I have encountered an algorithm problem for a long time and haven't solved it, help me

yunshenxin

The text document content is as follows

 // s1 s2 s3 s4 s5
20061217, D, B, DKDDA332021, ESA3332SS1
20061217, D, B, DKDDA332022, ESA3332SS2
20061217, D, B, DKDDA332023, ESA3332SS3
20061217, D, B, DKDDA332021, ESA3332SS2
20061217, D, B, DKDDA332025, ESA3332SS2
20061217, D, B, DKDDA332021, ESA3332SS3
20061217, D, B, DKDDA332022, ESA3332SS7

Among the text documents, s4 and s5 will have duplicate data. I want to delete all the duplicates.
But with SORT sort, you can only sort the entire row. Can you sort by s4 and s4, and delete the duplicate capacity (the entire row)! Thank you everyone

面面爱面面

Clarify the condition: "Delete the row as long as S4 and S5 are the same in a row" or "In all rows, where S4 is the same, only one row is left, and the remaining rows are deleted; where S5 is the same, only the left Delete one line and the rest "?

yunshenxin

Where S4 is the same, only one line is left, and the remaining lines are deleted; where S5 is the same, only one line is left, and the remaining lines are deleted

What do you mean

jk66255

The algorithm is roughly:

filelist: = TStringList.create;
filelist.loadfromfile (...);
newlist: = TStringList.Create;
s4list: = TStringList.Create;
s4list.sorted: = True; // sorting will speed up the search

for i: = 0 to filelist.count -1
begin
   // Separate a line of text file, extract s4,
.......

  if newlist.indexof (s4) <0 then
begin
  newlist.add (filelist [i]);
  s4list.add (s4);
end;
end;

newlist.savetofile (....)

jk66255

The above is to delete S4, S5 is similar