QUERY', 'REF', 'PROTOCOL', 'AUTHORITY', 'FILE', 
'USERINFO')。还可以通过将键作为第三个参数来提取 QUERY 中特定键的值 | 
parse_url('http://inlong.apache.org/p.php?k1=v1&k2=v2#Ref1', 'QUERY', 'k1') = 
v1<br>parse_url('http://inlong.apache.org/p.php?k1=v1&k2=v2#Ref1', 'PROTOCOL') 
= http<br>parse_url('k1=v1&k2=v2', 'QUERY','k2') = v2 |
+| printf(String strfmt [, Object obj, ...]) | - 从 printf 样式的格式字符串返回格式化字符串 | 
printf("User %s has %d points and a balance of %.2f.", "Bob", 1500, 99.99) = " 
User Bob has 1500 points and a balance of 99.99." |
+| regexp(String str, String regexp)<br>similar(String str, String 
regexp)<br>regexp_like(String str, String regexp) | - 如果任一参数为 NULL 或无效，返回 
""<br>- 如果 'str' 的任何（可能为空）子字符串匹配 Java 正则表达式 'regexp'，返回 TRUE | regexp("The 
quick brown fox", "quick") = true<br>regexp("The quick brown fox", "cold") = 
false |
+| regexp_substr(String source_string, String regexp) | - 如果任一参数为 NULL 或 regexp 
无效或模式未找到，返回 ""<br>- 返回 'str' 中匹配 'regexp' 的第一个子字符串 | regex_substr("abc123def", 
"(\\d+)") = 123 |
+| regexp_count(String str, String regexp) | - 如果任一参数为 NULL 或 'regexp' 无效，返回 
""<br>- 返回 'str' 匹配 'regexp' 模式的次数<br>注意：'regexp' 必须是 Java 正则表达式 | 
regexp_count("The quick brown fox quick", "quick") = 2 |
+| regexp_extract(String str, String regexp[, Integer extractIndex]) | - 
如果任一参数为 NULL 或无效，返回 ""<br>- 使用指定的正则表达式 'regexp' 和正则表达式匹配组索引 'extractIndex' 从 
'str' 中提取字符串<br>注意：'regexp' 必须是 Java 正则表达式。'extractIndex' 指示要提取的正则表达式组，从 1 
开始，如果未指定则为默认值。0 表示匹配整个 'regexp' 表达式。此外，正则表达式匹配组索引不应超过定义的组数 | 
REGEXP_EXTRACT("abc123def", "(\\d+)", 1) = 123<br>REGEXP_EXTRACT("Name: John, 
Age: 25, Location: NY", "Name: (\\w+), Age: (\\d+), Location: (\\w+)", 2) = 
25<br>REGEXP_EXTRACT("abc123def", "(\\d+)", 2) = null [...]
+| regexp_extract_all(String str, String regexp[, Integer extractIndex]) | - 
如果任一参数为 NULL 或无效，返回 ""<br>- 返回由 'str' 中匹配 'regexp' 表达式并对应于 'extractIndex' 
正则表达式组的所有子字符串组成的数组<br>注意：'regexp' 必须是 Java 正则表达式。'extractIndex' 指示要提取的正则表达式组，从 
1 开始，如果未指定则为默认值。0 表示匹配整个 'regexp' 表达式 | 
REGEXP_EXTRACT_ALL("abc123def456ghi789", "(\\d+)", 0) = [123, 456, 
789]<br>REGEXP_EXTRACT_ALL("Name: John, Age: 25, Location: NY", "Name: (\\w+), 
Age: (\\d+), Location: (\\w+)", 1) = [John]<br>REGEXP_EXTRACT_ALL("Name: John 
[...]
