While reviewing, I just came across this suggestion:


What, exactly is the suggested edit? (This is the edited question, and this is the suggested edit.)

A full resolution screenshot is available, too.

  • possible duplicate of Why does this suggested edit have no diff? Commented Mar 1, 2012 at 9:53
  • 1
    Looks like another occurrence of that bug which was marked status-completed but not fixed properly: some whitespace-only edits make it past the 6-character filter (that is probably the real bug), then show up with blank diffs. Oh, could you link to the suggested edit, please? Commented Mar 1, 2012 at 9:55
  • @Gilles: Added. Commented Mar 1, 2012 at 11:03
  • 1
    It appears everything was edited. Out.
    – user1228
    Commented Mar 1, 2012 at 14:40
  • I agree with Won't. Upon closer inspection, it looks like this is not a duplicate of the question Gilles linked, but rather an edit that removes all of the content of the original post. That tracks well with the fact that it comes from an anonymous user. And that would, of course, make this not a bug with the diff system, but rather a bug with human users. Filed under "deferred, fixing awaiting development of future technology". Commented Mar 1, 2012 at 19:20
  • And what about the missing "buttons" for showing a rendered output diff or a markdown diff? If the entire answer was erased, shouldn't there still be a diff of it? Commented Mar 1, 2012 at 20:27

2 Answers 2



This is what happened:

The original title was,

how to  convert .apk files to java

The title sanitizer converted it to

how to convert .apk files to java

Anonymous in her infinite wisdom decided to perform a no-op edit.

The "make sure that something in the title actually changed" code was a bit broken and did not properly squelch spaces between word boundaries. When it converted the title of the question with the "new" proposed title, it thought it changed cause a space was added.

I fixed it.


"java covert to android" shown in your screenshot (and in my review window as well) was not there in the question's revisions before: https://stackoverflow.com/posts/4573812/revisions

Given the above, it looks safe to assume that this is what was suggested in the edit. From the context, it is likely that it tries to clarify that the question covers both directions – from Java to Android and the other way around.

The meaning of the edit is vague, its spelling is bad and there is no information visible besides the Edit Summary – as if the author didn't even bother to look at the edited answer's preview. All the above sounds like too low effort to me. The edit appeared to deserve a rejection.

As for what could cause changes that appear invisible, these look entirely possible to me. In the previous edit of my answer (rev 2) I tested one particular trick to achieve that – by adding <text in angle brackets>. The result looks pretty similar to what is shown in your screenshot. The edit summary is shown and there's nothing else besides.


If the problematic edit was indeed of <hidden in angle brackets>, then this seems to be a known issue covered in another feature request: Add a view to see the raw source of suggested edits

  • 4
    "java covert toandroid" That's the summary for the suggested edit, not the content that was edited. The point Linus is trying to make is... that nothing appears to have been changed. Even a source diff isn't available when normally it should be. Commented Mar 1, 2012 at 9:17
  • @BoltClock'saUnicorn thanks I corrected the answer
    – gnat
    Commented Mar 1, 2012 at 10:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .