topic Re: CRLF issue in csv text fields in Application Development and Automation Discussions

CRLF issue in csv text fields

at09051 — Tue, 21 Jun 2022 06:32:39 GMT

Hi Experts,

I have a requirement in which I am uploading csv files in my abap program, the file has some text fields , for ex- supplier comment. records are separated by ||. user is writing comment in text field and pressing enter to write in new line. when I am uploading this file and split it into required internal table, the line is breaking from between( where the text field is) and next records are being generated in new row. do anyone hve any suggestions?

Re: CRLF issue in csv text fields

Sandra_Rossi — Tue, 21 Jun 2022 07:11:02 GMT

If I understand well, the user enters this::

my first line
my second line

And in the file:

my first line||my second line

Correct?

Then show your code and we'll tell you where you're wrong.

Re: CRLF issue in csv text fields

at09051 — Tue, 21 Jun 2022 07:43:36 GMT

actually file is like this..

abc|| my first line

my second line || cdef|| ghi and so on

the records in internal table should be like..

abc | my first line ##(crlf) my second line | cdef | ghi..

earlier I wrote a code..

CALL FUNCTION 'SCMS_XSTRING_TO_BINARY'

EXPORTING

buffer = iv_content

* APPEND_TO_TABLE = ' '

IMPORTING

output_length = gv_length

TABLES

binary_tab = gt_upload.

CALL FUNCTION 'SCMS_BINARY_TO_TEXT'

EXPORTING

input_length = gv_length

TABLES

binary_tab = gt_upload

text_tab = gt_str

EXCEPTIONS

failed = 1

OTHERS = 2.

LOOP AT gt_str INTO DATA(gs_str).

" split csv data to convert it into standard table format

SPLIT gs_str-line AT '||' INTO gs_data_tech-statebox

gs_data_tech-taskbox

gs_data_tech-validationbox

gs_data_tech-prioritybox

gs_data_tech-rfrno

gs_data_tech-nctext

gs_data_tech-msn

gs_data_tech-version

gs_data_tech-nctype

gs_data_tech-program

gs_data_tech-plant

gs_data_tech-state

gs_data_tech-ncopened

gs_data_tech-ncclosed

gs_data_tech-supplier

gs_data_tech-activeoss

gs_data_tech-taskopened

gs_data_tech-priority

gs_data_tech-productgrp

gs_data_tech-partdesc

gs_data_tech-partno

gs_data_tech-category

gs_data_tech-airbuscomment

gs_data_tech-liability

gs_data_tech-liabilitycat

gs_data_tech-liabilityval

gs_data_tech-validationstatus

gs_data_tech-validationcomment

gs_data_tech-suppliercomment

gs_data_tech-completiondate

gs_data_tech-prod_deadline_datetime

gs_data_tech-matavailabledate

gs_data_tech-affectedsubcomp

gs_data_tech-trackingno

gs_data_tech-currency

gs_data_tech-labourhrsoss

gs_data_tech-adminhrsoss

gs_data_tech-matcostsoss

gs_data_tech-shipcostsoss

gs_data_tech-datasource

gs_data_tech-notificationid

gs_data_tech-dossierid

gs_data_tech-tasknumber

gs_data_tech-qrpref

gs_data_tech-cfifilter

gs_data_tech-technical_fields_marker.

IF sy-subrc <> 0.

ELSE.

APPEND gs_data_tech TO gt_data_tech. " append lines of all csv files to one table

CLEAR gs_data_tech.

ENDIF.

ENDLOOP.

there should be 2 records only(one is header file).. but you see there are three

Re: CRLF issue in csv text fields

at09051 — Tue, 21 Jun 2022 07:45:31 GMT

Then I tried like this
DATA: ls_string TYPE string,

lv_rec TYPE i,

gv_rec TYPE i VALUE 0.

FIELD-SYMBOLS: <fs_val> TYPE any.

DATA: it_st TYPE STANDARD TABLE OF ts_data_tech,

wa_st TYPE ts_data_tech.

CALL FUNCTION 'HR_KR_XSTRING_TO_STRING'

EXPORTING

in_xstring = iv_content

IMPORTING

out_string = ls_string.

SPLIT ls_string AT '||' INTO TABLE it_text.

DESCRIBE TABLE it_text LINES lv_rec.

DATA(lv_count) = lv_rec / 43.

lo_struct ?= cl_abap_structdescr=>describe_by_data( gs_data_tech2 ).

DO lv_count TIMES.

APPEND INITIAL LINE TO gt_data_tech2 ASSIGNING FIELD-SYMBOL(<fs_data_tech>).

LOOP AT lo_struct->components ASSIGNING FIELD-SYMBOL(<fs_components>).

gv_rec = gv_rec + 1.

READ TABLE it_text INTO DATA(wa_text) INDEX gv_rec.

ASSIGN COMPONENT <fs_components>-name

OF STRUCTURE <fs_data_tech> TO <fs_val>.

<fs_val> = wa_text-line.

ENDLOOP.

ENDDO.

*_______________________________________
but for this to work I hve to put delimeter || at the end of every record.. but my manager said it is not a good solution