+| regexp_instr(String str, String regexp) | - 如果任一参数为 NULL 或无效，返回 ""<br>- 返回 
'str' 中匹配 'regexp' 的第一个子字符串的位置 | regexp_instr("abc123def", "(\\d+)") = 4 |
+| regexp_matches(String source_string, String regexp[, String flags]) | - 
如果任一参数为 NULL 或无效，返回 ""<br>- 返回指定正则表达式 'regexp' 从 'source_string' 
的第一次匹配结果<br>注意：'flags' 是以下之一 ('g' -> 当我们想要匹配所有出现的子字符串时可以使用，'i' -> 忽略大小写匹配，'x' 
-> 扩展语法（忽略正则表达式中的空格和注释），'m' 和 'n' -> 允许正则表达式跨多行匹配） | regexp_matches("The quick 
brown fox", "quick") = [{"quick"}]<br>regexp_matches("foo 123 bar 456", "\\d+", 
"g") = [{"123"},{"456"}] |
+| regexp_replace(String source_string, String regexp, String replacement) | - 
如果任一参数为 NULL 或无效，返回 ""<br>- 返回将 'source_string' 中匹配正则表达式 'regexp' 的所有连续子字符串替换为 
'replacement' 的字符串 | regexp_replace('foobarbaz', 'b..', 'X') = "fooXbaz" |
+| regexp_split_to_array(String source_string, String regexp[, String flags]) | 
- 如果任一参数为 NULL 或无效，返回 ""<br>- 返回将 'source_string' 中匹配正则表达式 'regexp' 
的所有连续子字符串替换为 'replacement' 的字符串<br>注意：'flags' 是以下之一 ('g' -> 
当我们想要匹配所有出现的子字符串时可以使用，'i' -> 忽略大小写匹配，'x' -> 扩展语法（忽略正则表达式中的空格和注释），'m' 和 'n' -> 
允许正则表达式跨多行匹配） | regexp_split_to_array("hello world","\s+") = {hello, world} |
+| repeat(String str, Integer times)<br>replicate(String str, Integer times) | 
- 如果任一参数为 NULL，返回 ""<br>- 返回将 'str' 重复一定 'times' 次数的新字符串 | repeat('apple', 2) = 
"appleapple" |
+| replace(String s, String s1, String s2) | - 如果任一参数为 NULL，返回 ""<br>- 返回将字符串 
's' 中的字符串 's1' 替换为字符串 's2' 的结果 | replace('Hello World', '', 'J') = "JHJeJlJlJoJ 
JWJoJrJlJdJ" |
+| reverse(String str) | - 如果 'str' 为空字符串或 NULL，返回 ""<br>- 返回字符顺序反转的 'str' | 
reverse('apple') = "elppa" |
+| right(String str, Integer len) | - 如果 'str' 或 'len' 为 NULL，返回 ""<br>- 如果 
'len' 小于或等于零，返回 ""<br>- 返回 'str' 右侧开始的 len 长度的子字符串 | right('hello world',100) = 
"hello world" |
+| rpad(String str1, Integer len, String str2) | - 如果三个参数中任一为 NULL 或 'len' 小于 
0，返回 ""<br>- 如果 'len' 小于或等于 's1' 的长度，返回下标在 [0, 'len') 范围内的 's1' 子字符串<br>- 如果 
's2' 为 "" 且 'len' 长于 's1' 的长度，返回 ""<br>- 如果 's2' 不为 ""，返回在字符串 's1' 末尾填充字符串 's2' 
使字符串长度为 'len' 的结果字符串 | rpad('he',1,'xxd') = "h"<br>rpad('he',7,'') = "" |
+| rtrim(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回去除尾随空格的 'str' 字符串 | 
rtrim(' in long ') = " in long" |
+| soundex(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回表示 'str' 发音的四字符代码 | 
soundex('hello world') = "H464" |
+| space(Integer N) | - 如果 'N' 为 NULL 或小于或等于 0，返回 ""<br>- 返回由 'N' 个空格字符组成的字符串 | 
space(5) = "     " |
