I have a dynamic internal table <ft_dyn_tab>. I want to cast each row of the internal table to the type string via the field symbol <lf_string>:
LOOP AT <ft_dyn_tab> ASSIGNING <fs_dyn_wa>.
ASSIGN <fs_dyn_wa> to <lf_string> CASTING.
...
"other logic
...
ENDLOOP.
Normally, CASTING works fine when all fields of the structure are of type character. But when one field is of type string, it gives a runtime error. Can anyone explain why? And how to resolve this issue?
Why a structure with only character-like and String components can't be "casted" as a text variable
The reason is given by the ABAP documentation of Strings:
"A structure that contains a string is a deep structure and cannot be used as a character-like field in the same way as a flat structure.".
and of Deep:
"Deep: [...] the content [...] is addressed internally using references ([...], strings..."
and of Memory Requirement for Deep Data Objects:
"The memory requirement for the reference is 8 byte. [...] In strings, [...] an implicit reference is created internally."
and of ASSIGN - casting_spec:
"If the data type determined by CASTING is deep or if deep data objects are stored in the assigned memory area, the deep components must appear with exactly the same type and position in the assigned memory area. In particular, this means that individual reference variables can be assigned to only one field symbol that is typed as a reference variable by the same static type."
Now, the reason why the compiler and the run time don't let you do that, is that if you cast a whole deep structure, you could change the 8-bytes reference to access any place in the memory, that could be dangerous (How dangerous is it to access an array out of bounds?) and very difficult to analyze the subsequent bugs. In all programming languages, as far as possible, the compiler prevents out-of-bounds accesses or the checks are done at run time (Bounds checking).
Workaround
Your issue happens at run time because you use dynamically-created data objects, but you'd have exactly the same issue at compile time with statically-defined data objects. Below is a simple solution with a statically-defined structure.
You can access each field of the structure and concatenate it to a string:
DATA: BEGIN OF dyn_wa,
country TYPE c LENGTH 3,
city TYPE string,
END OF dyn_wa,
lf_string TYPE string.
FIELD-SYMBOLS: <lf_field> TYPE clike.
dyn_wa = VALUE #( country = 'FR' city = 'Paris' ).
DO.
ASSIGN COMPONENT sy-index OF STRUCTURE dyn_wa TO <lf_field>.
IF sy-subrc <> 0.
EXIT.
ENDIF.
CONCATENATE lf_string <lf_field> INTO lf_string RESPECTING BLANKS.
ENDDO.
ASSERT lf_string = 'FR Paris'. " one space because country is 3 characters
RESPECTING BLANKS keeps trailing spaces, to mimic ASSIGN ... CASTING.
Sounds like you want to assign the complete structured row to a plain string field symbol. This doesn't work. You can only assign the individual type-compatible components of the structured row to the string field symbol.
Otherwise, this kind of assignment works fine. For a table with a single column with type string:
TYPES table_type TYPE STANDARD TABLE OF string WITH EMPTY KEY.
DATA(filled_table) = VALUE table_type( ( `Test` ) ).
ASSIGN filled_table TO FIELD-SYMBOL(<dynamic_table>).
FIELD-SYMBOLS <string> TYPE string.
LOOP AT <dynamic_table> ASSIGNING FIELD-SYMBOL(<row>).
ASSIGN <row> TO FIELD-SYMBOL(<string>).
ENDLOOP.
For a table with a structured row type:
TYPES:
BEGIN OF row_type,
some_character_field TYPE char80,
the_string_field TYPE string,
END OF row_type.
TYPES table_type TYPE STANDARD TABLE OF row_type WITH EMPTY KEY.
DATA(filled_table) = VALUE table_type( ( some_character_field = 'ABC'
the_string_field = `Test` ) ).
ASSIGN filled_table TO FIELD-SYMBOL(<dynamic_table>).
FIELD-SYMBOLS <string> TYPE string.
LOOP AT <dynamic_table> ASSIGNING FIELD-SYMBOL(<row>).
ASSIGN <row>-the_string_field TO <string>.
ENDLOOP.
I have just tested this and it gives runtime error also when the structure does not have any string typed field.
I change the ASSIGN to a simple MOVE to a string variable g_string and it fails with runtime. If this fail it means that such an assignment is not possible, so the casting will not be either.
REPORT ZZZ.
TYPES BEGIN OF t_test.
TYPES: f1 TYPE c LENGTH 2,
f2 TYPE n LENGTH 4,
f3 TYPE string.
TYPEs END OF t_test.
TYPES BEGIN OF t_test2.
TYPES: f1 TYPE c LENGTH 2,
f2 TYPE n LENGTH 4,
f3 TYPE c LENGTH 80.
TYPES END OF t_test2.
TYPES: tt_test TYPE STANDARD TABLE OF t_test WITH EMPTY KEY,
tt_test2 TYPE STANDARD TABLE OF t_test2 WITH EMPTY KEY.
DATA(gt_test) = VALUE tt_test( ( f1 = '01' f2 = '1234' f3 = `Test`) ).
DATA(gt_test2) = VALUE tt_test2( ( f1 = '01' f2 = '1234' f3 = 'Test') ).
