Using INPUTCSV

[pandawdy]

Is it possible to select lines in a CSV file to load?

For example, lets say this is the CSV file I'm working with:
one, two, three
four, five, six

When using INPUTCSV, is it possible to select the second line only? Or do all values have to be loaded first?

It seems like this should involve an array but I don't understand how to use arrays very well.

[Rod]

Spend some time reading and playing with the RAF file handling example in the help file. That lets you put and get individual lines of data from deep within a file.

[alincon]

Have you studied the inputcsv command in the LB help file?
The examples make it pretty clear how to get any particular line from a csv file.

r.m.

[pandawdy]

Jul 16, 2019 14:40:30 GMT -5 alincon said:

Have you studied the inputcsv command in the LB help file?
The examples make it pretty clear how to get any particular line from a csv file.

r.m.

Yes I did read the help file. I must have been missing something. Still don't see it. 

[metro]

It is a bit hard to find because it is under new features in LibertyBasic 4.5.0

I think the online help needs to be updated  to match the help document you have with your install.

INPUTCSV

INPUTCSV #handle, var[, var2[, varn ] ]

Description:

Read a comma separated (or delimited) value or values from the file for #handle. This is a new feature for Liberty BASIC v4.5.0.

This allows your program to read a line of text where the text has commas to separate the different values, like the regular INPUT #handle statement does but it also pays attention to quotation marks. This makes it possible to read files that have commas between the quotation marks and those commas are part of the data.

For example if you have a data item which is last name, first name format it can be read using this command. Here is an example with a name, a phone number, and a date:

"Jones, Timothy", 444-6823, 04/01/1970

You can read this data with the following statement:

INPUTCSV #handle, memberName$, phone$, birthday$

Usage:

The example below writes these two lines of text containing 4 comma separated items. Notice the quotation marks.

this,"is, a test, of"
"the emergency",broadcast system

Then it reads those 4 items and prints them out one line each.

open "somefile.txt" for output as #writeFile
print #writeFile, "this,";chr$(34);"is, a test, of";chr$(34)
print #writeFile, chr$(34);"the emergency";chr$(34);",broadcast system"
close #writeFile

open "someFile.txt" for input as #readFile
inputcsv #readFile, a$, b$, c$, d$
print a$
print b$
print c$
print d$
close #readFile
end


See also: INPUT$(#h, n), INPUTTO$(#h, c$), Line Input, INPUT

[metro]

down here we would say you're one kangaroo short in the top paddock nah only joking

for me I would either use an array and line input the data
you can then use WORD$ to find your match within your array based upon your search criteria

or for huge data sets insert them into a SQlite DB and search for s specific field/record

Arrays:
Line Input now allows input directly multiple variables and arrays.
j=0 : i=0
filedialog "Open","*.txt",file$
if file$="" then end
dim a$(3000)

open file$ for input as #f

while not(eof(#f))
   line input #f, a$(i)
   i=i+1
wend

close #f

for j=0 to i
   print a$(j) ' for the fifth line print a$(4)
next

end


[alincon]

I believe that inputcsv retrives as many items from a file as there are variable names following "inputcsv"
(if you have more variable names than file items, you'll get an error)
But you don't need an array, you can work with the variables directly; something like this:

If memberName$ = "Smith" then memberCount = memberCount + 1

r.m.

[metro]

Jul 16, 2019 20:40:27 GMT -5 alincon said:

I believe that inputcsv retrives as many items from a file as there are variable names following "inputcsv"
(if you have more variable names than file items, you'll get an error)
But you don't need an array, you can work with the variables directly; something like this:

If memberName$ = "Smith" then memberCount = memberCount + 1

r.m.

Now I'm curious,

If I am understanding the problem correctly pandawdy wants to choose a line from the CSV file, lets say line 3 ie a$(2)

I guess he could just iterate through the file till the index hits the line number he wants and just pull that line to print or manipulate.

But I assumed that this maybe a repeatable query of the same data file (maybe I should not assume) so therefore

having it sit in an array may be easier to repeat the process.

Or if he was looking for a line that contained JR (for example)  as the first field maybe the following would work

a$(0)="QW,AZ,PB,YG,LE,EW,TV,RB,NW,VZ,ER,TP,MF,ZS"
a$(1)="TE,TI,ZO,BU,KQ,UT,RH,IU,BG,YO,EF,SM,UW,QF"
a$(2)="JR,VH,IO,YI,CQ,DJ,JL,CO,WN,OU,OV,YT,YD,AY"
a$(3)="MI,JB,OW,TF,EM,UM,RB,YO,SH,YQ,BH,YW,LB,QN"

find$="JR"

FOR count = 0 to 3
if WORD$(a$(count),1,",")=find$ then
PRINT a$(count)
end if
next
FOR count = 0 to 3
if count = 2 then PRINT a$(count)

next





[Rod]

What if CSV was the wrong tool for the job?

[pandawdy]

I could have swore I wrote a response and it never showed up.. anyhow.. I think in this case a random access data file might work better. Appreciate all the help.