CDS031
This program can be used to convert a tab-separated, comma-separated,
or pipe-separated string into a string array. These CSV strings are often
created using a spreadsheet program or other data export tool. When used with the CHAN option, CDS031 can be used to read the data from the csv file, which provides support for field and line separators imbedded within quoted text strings. This enables quoted text strings to contain commas, tabs, carriage-return, and line-feed characters.
Calling Format:
CALL "CDS031", CSV$, CVS,
FLD$[ALL], FLDS {, FLDS$, CHAN }
| CDS031 |
CSV$
|
CHAN not passed |
Input string containing the separated fields. If passed as null then CDS031 will assume a single null string |
CHAN passed |
Output string read from file opened to channel CHAN. Will be null when end of file is reached. |
|
CVS
|
used by CDS031 with the CVS() function when assigning fields, and should
be set to the sum of the functions desired below. See CVS() in the
BBx documentation for additional information| 0 | no action | | 1 | strip leading blanks | | 2 | strip trailing blanks | | 4 | convert to upper case | | 8 | convert to lower case | | 16 | replace non-displayable characters with blanks | | 32 | replace multiple spaces with a single space | | 64 | replace "," and "." characters with values specified in OPTS |
|
FLD$[ALL]
|
will be returned by CDS031 with one element of the array for each field
in the record. The leading and trailing quotation marks will be
removed from field that has a quotation mark at the beginning and at the end
of the field. Since some export programs place quotes around strings,
they often store a quote contained within the field as two quotes.
CDS031 converts a pair of adjacent quote marks into a single quote mark.
The first field is FLD$[1] and the last field is FLD$[FLDS]. |
FLDS
|
set by CDS031 to the number of elements in the
FLD$[ALL] array. |
FLDS$
|
optional variable set by CDS031 which is similar to FLD$[ALL] except
that it is in template format. The first field is FLDS.FLD$[1] and the
last field is FLDS.FLD$[FLDS]. |
CHAN
|
Optional channel number which can be passed to CDS031 to have it perform the read operations to a string file used as input. Some CSV files include fields that have line feeds that within a quoted field indicating multiple lines of text. Since a line feed also separates records, filtering out these line feeds imbedded within a quotes field can be tricky. By passing CHAN instead of CSV$, CDS031 will read the file opened by the calling program to channel CHAN, and merge the multiple lines within a quoted field into a single field. You can use this option even when you do not expect to have line feeds imbedded within a single field.
When the end of the file is reached, CDS031 will return CSV$ as null.
|
CDS031 determines the separation character automatically, which can be a tab ($09$), a comma, or the pipe (|) symbol.
When using comma separated fields, the field can contain commas if the field
starts and ends with a quotation mark.
It is generally preferred to use a tab as the field separator, as unlike a
comma or the pipe symbol, a field cannot contain the tab character.
| Example using comma separator |
>CALL "CDS031","123,456,789",0,FLD$[ALL]
>WRITE FLD$[ALL]
123
456
789
|
| Example using pipe separator |
>CALL "CDS031","abc|d|efghi|jkl",4,FLD$[ALL],FLDS
>?FLDS
4
>WRITE FLD$[ALL]
ABC
D
EFGHI
JKL
|
| Example of a field that contains a pair of adjacent quotes |
>CSV$="12"""" NAILS,1.23,SMALL"
>?CSV$
12"" NAILS,1.23,SMALL
>CALL "CDS031",CSV$,0,FLD$[ALL],FLDS
>?FLDS
3
>WRITE FLD$[ALL]
12" NAILS
1.23
SMALL
|
| Example using comma separator with a field that contains commas |
>CSV$="""Cable, Parallel Printer, 6 ft"",7.50,33"
>?CSV$
"Cable, Parallel Printer, 6 ft",7.50,33
>call "CDS031",CSV$,0,FLD$[ALL],FLDS
>?FLDS
3
>WRITE FLD$[ALL]
Cable, Parallel Printer, 6 ft
7.50
33
|