DATA: g_string TYPE string.
FIELD-SYMBOLS: <g_any_table> TYPE ANY TABLE,
<g_string> TYPE string.
ASSIGN gt_test2 TO <g_any_table>.
ASSERT <g_any_table> IS ASSIGNED.
LOOP AT <g_any_table> ASSIGNING FIELD-SYMBOL(<g_any_wa2>).
* ASSIGN <g_any_wa2> TO <g_string> CASTING.
g_string = <g_any_wa2>.
ENDLOOP.
UNASSIGN <g_any_table>.
ASSIGN gt_test TO <g_any_table>.
ASSERT <g_any_table> IS ASSIGNED.
LOOP AT <g_any_table> ASSIGNING FIELD-SYMBOL(<g_any_wa>).
* ASSIGN <g_any_wa> TO <g_string> CASTING.
g_string = <g_any_wa>.
ENDLOOP.
Related
the thing is to replace whatever observation that hold the number 999.9 as NULL
but i dont know the utility of CAST here as i know it helps change the data type but WDSP in the description is STRING
IF(
wdsp="999.9",
NULL,
CAST(wdsp AS Float64)) AS wind_speed,
nothing yet the cast should it come before IF or within IF
The CAST function in being used to convert the wdsp column to a number (specifically FLOAT64) instead of returning it as a STRING (which you mention is the declared type). This is likely so that whatever is issuing the query can get the aliased column wind_speed as a numerical type that can be processed as a number and not a string.
I have a definition like this:
TYPES: BEGIN OF ty_workarea,
data1 TYPE string,
data2 TYPE string,
data3 TYPE string,
END OF ty_workarea.
DATA gt_data TYPE TABLE OF ty_workarea.
Fields (data1, data2, data3) don't have any column name when I output it (gt_data) in ALV (cl_salv_table). How can I put name for them?
And since they have empty column names, I'm unable to do this trick:
lo_columns = go_alv->get_columns( ).
lo_column = lo_columns->get_column( 'CURRENT_NAME' ).
lo_column->set_long_text( 'NEW_NAME' ).
The get_column method expects the name of the field it represents, not the current heading text. So
lo_column = lo_columns->get_column( 'DATA1' ).
should get you the column you want.
Alternatively, the class CL_SALV_COLUMNS_TABLE (the one behind the lo_columns object) also has a method get which returns an internal table with the names and corresponding CL_SALV_COLUMN objects of all the columns. This can be useful in a context where you don't know the names of the columns you want to modify.
Columns names are taken by default from dict. You can try to use Sap elements or if you want to have custom title , you may define your own data element.
e.g use type MATNR instead of STRING for DataX and the title will be displayed for that column.
My function module receives a table name and a column name at runtime.
I would like to get the length of the column: How many characters are allowed in the transparent table?
I used my favorite search engine and found RTTS.
But the examples in the documentation pass a variable to the RTTS method DESCRIBE_BY_DATA; in my case, I don't have a variable, I just have the type names in table_name and column_name.
How to get the length?
For retrieving the type of a given DDIC type only known at runtime, use the method DESCRIBE_BY_NAME. The RTTI length is always returned as a number of bytes.
Example to get the type of the column CARRID of table SFLIGHT (I know it's a column of 3 characters) :
cl_abap_typedescr=>describe_by_name(
EXPORTING
p_name = 'SFLIGHT-CARRID'
RECEIVING
p_descr_ref = DATA(lo_typedescr)
EXCEPTIONS
type_not_found = 1 ).
" you should handle the error if SY-SUBRC <> 0
" Because it's SFLIGHT-CARRID, I expect 6 BYTES
ASSERT lo_typedescr->length = 6. " 3 characters * 2 bytes (Unicode)
" Length in CHARACTERS
CASE lo_typedescr->type_kind.
WHEN lo_typedescr->typekind_char
OR lo_typedescr->typekind_num
OR lo_typedescr->typekind_date
OR lo_typedescr->typekind_time
OR lo_typedescr->typekind_string.
DATA(no_of_characters) = lo_typedescr->length / cl_abap_char_utilities=>charsize.
ASSERT no_of_characters = 3.
ENDCASE.
You don't need RTTS for this. You can Do one of this
Select table DD03L with TABNAME and FIELDNAME
Use Function DDIF_FIELDINFO_GET
I want to split my code into a workarea and append this to a an internal table for a later perform.
But sometimes the text contains more than 3 numbers for example 3;5;3;6;2;5 but its always 3,6,9,12... number. How can I solve the problem that I want to loop 3 times, then the next 3 numbers and so on?
DATA: text(100) type c,
it_1 TYPE STANDART TABLE LIKE text,
it_2 TYPE STANDART TABLE LIKE text,
it_3 TYPE STANDART TABLE LIKE text,
string(100) TYPE c.
text = '123;2;2'.
SPLIT text AT ';' INTO wa_1-c1 wa_1-c2 wa_1-c3.
APPEND wa_1-c1 to it_1.
APPEND wa_1-c2 to it_2.
APPEND wa_1-c3 to it_3.
