From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 65742 invoked by alias); 6 Sep 2019 21:53:13 -0000 Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner@cygwin.com Mail-Followup-To: cygwin@cygwin.com Received: (qmail 65735 invoked by uid 89); 6 Sep 2019 21:53:13 -0000 Authentication-Results: sourceware.org; auth=none X-Spam-SWARE-Status: No, score=-4.5 required=5.0 tests=AWL,BAYES_00,GIT_PATCH_2,L_8BIT_MISMATCH,NORMAL_HTTP_TO_IP,NUMERIC_HTTP_ADDR,RCVD_IN_DNSWL_LOW,SPF_PASS autolearn=ham version=3.3.1 spammy=H*MI:smtp, Device, birth, reaction X-HELO: lb3-smtp-cloud9.xs4all.net Received: from lb3-smtp-cloud9.xs4all.net (HELO lb3-smtp-cloud9.xs4all.net) (194.109.24.30) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Fri, 06 Sep 2019 21:53:10 +0000 Received: from tmp.5AeDEL7Ske ([83.162.234.136]) by smtp-cloud9.xs4all.net with ESMTPSA id 6MAAiDuS59b7T6MABiatQH; Fri, 06 Sep 2019 23:53:05 +0200 Date: Fri, 06 Sep 2019 21:53:00 -0000 Message-ID: <8e310c59702b57501ca2fd80edf19d2e@smtp-cloud9.xs4all.net> From: Houder Reply-To: cygwin@cygwin.com To: cygwin@cygwin.com Subject: Solved. Odd, is it not? mkdir 'e:\' cannot be undone by rmdir 'e:\' ... Content-Type: text/plain; charset=UTF-8; format=fixed User-Agent: mua.awk 0.99 X-SW-Source: 2019-09/txt/msg00062.txt.bz2 To those still interested! :-P I expressed surprise that mkdir e:/ does NOT refer to the drive, but rmdir e:/ does. Likewise do ls, stat ... https://cygwin.com/ml/cygwin/2019-08/msg00334.html ( Odd, is it not? mkdir 'e:\' cannot be undone by rmdir 'e:\' ... ) Why could mkdir not be made symmetrical to rmdir w/ regard to e:/ ????? After glancing over path_conv::check() in winsup/cygwin/path.cc, https://cygwin.com/ml/cygwin/2019-08/msg00418.html ( Date: Fri, 30 Aug 2019 11:54:27 +0200 ) I decided to remove Eric B.'s code snippet (2009, 26th Sep) from mkdir() in winsup/cygwin/dir.cc, i.e. I decided to delete the following lines: if (isdirsep (dir[strlen (dir) - 1])) { /* This converts // to /, but since both give EEXIST, we're okay. */ char *buf; char *p = stpcpy (buf = tp.c_get (), dir) - 1; dir = buf; while (p > dir && isdirsep (*p)) *p-- = '\0'; } While I took a closer look at the source code, I found a BUG in path_conv::check() in winsup/cygwin/path.cc https://cygwin.com/ml/cygwin/2019-08/msg00418.html ( Date: Sun, 01 Sep 2019 19:38:11 +0200 ) if (dev.isfs ()) { //if (strncmp (path, "\\\\.\\", 4)) <==== 1171 if ( ! strncmp (path, "\\\\.\\", 4)) // <==== [1] .. [1] this code should be executed only if path == '\\.\' !! On September 3rd, I discovered that dropping Eric B.'s code snippet, would introduce a BUG: https://cygwin.com/ml/cygwin/2019-09/msg00015.html ( Date: Tue, 03 Sep 2019 10:39:54 +0200 ) 64-@@ ln -s aap noot .. 64-@@ rmdir aap 64-@@ mkdir noot mkdir: cannot create directory ‘noot’: File exists 64-@@ mkdir noot/ <==== Whao! So that is what Eric indicated in his commit! 64-@@ ls -ld aap <==== WRONG! WRONG! drwxr-xr-x+ 1 Henri None 0 Sep 3 10:28 aap Different from Posix, Linux does not allow the creation of the directory aap ... (btw, neither should rmdir delete an existing directory aap if noot/ is specified) While waiting for a reaction by Eric Blake, I decided to take a closer look at path_conv::check() ... Could a solution be found in this method? (path arguments to (all?) commands are processed by this method) Basically, this method consists of a 'double loop', as follows: for (;;) // outer loop for (;;) // inner loop - the inner loop tests whether or not a path component is a symlnk - if it is, the outer loop is reentered, where the symlnk part of the path is replaced by the target - finally, the algorithm bails out of