Stata Programming help _ grouping do files by their size

[Myron Merton]

Dear all,

I am struggling with one task. I have lets say around 100 do files of different sizes (some are small, less than 1 mb, some are up to 126mb). I need to split all the files whose weight is more than 3 mb into smaller files with weight up to 3 mb and then be able to put them back together

It would be greatly appreciated if you could help me with some ideas

[William Lisowski]

Welcome to Statalist.

Are you sure you mean Stata do-files (suffix .do), which typically contain Stata commands, as opposed to Stata datasets (suffix .dat) which contain Stata data?

Because a 126mb do-file, if in ASCII, would contain about 1.5 million 80-character lines of Stata code, which, unless is was generated by some other program rather than typed in by an army of research assistants, is difficult to imagine.

[Jorrit Gosens]

cd "path where your files are stored"
local myfilelist: dir . files "*.dta"
foreach file of local myfilelist {
  use `"`file'"'
  local outfile = subinstr(`"`file'"',".dta","",.)
  local N = _N
  local i = 1
  forvalues obs = 1(150)`N' {
      local lastobs= `obs'+149
      if `lastobs'>`N'{
        local lastobs= `N'
        }
      preserve
      keep in `obs'/`lastobs'
      save "`outfile'part`i'"
      restore
      local ++i
      }
}