+| split_index(String str, String delimiter, Integer 
index)<br>splitindex(String str, String delimiter, Integer index) | - 
如果索引为负数或任一参数为 NULL，返回 ""<br>- 如果索引超出拆分后字符串的范围，返回 NULL<br>- 将 'str' 按 
'delimiter' 分割后，返回给定整数索引（从0开始）处的字符串 | split_index('a,b,c', ',', 1) = "b" |
+| startswith(String s1, String s2) | - 如果任一参数为 NULL，返回 ""<br>- 返回 's1' 是否以 
's2' 开头 | startswith('Apache InLong', 'A') = true |
+| str_to_map(String str[, String pairDelimiter, String keyValueDelimiter]) | - 
如果 'str' 为 NULL，返回 ""<br>- 使用 'pairDelimiter'（默认是 ','）和 'keyValueDelimiter'（默认是 
'='）将 'str' 分割成键/值对后返回一个映射<br>注意：'pairDelimiter' 和 'keyValueDelimiter' 
都被视为正则表达式。因此，特殊字符（如 <([{^-=$!\|]})?*+.>）在用作字面分隔符前需要正确转义 | 
str_to_map('key1=value1,key2=value2,key3=value3') = {key1=value1, key2=value2, 
key3=value3}<br>str_to_map("name->John!age->30!city->China" , "!" , "->") = 
{name=John, age=30, city=China} |
+| strcmp(String s1, String s2) | - 如果任一参数为 NULL，返回 ""<br>- 如果字符串相同，返回 0<br>- 
如果根据当前排序顺序 's1' 小于 's2'，返回 -1<br>- 否则返回 1 | strcmp('hello world','banana') = 1 |
+| substring(String s1, Integer pos[, Integer len])<br>substr(String s1, 
Integer pos[, Integer len])<br>mid(String s1, Integer pos[, Integer len]) | - 
如果 'str' 为 NULL，返回 ""<br>- 返回 'str' 从位置 'pos' 开始、长度为 'len' 
的子字符串（默认到末尾）<br>注意：此函数也支持 "substring(str FROM pos [ FOR len ])" 语法 | 
substring('apple', 1, 3) = "app" |
+| substring_index(String str, String delim, Integer count) | - 如果任一参数为 NULL，返回 
""<br>- 如果 'count' 为正数，返回从左边开始计数，最后一个 'delim' 出现次数之前的全部内容<br>- 如果 'count' 
为负数，返回从右边开始计数，最后一个 'delim' 出现次数之后的全部内容 | SUBSTRING_INDEX('AA. ', ' ', 1) = 
"AA." |
+| translate(String origin_string, String find_chars, String replace_chars) | - 
如果任一参数为 NULL，返回 ""<br>- 返回将 'origin_string' 中所有出现的 'find_chars' 中的字符替换为 
'replace_chars' 中对应字符的结果 | translate('apa...@inlong.com', '@', '.') = 
"apache.inlong.com"<br>translate('hello WorD', 'WD', 'wd') = "hello word" |
+| trim(String str)<br>btrim(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 返回删除 
'str' 前后空格的结果 | trim(' in long ') = "in long" |
+| unhex(String str) | - 如果 'str' 为 NULL，返回 ""<br>- 
返回将参数中的每对字符解释为其十六进制数对应的字符的结果 | unhex("696E6C6F6E67") = "inlong" |
+| upper(String s)<br>ucase(String s) | - 如果 's' 为 NULL，返回 ""<br>- 返回将 's' 
转换为大写的结果 | upper("ApPlE") = "APPLE" |
+| url_decode(String str[, String charset]) | - 如果 'str' 为 
NULL，或解码过程有问题（例如遇到非法转义模式），或编码方案不受支持，返回 ""<br>- 使用字符集（默认：UTF-8）编码方案解码给定的 
'application/x-www-form-urlencoded' 格式的 'str' | 
