Parsing a line from a file.

[Rock Ridge Farm (Larry)]

I have an input file that uses '|' as the separator.
I want to parse it for storage in a database..
I thought of converting it to ',' and saving as a .csv but that does not work since
some of the fields contain ',' separated data.
Is there an easy way to do this?
A data line could look like:
aaa|bbb|ccc,ddd,eee,fff|ggg
Any help?

[Quentin]

you can use ";" instead of ",". I think this should work too for csv files.

[Rock Ridge Farm (Larry)]

Problem is that '|' is the separator and I still need a method to parse the file.

[Quentin]

not sure if I understand the problem.
Just read the file line by line and split the line into the fields

e.g. like this

Code Select Expand


const false = 0
const true = 1

lin$ = "aaa|bbb|ccc,ddd,eee,fff|ggg"
start = 1
finished = false

openconsole
while finished = false
  pos = instr(lin$, "|", start)
  field$ = mid$(lin$, start, pos - start)
  print field$
  start = pos + 1
  if pos = 0
    finished = true
  endif
wend
DO:UNTIL INKEY$ <> ""
closeconsole
end


sorry if I didn't understand the prob.

[LarryMc]

fletchies ctl.lib has a SplitTwo function that was designed to do exactly what you want to do.

I use it all the time.

Get it here: http://www.codingmonkeys.com/index.php?action=tpmod;dl=item108


Larry

[JoaoAfonso]

This comes from noteworx program that comes in the IB STD archive, at least: in order to separate data, try to use the character `. it is a `, not a ÂÃ,´ or a ' last one is apostrophe, first 2 are accents, just like ~ or ^, characters that almost for sure you wont use alone in the data you want to sabe in your file, so it will be good separators. In noteworx there is also a routine, to prevent wrong data separators, that before write data into file, it checks that string in search for ÂÃ,´ (accent) characters and if any found, it replaces by a ' (apostrophe).
Hope it helps.

[Jerry Muelver]

Here's a function I've been using since about 1997 in one form or another....

Code Select Expand


'------------------------------------------------------------------
SUB splitstr(tline as string, splitter as string),HEAP
'splits string tline on delimiter splitter
' returns split-off chunk, modifies (chops)
' original string as well
DEF tok as POINTER
tok = AllocHeap(64000)
DEF tokpos:INT
tline = LTRIM$(tline)
tokpos = INSTR(tline,splitter)
IF(tokpos = 0)
    #<STRING>tok = tline
    tline = ""
ELSE
    #<STRING>tok = LEFT$(tline,tokpos-1)
    tline = MID$(tline,tokpos+LEN(splitter))
ENDIF
RETURN #<STRING>tok
ENDSUB



[aurelCB]

Great code Jerry

[Jerry Muelver]

Thank you. You can find a couple more string-manipulation routines in the ibHash demo, including substring extraction by delimiters, and search-replace.

[aurelCB]

This is differant from i do in Abasic i will try and thanks for link.
zlatko

[Rock Ridge Farm (Larry)]

Can someone put ctl.lib on this site or point me to another location.
I can not get into C.M. It keeps making me change my PW and then
will not let me log in.

[Rock Ridge Farm (Larry)]

Never mind - I found another copy.