Thanks Jim and Eric for your replies.<div><br></div><div>Jim, perhaps info&#39;s version of option -d should be used in uniq&#39;s man page?</div><div><br></div><div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">


In isolation, uniq prints the last instance of the duplicated line, and<br></blockquote><div><br></div><div>I think it prints the first line, not the last:</div><div><br></div><div><div>printf &quot;a1\na2\na3\na4&quot; | uniq -w 1</div>

<div>a1</div></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
uniq -u suppresses the output of the 4th line.</blockquote><div><br></div><div>I have lost you here. -u suppresses any lines with duplicates.</div><div><br></div><div><div>printf &quot;a1\na2\na3\na4&quot; | uniq -u -w 1</div>

</div><div>(no output)</div><div><br></div><div>I suspect you mean -Du?</div><div><br></div><div><div>printf &quot;a1\na2\na3\na4&quot; | uniq -Du -w 1</div><div>a1</div><div>a2</div><div>a3</div></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">

  In isolation, -D says to<br>
output the first three lines which are normally omitted because they<br>
have duplicates, in addition to the 4th line that is printed by default.<br></blockquote><div><br></div><div>But,</div><div><br></div><div>printf &quot;a1\na2\na3\na4&quot; | uniq -D -w 1</div><div>a1</div><div>a2</div><div>

a3</div><div>a4</div><div><br></div><div>and default is this:</div><div><br></div><div><div>printf &quot;a1\na2\na3\na4&quot; | uniq -w 1</div><div>a1</div></div><div><br></div><div>So, if I understand your logic correctly and if I correct your logic by referring to the 1st and not the 4th duplicate then -Du should give me</div>

<div><br></div><div>a2</div><div>a3</div><div>a4</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
 So in combination, -Du says to print the lines with subsequent<br>
duplicates (the first three lines) but to suppress the output line that<br>
corresponds to the last input line that ends a sequence of duplicates<br>
(the 4th line).<br>
<br></blockquote><div><br></div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Perhaps we can document this behavior better.  Or perhaps we can change<br>
the behavior of -D (but at risk of breaking existing clients that depend<br>
on the current behavior).  But we can&#39;t change -u or -d behavior.<br>
<br></blockquote><div><br></div><div>I think changing behaviour of a utility is dangerous - I thought it was a bug, but both you and Jim have indicated that it is poor documentation.</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">


Put another way, per POSIX, the default behavior is subtractive (remove<br>
any line with a subsequent duplicate), -d is subtractive (remove any<br>
line with no duplicate), and -u is subtractive (remove any last line<br>
that had a prior duplicate), and GNU -D is additive (print any line with<br>
a subsequent duplicate, to counter the initial default).<br>
<br></blockquote><div><br></div><div>Whilst not exactly following your previous notes, I don&#39;t think this explains uniq&#39;s behaviour.</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">

At this point, I will claim that the behavior is intended, and therefore<br>
close out the bug.  But if you are willing to submit documentation<br>
patches, or even code patches accompanied by extensive test cases to<br>
demonstrate the corner cases of any new behavior, feel free to continue<br>
to reply to this bug report.<br>
<span class="HOEnZb"><font color="#888888"><br></font></span></blockquote><div> </div><div>Having now read info&#39;s uniq pages, I think that</div><div><br></div><div>1) -Du is undefined behaviour</div><div>2) current output makes no sense.</div>

<div><br></div><div>eg.</div><div><br></div><div><div>printf &quot;1\na1\na2\na3\na4\n2&quot; | uniq -w 1</div><div>1</div><div>a1</div><div>2</div><div>(comply)</div></div><div><br></div><div><div>printf &quot;1\na1\na2\na3\na4\n2&quot; | uniq -u -w 1</div>

<div>1</div><div>2</div></div><div>(comply)</div><div><br></div><div><div>printf &quot;1\na1\na2\na3\na4\n2&quot; | uniq -D -w 1</div><div>a1</div><div>a2</div><div>a3</div><div>a4</div></div><div>(comply)</div><div><br>
</div>
<div><div>printf &quot;1\na1\na2\na3\na4\n2&quot; | uniq -Du -w 1</div><div>a1</div><div>a2</div><div>a3</div></div><div><br></div><div>I think this one makes no sense.</div><div><br></div><div>But... this behaviour IS exactly what I need so if this can be documented then I would be happy - others might not be.</div>

<div><br></div><div><br></div><div>On a side point, why can&#39;t the official info pages be automagically converted into man pages to avoid discrepancies?</div><div><br></div></div></div>