url_decode('https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+encode') = 
"https://apache.inlong.com/search?q=java url 
encode"<br>url_decode('https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+encode','UTF-8')
 = "https://apache.inlong.com/search?q=java url encode" |
+| url_encode(String str[, String charset]) | - 如果 'str' 为 
NULL，或编码过程有问题，或编码方案不受支持，返回 ""<br>- 使用字符集（默认：UTF-8）编码方案将 'str' 转换为 
'application/x-www-form-urlencoded' 格式 | 
url_encode('https://apache.inlong.com/search?q=java url encode') = 
"https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+encode"<br>url_encode('https://apache.inlong.com/search?q=java
 url encode','UTF-8') = 
"https%3A%2F%2Fapache.inlong.com%2Fsearch%3Fq%3Djava+url+encode" |
+| uuid() | - 返回基于 RFC 4122 类型 4（伪随机生成）的 UUID（通用唯一标识符）字符串 | uuid() |
+
+## Temporal (时间日期函数)
+
+| 函数 | 描述 | 示例 |
+|------|------|------|
+| convert_tz(String datetime_str, String from_tz, String to_tz) | - 如果任一参数为 
NULL，返回 NULL<br>- 将日期时间字符串 'datetime_str'（默认 ISO 时间戳格式 yyyy-MM-dd HH:mm:ss）从时区 
'from_tz' 转换为时区 'to_tz'<br>注意：时区格式应为缩写（如 "PST"）、全名（如 "America/Los_Angeles"）或自定义 
ID（如 "GMT-08:00"） | CONVERT_TZ('1970-01-01 00:00:00', 'UTC', 
'America/Los_Angeles') = "1969-12-31 16:00:00" |
+| date_add(String dateStr, String intervalExprStr) | - 如果 'dateStr' 为 NULL，返回 
""<br>- 如果 'dateStr' 是 DATE 值且计算仅涉及 YEAR, MONTH, DAY 部分（即不涉及时间部分），返回 DATE 
类型<br>- 如果 'dateStr' 是 TIME 值且计算仅涉及 HOURS, MINUTES, SECONDS 部分（即不涉及日期部分），返回 
TIME 类型<br>- 如果第一个参数是 DATETIME（或 TIMESTAMP）值，或者第一个参数是 DATE 但单位值使用了 HOURS, 
MINUTES, SECONDS，或者第一个参数是 TIME 类型但单位值使用了 YEAR, MONTH, DAY，返回 DATETIME 类型<br>- 
否则返回 String 类型（VARCHAR）<br>注意：关于 intervalExpr，请参考 MySQL 官网 | 
date_add('2020-12-31 23:59:59',INTERVAL 999 DAY) =  [...]
+| date_format(String timestampStr, String formatStr) | - 如果任一参数为 NULL，返回 
""<br>- 将时间戳（以秒为单位）转换为指定格式的日期格式字符串<br>注意：格式字符串与 Java 的 SimpleDateFormat 兼容 | 
date_format('2024-08-01 22:56:56', 'yyyy/MM/dd HH:mm:ss') = "2024/08/01 
22:56:56" |
+| date_sub(String dateStr, String intervalExprStr)<br>datesub(String dateStr, 
String intervalExprStr) | - 如果 'dateStr' 为 NULL，返回 ""<br>- 如果日期参数是 DATE 值且计算仅涉及 
YEAR, MONTH, DAY 部分（即不涉及时间部分），返回 DATE 类型<br>- 如果日期参数是 TIME 值且计算仅涉及 HOURS, 
MINUTES, SECONDS 部分（即不涉及日期部分），返回 TIME 类型<br>- 如果第一个参数是 DATETIME（或 
TIMESTAMP）值，或者第一个参数是 DATE 但单位值使用了 HOURS, MINUTES, SECONDS，或者第一个参数是 TIME 
类型但单位值使用了 YEAR, MONTH, DAY，返回 DATETIME 类型<br>- 否则返回 String 类型（VARCHAR）<br>注意：关于 
'intervalExpr'，请参考 MySQL 官网 | DATE_SUB('19 [...]
