I've got something that works, but it seems like it would be "nicer" if I didn't have to use Table (or For, etc.).

I've got 3 Excel files, each with multiple sheets. Each file also has a sheet called "Paste Here" that I don't want. Each sheet contains an array. I'm reading in the file names from the folder, then from each file I read in the sheet names.

Now, I'd like to read the array from each sheet and store them all in a single (nested) list. I'm currently doing the following, which works fine:

pmlPath = ParentDirectory[NotebookDirectory[]] <> pmlFolder;
fileNames = Select[FileNames["*.xlsm", pmlPath], StringCount[#, "~"] == 0 &];
sheetNames = DeleteCases[(Import[#, "Sheets"] & /@ fileNames), "Paste Here", {2}];
pmls = Table[Import[fileNames[[i]], 
       {"Sheets", sheetNames[[i, j]]}], {i,Length[fileNames]}, {j,Length[sheetNames[[i]]]}];

It seems like I would have to use multiple slots to accomplish that, if it's possible, and I really don't know how to do that.

  • 3
    $\begingroup$ Maybe something like Import[#, {"Sheets", #2}] & @@@ Transpose[{fileNames, sheetNames}]? $\endgroup$
    – kglr
    Jun 11 '14 at 20:26
  • 2
    $\begingroup$ Or Import[#1, {"Sheets", #2}] & ~MapThread~ {fileNames, sheetNames} $\endgroup$ Jun 11 '14 at 20:33
  • 1
    $\begingroup$ Okay, I'll toss in: Inner[Import[#, {"Sheets", #2}] &, fileNames, sheetNames, List] $\endgroup$
    – Mr.Wizard
    Jun 11 '14 at 21:42
fileNames = "file" <> # & /@ RandomSample[CharacterRange["A", "Z"], 4]

(* {"fileS","fileU","fileJ","fileO"} *)

sheetNames = "sheet" <> # & /@ # & /@ (RandomSample[CharacterRange["A", "Z"], 
                                            RandomInteger[{1, 3}]] & /@ Range[4])

(* {{"sheetB","sheetR","sheetK"},{"sheetA"},{"sheetC","sheetO","sheetW"}, 

Use Import in place of foo in the following:

foo[#, {"Sheets", #2}] & @@@ Transpose[{fileNames, sheetNames}]

(* {foo["fileS",{"Sheets",{"sheetB","sheetR","sheetK"}}],
    foo["fileO",{"Sheets",{"sheetY","sheetU"}}]} *)
  • $\begingroup$ I don't understand the second "# & /@". I see that it doesn't work if I remove it. Denote "(RandomSample[CharacterRange" and everything to it's right as X. If I execute X, I get the same thing as when I execute # & /@ X. I really don't understand what's happening. $\endgroup$ Jun 12 '14 at 18:52
  • 1
    $\begingroup$ @MitchellKaplan, the function f="sheet"<>#& concatenates "sheet" with the input argument. (Btw you can get the same result using Map["sheet" <> # &, X, {2}] -- map f on X at level 2.) If you compare the outputs of f/@X and f/@#&/@X, you see that, when X is a list of lists, the first one applies f to the rows of X while the second applies it to the deeper elements inside each sublist. Hope this helps. $\endgroup$
    – kglr
    Jun 12 '14 at 22:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.