Seen on 7-Zip [64] 16.00, windows 10 and p7zip Version 16.02 under Fedora.
So if we happen to have just an a.7z in the current directory and we issue simple "7z a a.7z" command
then new iteration of "a.7z" will devour former itself and so on.
p7zip process output:
Open archive: a.7z -- Path = a.**7z Type = 7z Physical Size = 111 Headers Size = 106 Method = LZMA2:12 Solid = - Blocks = 1 Scanning the drive: 1 file, 111 bytes (1 KiB) Updating archive: a.7z Items to compress: 1 Files read from disk: 1 Archive size: 258 bytes (1 KiB) Everything is Ok
List after some iterations:
$ 7z l a.7z
7-Zip [64] 16.02 : Copyright (c) 1999-2016 Igor Pavlov : 2016-05-21
p7zip Version 16.02 (locale=en_US.UTF-8,Utf16=on,HugeFiles=on,64 bits,2 CPUs Intel(R) Celeron(R) CPU G3900 @ 2.80GHz (506E3),ASM,AES-NI)
Scanning the drive for archives:
1 file, 428 bytes (1 KiB)
Listing archive: a.7z
--
Path = a.7z
Type = 7z
Physical Size = 428
Headers Size = 161
Method = LZMA2:12
Solid = -
Blocks = 2
Date Time Attr Size Compressed Name
2017-01-30 00:24:16 ....A 1 5 1
2017-01-30 00:24:31 ....A 258 262 a.7z
2017-01-30 00:24:31 259 267 2 files
Side note: midc's 7z plugin under fedora is supplying strange results with this: it present archive list without the copy's filename in it. just other files that's happens to be there also. Considering this list probably come from the 7z's lib then some problems are there too, but that's another story.
It's not bug.
Use
-x!
switch, if you don't like it.