+| datediff(String dateStr1, String dateStr2) | - 如果两个参数中有一个为 null 或 ""，返回 
""<br>- 如果两个参数中有一个的日期格式不正确，返回 ""<br>- 返回日期 'dateStr1' 和 'dateStr2' 
之间的天数差（dateStr1 - dateStr2） | datediff('2018-12-10 12:30:00', '2018-12-09 
13:30:00') = 1<br>datediff('2018-12', '2018-12-12') = "" |
+| day_name(String dateStr)<br>dayname(String dateStr) | - 如果 'dateStr' 为 
null，返回 ""<br>- 从 'dateStr' 返回星期几的名称 | dayname('2024-02-29') = THURSDAY |
+| day_of_month(String dateStr)<br>dayofmonth(String dateStr) | - 如果 'dateStr' 
为 null，返回 ""<br>- 从 'dateStr' 返回一个月中的第几天（1 到 31 之间的整数） | 
dayofmonth('2024-02-29') = 29 |
+| day_of_week(String dateStr)<br>dayofweek(String dateStr) | - 如果 'dateStr' 为 
null，返回 ""<br>- 从 'dateStr' 返回一周中的第几天（1（星期日）到 7（星期六）之间的整数） | 
dayofweek('2024-02-29') = 5 |
+| day_of_year(String dateStr)<br>dayofyear(String dateStr) | - 如果 'dateStr' 为 
null，返回 ""<br>- 从 'dateStr' 返回一年中的第几天（1 到 366 之间的整数） | dayofyear('2024-02-29') 
= 60 |
+| from_unix_time(Numeric unixtime[, String 
formatStr])<br>from_unixtime(Numeric unixtime[, String formatStr]) | - 如果 
'unixtime' 为 NULL，返回 ""<br>- 将数字参数表示为字符串值返回<br>注意：'formatStr' 默认为 'yyyy-MM-dd 
HH:mm:ss'。unixtime 是表示自 UTC '1970-01-01 00:00:00' 以来的秒数的内部时间戳值，例如由 
UNIX_TIMESTAMP() 函数生成的值 | from_unix_time(44) = "1970-01-01 08:00:44" |
+| hour(String timestamp) | - 如果 'timestamp' 为 null，返回 ""<br>- 从 SQL 
'timestamp' 返回一天中的小时（0 到 23 之间的整数） | hour('2024-08-12 12:23:34') = 12 |
+| localdate([String timeZoneStr])<br>currentdate([String 
timeZoneStr])<br>current_date([String timeZoneStr])<br>curdate([String 
timeZoneStr]) | - 返回指定时区 'timeZoneStr' 的当前日期 | localDate("UTC") = 
currentDate<br>localDate() = currentDate |
+| localtime([String timeZoneStr])<br>current_time([String timeZoneStr]) | - 
返回指定时区的当前时间 | localTime() = currentTime<br>currentTime("UTC") = currentTime |
+| minute(String timestamp) | - 如果 'timestamp' 为 null，返回 ""<br>- 从 SQL 
'timestamp' 返回一小时中的分钟（0 到 59 之间的整数） | minute('2024-08-12 12:23:34') = 23 |
+| month(String dateStr) | - 如果 'dateStr' 为 null，返回 ""<br>- 从 'dateStr' 
返回一年中的月份（1 到 12 之间的整数） | month('2024-08-08') = 8 |
+| now() | - 返回本地时区的当前 SQL 时间戳 | now() |
+| quarter(String dateStr) | - 如果 'dateStr' 为 null，返回 ""<br>- 从 'dateStr' 
返回一年中的季度（1 到 4 之间的整数） | quarter('2024-08-08') = 3 |
+| second(String timestamp) | - 如果 'timestamp' 为 null，返回 ""<br>- 从 SQL 
'timestamp' 返回一分钟中的秒数（0 到 59 之间的整数） | second('2024-08-12 12:23:34') = 34 |
