Subject Re: importing csv file From Mervyn Bick <invalid@invalid.invalid> Date Wed, 30 Jun 2021 10:37:18 +0200 Newsgroups dbase.getting-started On 2021/06/29 19:33, Mustansir Ghor wrote: > Dear Mervyn Sir > > I read your below remark on date but could not figured the code to add. Will you help me on this. > > That code will work with most .csv files.  If, however, you have to > import dates you may need to add code to format the date before it gets > saved to the .dbf file. > I've added some comments to a snippet of the code I posted before. ...... do while not f.eof()    cRead = f.gets(100000) //Fetch a line from the .csv file.    aRecord = new stringex(cRead).breakstring(";",true) //check separator in .csv file. It is normally a comma but, //because I use a comma as a decimal point, Excel made the .csv //file with ; as the separator.  Change the ";" to "," if your //.csv uses a comma as the separator. // //stringex() creates an array with each field from the .csv in //an element.  The array is the same size as the rowset's fields array. //The following code places the contents of each element in the array //created by stringex() into the corresponding field of the table. q.rowset.beginAppend()    for n = 1 to aRecord.size      q.rowset.fields[n].value = aRecord[n]    next    q.rowset.save() enddo ...... Where the fields in the table are numeric or character then the data in each element of the array created by stringex() can be placed directly into each field in the fields array. Where the field type in the table is DATE or DATETIME the string in the array created by stringex() needs to be converted to a DATE or a DATETIME type first.  The actual conversion depends on the format of the string in the .csv file.  This is why Gaetano has asked for a sample. It is extremely difficult to write a universal program to save data from a .csv file to a table.  It is, however, very easy to write a program to save data from a specific .csv file to a specific table using stringex() from the dUFLP.  The .csv file you were using previously only had three character fields.  There was, therefore, no need to deal with anything else.  If you have fields other than numeric or character to deal with it is easy enough to test for them and then convert the string for the .csv file to the appropriate type. q.rowset.beginAppend()    for n = 1 to aRecord.size      if q.rowset.fields[n].type = 'DATETIME'        //code to convert string to dtDATETIME        q.rowset.fields[n].value = dtDATETIME      elseif q.rowset.fields[n].type = 'DATE'        //code to convert string to dDATE        q.rowset.fields[n].value = dDATE      elseif q.rowset.fields[n].type = 'TIME'        //code to convert string to tTIME        q.rowset.fields[n].value = tTIME      else //all numeric and character field types can simply be saved        q.rowset.fields[n].value = aRecord[n]      endif    next    q.rowset.save() enddo Mervyn.