LOOP at it_1 INTO string.
PERFORM task using string.
ENDLOOP.
You should use the INTO TABLE addition to the split keyword rather than hard coding the fields.
DATA: text_s TYPE string.
text_s = '123;2;2'.
DATA: text_tab TYPE TABLE OF string.
SPLIT text_s AT ';' INTO TABLE text_tab.
LOOP AT text_tab ASSIGNING FIELD-SYMBOL(<line>).
"do whatever on each token here
ENDLOOP.
This will split the string in 3-er blocks, while overwriting it with the rest:
WHILE text IS NOT INITIAL.
SPLIT AT ';'
INTO wa_1-c1
wa_1-c2
wa_1-c3
text.
APPEND: wa_1-c1 to it_1,
wa_1-c2 to it_2,
wa_1-c3 to it_3.
ENDWHILE.
Please note, the string variable text will be initial at the end, if its original value is still needed, than you can define another string, copy the value and use that one for the split.
You can try using Sy-tabix if you want to control the iterations three times and since you are saving the text values in 3 different internal tables.
DATA: text(100) type c,
it_1 TYPE STANDARD TABLE OF text,
it_2 TYPE STANDARD TABLE OF text,
it_3 TYPE STANDARD TABLE OF text,
string(100) TYPE c.
text = '123;2;2'.
SPLIT text AT ';' INTO TABLE it_1.
LOOP at it_1 INTO string WHERE sy-tabix = 3.
WRITE : string.
ENDLOOP.
if sy-tabix = 3.
LOOP AT it_2 INTO string WHERE sy-tabix = sy-tabix+3.
"do the next loop
ENDLOOP.
ENDIF.
I'm using the conditional operator COND in a string template to create a string to concatenate a variable Alpha and Beta. If Alpha is initial the string is initial but if it isn't the result should be 'Alpha' if Beta is initial but 'Alpha-Beta' if Beta is not initial.
This is the code:
DATA:
lv_alpha TYPE string VALUE 'Alpha',
lv_beta TYPE string VALUE 'Beta',
lv_result TYPE string.
lv_result = COND #( WHEN lv_alpha IS INITIAL THEN '' ELSE
|{ lv_alpha }{ COND #( WHEN lv_beta IS INITIAL THEN '' ELSE |-{ lv_beta }| ) }|
).
The problem now is that when I run this, lv_result actually returns: 'Alpha-', so the value of lv_beta is not included, even though I know that the code entered the second part of the inner COND statement. What is causing this?
You see this behaviour when you use # as the operand type and the result of the type inference doesn't match what you want it to.
If you look at the documentation on Type Inference for Actual Parameters it explains:
If a constructor expression COND|SWITCH #( ... THEN ... ) is passed to generically typed formal parameters as an actual parameter using the character # as a symbol for the operand type, the following type inference is performed for the character #:
If the data type of the operand after the first THEN is identifiable statically and matches the generic type of the formal parameter, this data type is used.
If the data type of the operand after the first THEN is identifiable statically and does not match the generic type of the formal parameter or of it is not identifiable statically, the type is derived from the generic type as follows:
The fact that you're seeing this on the inner COND and not the outer one is because the outer type inference looks to the type of lv_result, a string. The inner COND has no direct type to look at so instead uses the statically identifiable type '', defaulting to a CHAR(1).
You can easily observe this behaviour by running the following commands. The output of each is behind the ".
DATA(lv_test) = 'ABCDEFG'.
WRITE :/ COND #( WHEN lv_alpha IS INITIAL THEN '' ELSE lv_test ). " A
WRITE :/ COND #( WHEN lv_alpha IS INITIAL THEN '123' ELSE lv_test ). " ABC
WRITE :/ COND string( WHEN lv_alpha IS INITIAL THEN '' ELSE lv_test ). " ABCDEFG
In both of the first commands, the type is derived from the first operand after the THEN and as a result the length of the output is limited based on the length of that first operand. The last line correctly specifies that we want to see a string as the result and ignores whatever type the first operand has. You're more likely to see this in nested statements because of how the type is inferred. Most programmers always use fully typed variables which means that errors like in the first two statements here are rare.
The # operand type is very convenient and certainly speeds up programming but you need to keep in mind what it's actually doing or risk running into strange bugs like these. These last examples show what happens when you rely on an unspecified type and you're passing in a value of a different type than what you're expecting, but the result won't always be this obviously incorrect, especially since this can happen with all data types, not just CLIKEs. Especially in nested statements where you can't view the type of the intermediate objects, this can be very difficult to trace.
DATA: lv_date TYPE dats.
WRITE :/ COND #( WHEN lv_alpha IS INITIAL THEN lv_date ELSE lv_test ). " G .EF.ABCD
WRITE :/ COND dats( WHEN lv_alpha IS INITIAL THEN '' ELSE lv_test ). " G .EF.ABCD
WRITE :/ COND string( WHEN lv_alpha IS INITIAL THEN lv_date ELSE lv_test ). " ABCDEFG
WRITE :/ COND string( WHEN lv_alpha IS INITIAL THEN '' ELSE lv_test ). " ABCDEFG