+| timediff(String dateStr1, String dateStr2)<br>time_diff(String dateStr1, 
String dateStr2) | - 如果 'dateStr1' 或 'dateStr2' 为 NULL，并且 'dateStr1' 和 
'dateStr2' 的转换类型不同，返回 ""<br>- 返回 'dateStr1' - 'dateStr2' 
表示的时间值<br>注意：'dateStr1' 和 'dateStr2' 是转换为 TIME 或 DATETIME 表达式的字符串 | 
timediff('23:59:59.000001','01:01:01.000002') = "22:58:57.999998" |
+| timestamp(String datetime_expr1[, String datetime_expr2]) | - 如果 
'datetime_expr1' 或 'datetime_expr2' 为 NULL，返回 ""<br>- 如果只有一个参数：将日期或日期时间表达式 expr 
作为日期时间值返回<br>- 如果有两个参数：返回日期或日期时间表达式 'datetime_expr1' 加上时间表达式 'datetime_expr2' 
的结果 | timestamp('2003-12-31 12:00:00.600000','12:00:00') = "2004-01-01 
00:00:00.600000" |
+| timestamp_add(String unit, Integer interval, String 
baseDateStr)<br>timestampadd(String unit, Integer interval, String baseDateStr) 
| - 如果任一参数为 null，返回 ""<br>- 返回将整数表达式间隔 interval 添加到日期或日期时间表达式 'baseDateStr' 
的结果<br>注意：'unit' 是以下之一 (MICROSECOND, SECOND, MINUTE, HOUR, DAY, WEEK, MONTH, 
QUARTER, YEAR) | timestamp_add('MICROSECOND',3,'1970-01-01 00:00:44') = 
"1970-01-01 00:00:44.000003" |
+| timestamp_diff(String unit, String datetime_expr1, String 
datetime_expr2)<br>timestampdiff(String unit, String datetime_expr1, String 
datetime_expr2) | - 如果任一参数为 null，返回 ""<br>- 返回 'datetime_expr2' − 
'datetime_expr1'，其中 'datetime_expr1' 和 'datetime_expr2' 
是日期或日期时间表达式<br>注意：'unit' 是以下之一 (MICROSECOND, SECOND, MINUTE, HOUR, DAY, WEEK, 
MONTH, QUARTER, YEAR) | timestampdiff('MONTH','2003-02-01','2003-05-01') = 3 |
+| to_date(String str [,String format]) | - 如果 'str' 为 NULL，返回 ""<br>- 将日期字符串 
'str' 转换为指定 'format' 格式的日期（默认 'yyyy-MM-dd'） | to_date('20240815', 'yyyyMMdd') = 
"2024-08-15" |
+| to_timestamp(String str [,String format]) | - 如果 'str' 为 NULL，返回 ""<br>- 在 
'UTC+0' 时区下，将日期时间字符串 'str' 转换为指定 'format' 格式（默认未指定则为 yyyy-MM-dd HH:mm:ss）的时间戳 | 
to_timestamp('1970/01/01 00:00:44', 'yyyy/MM/dd HH:mm:ss') = "1970-01-01 
00:00:44.0" |
+| unix_timestamp([String dateStr [, String format]]) | - 如果未指定参数，返回当前的 Unix 
时间戳（秒）<br>- 如果指定了参数，将日期时间字符串 'dateStr' 转换为指定 'format' 格式（默认未指定则为 yyyy-MM-dd 
HH:mm:ss）的 Unix 时间戳（秒） | unix_timestamp('1970/01/01 08:00:44', 'yyyy/MM/dd 
HH:mm:ss') = 44 |
+| week(String dateStr) | - 如果 'dateStr' 为 null，返回 ""<br>- 从 'dateStr' 
返回一年中的第几周（1 到 53 之间的整数） | week('2024-02-29') = 9 |
+| year(String dateStr) | - 如果 'dateStr' 为 null，返回 ""<br>- 从 SQL 日期返回年份 | 
year('2024-08-08') = 2024 |
\ No newline at end of file