both loops if a "real" path is found (or not) Or something very near to this explanation ... In case the last component is a symlnk, the name of the symlnk is saved internally if the path had not been specified w/ a trailing slash. Otherwise the name of the target is saved internally. In short, there is a basic difference between specifying a path w/ a trailing slash or not ... I decided that modifying this method was not an option; I turned my attention again to mkdir() in winsup/cygwin/dir.cc To make mkdir behave as in Linux, and in case the last component is a symlnk, the name of the symlnk must be saved internally, not the name of the target. This was achieved by stripping trailing slashes ... The same patch should have been applied to rmdir ... However an exception can be made for e:/ (or e:\), as follows: -- char flag = '\0'; // strip trailing dirsep's, while remembering the last one if (isdirsep (dir[strlen (dir) - 1])) { flag = dir[strlen (dir) - 1]; /* This converts // to /, but since both give EEXIST, we're okay. */ char *buf; char *p = stpcpy (buf = tp.c_get (), dir) - 1; dir = buf; while (p > dir && isdirsep (*p)) { flag = *p; *p-- = '\0'; } } // reattach dirsep in case of x: and flag != '\0' if ( (strlen (dir) == 2) && (dir[1] == ':') && isalpha (dir[0]) && flag != '\0' ) { char *buf = tp.c_get (); buf[0] = dir[0]; buf[1] = ':'; buf[2] = flag; buf[3] = '\0'; dir = buf; } -- I applied the above patch to both mkdir() and rmdir() ... Henri ----- Tests: 3.1.0-0.2-patched => both mkdir and rmdir/ dir.cc have been modified (plus bug removed in path.cc) 64-@@ ln -s aap noot 64-@@ ls -l noot lrwxrwxrwx 1 Henri None 3 Sep 5 10:39 noot -> aap 64-@@ stat noot File: noot -> aap Size: 3 Blocks: 1 IO Block: 65536 symbolic link Device: 33d91880h/869865600d Inode: 23643898043927143 Links: 1 Access: (0777/lrwxrwxrwx) Uid: ( 1000/ Henri) Gid: ( 513/ None) Access: 2019-09-05 10:39:48.879895700 +0200 Modify: 2019-09-05 10:39:48.879895700 +0200 Change: 2019-09-05 10:39:48.879895700 +0200 Birth: 2019-09-05 10:39:48.879895700 +0200 64-@@ ls -l noot/ ls: cannot access 'noot/': No such file or directory 64-@@ stat noot/ stat: cannot stat 'noot/': No such file or directory 64-@@ mkdir noot mkdir: cannot create directory ‘noot’: File exists 64-@@ mkdir noot/ mkdir: cannot create directory ‘noot/’: File exists 64-@@ rmdir noot rmdir: failed to remove 'noot': Not a directory 64-@@ rmdir noot/ rmdir: failed to remove 'noot/': Not a directory -- 64-@@ mkdir aap 64-@@ ls -l noot lrwxrwxrwx 1 Henri None 3 Sep 5 10:39 noot -> aap 64-@@ stat noot File: noot -> aap Size: 3 Blocks: 1 IO Block: 65536 symbolic link Device: 33d91880h/869865600d Inode: 23643898043927143 Links: 1 Access: (0777/lrwxrwxrwx) Uid: ( 1000/ Henri) Gid: ( 513/ None) Access: 2019-09-05 10:39:48.879895700 +0200 Modify: 2019-09-05 10:39:48.879895700 +0200 Change: 2019-09-05 10:39:48.879895700 +0200 Birth: 2019-09-05 10:39:48.879895700 +0200 64-@@ ls -l noot/ total 0 64-@@ stat noot/ File: noot/ Size: 0 Blocks: 0 IO Block: 65536 directory Device: 33d91880h/869865600d Inode: 10977524091947662 Links: 1 Access: (0755/drwxr-xr-x) Uid: ( 1000/ Henri) Gid: ( 513/ None) Access: 2019-09-05 10:41:13.759644800 +0200 Modify: 2019-09-05 10:41:13.759644800 +0200 Change: 2019-09-05 10:41:13.759644800 +0200 Birth: 2019-09-05 10:41:13.759644800 +0200 64-@@ mkdir noot mkdir: cannot create directory ‘noot’: File exists 64-@@ mkdir noot/ mkdir: cannot create directory ‘noot/’: File exists 64-@@ rmdir noot rmdir: failed to remove 'noot': Not a directory 64-@@ rmdir noot/ rmdir: failed to remove 'noot/': Not a directory ===== -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple