The file contains:
|
#### 18) Deleting whole observations with wildcarded dimensions: all observations are deleted for all dimension combinations where: |
|
- row 2: DIM2=A |
|
- row 3: DIM2=B and DIM3=C |
|
|
|
STRUCTURE,STRUCTURE_ID,ACTION,DIM_2,DIM_3 |
|
dataflow,ESTAT:NA_MAIN(1.6.0),D,A,, |
|
dataflow,ESTAT:NA_MAIN(1.6.0),D,B,C, |
and also says:
the RFC specifies that all records must have an identical structure (determined column number)
The second and third lines of the snippet contain erroneous trailing commas:
| STRUCTURE |
STRUCTURE_ID |
ACTION |
DIM_2 |
DIM_3 |
(nothing; end of line) |
| dataflow |
ESTAT:NA_MAIN(1.6.0) |
D |
A |
(empty) |
(additional ,) |
| dataflow |
ESTAT:NA_MAIN(1.6.0) |
D |
B |
C |
(additional ,) |
These lines thus have six columns (the sixth is empty), while the header has five.
The file contains:
sdmx-csv/data-message/docs/sdmx-csv-field-guide.md
Lines 257 to 263 in 3f8211f
and also says:
The second and third lines of the snippet contain erroneous trailing commas:
,),)These lines thus have six columns (the sixth is empty), while the header has five.