Sprintf equivalent in Mathematica?

wolfram-mathematica

Solution

I wouldn't use `PaddedForm` for this. In fact, I'm not sure that `PaddedForm` is good for much of anything. Instead, I'd use good old `ToString`, `Characters` and `PadLeft`, like so:

toFixedWidth[n_Integer, width_Integer] := 
  StringJoin[PadLeft[Characters[ToString[n]], width, "0"]]

Then you can use `StringForm` and `ToString` to make your file name:

toNumberedFileName[n_Integer] :=
  ToString@StringForm["filename_``", toFixedWidth[n, 5]]

Mathematica is not well-suited to this kind of string munging.

EDIT to add: Mathematica proper doesn't have the required functionality, but the `java.lang.String` class has the static method `format()` which takes `printf`-style arguments. You can call out to it using Mathematica's JLink functionality pretty easily. The performance won't be very good, but for many use cases you just won't care that much:

Needs["JLink`"];
LoadJavaClass["java.lang.String"];
LoadJavaClass["java.util.Locale"];
sprintf[fmt_, args___] :=
 String`format[Locale`ENGLISH,fmt,
  MakeJavaObject /@
   Replace[{args},
    {x_?NumericQ :> N@x,
     x : (_Real | _Integer | True | 
         False | _String | _?JavaObjectQ) :> x,
     x_ :> MakeJavaExpr[x]},
    {1}]]

You need to do a little more work, because JLink is a bit dumb about Java functions with a variable number of arguments. The `format()` method takes a format string and an array of Java `Object`s, and Mathematica won't do the conversion automatically, which is what the `MakeJavaObject` is there for.

Problem

I don't know why Wikipedia lists Mathematica as a programming language with printf. I just couldn't find the equivalent in Mathematica. My specific task is to process a list of data files with padded numbers, which I used to do it in bash with ``` fn=$(printf "filename_%05d" $n) ``` The closest function I found in Mathematica is `PaddedForm`. And after some trial and error, I got it with ``` "filename_" <> PaddedForm[ Round@#, 4, NumberPadding -> {"0", ""} ]& ``` It is very odd that I have to use the number 4 to get the result similar to what I get from "%05d". I don't understand this behavior at all. Can someone explain it to me? And is it the best way to achieve what I used to in bash?

Original source