From nerijus at users.sourceforge.net Fri Mar 1 00:08:29 2002 From: nerijus at users.sourceforge.net (Nerijus Baliunas) Date: Fri, 1 Mar 2002 02:08:29 +0200 (EET) Subject: Problems with vfs / ftpfs In-Reply-To: <20020228174342.B3062@prog1> References: <20020227202729.A6604@prog1> <3C7E0132.7010603@it.efp.com.ua> <20020228174342.B3062@prog1> Message-ID: On Thu, 28 Feb 2002 17:43:42 +0400 Walery Studennikov wrote: WS> > >Cannot chmod target file "/mnt/d/file.txt" WS> > >Operation not permitted (1) WS> > WS> > You can unset "preserve Attributes" when copy. WS> > In January, 2002, I sent a patch to enable this feature for Move operation. WS> WS> This doesn't work for me. I've tried several versions, WS> now I'm using CVS version. The easiest thing is to mount vfat quiet. My /etc/fstab has: defaults,user,showexec,umask=0,quiet Regards, Nerijus From avarakin00 at hotmail.com Fri Mar 1 04:35:03 2002 From: avarakin00 at hotmail.com (Alexander Varakin) Date: Thu, 28 Feb 2002 23:35:03 -0500 Subject: Subshell hangs References: <20020227202729.A6604@prog1> <3C7E0132.7010603@it.efp.com.ua> <20020228174342.B3062@prog1> Message-ID: Hi, I am trying to port mc to cygwin. I was able to build it from CVS, then fixed problem in regex which was making mc useless. At the moment my main problem is that subshell, invoked from mc hangs very often. It hangs when a proces, invoked from shell is about to return and shell must display command prompt, but this doesn't happen. Control-O, Control-C, Control-Z, Control-D, Control-break don't help, and the only fix is just to close the terminal window with mc.... This also happens if just hit Enter few times. Interestingly enough this happens much more frequently while running mc in rxtvt or xterm, than while running from command prompt box. I attached gdb to mc when this happened, the call stack of mc is as follows: #0 0x77f827e8 in _libkernel32_a_iname () #1 0x77e86a15 in _libkernel32_a_iname () #2 0x77e86a3d in _libkernel32_a_iname () #3 0x6100df50 in _libkernel32_a_iname () #4 0x61058bfc in _libkernel32_a_iname () #5 0x0044b4cc in synchronize () at subshell.c:1098 #6 0x0044b32c in feed_subshell (how=1, fail_on_error=0) at subshell.c:1039 #7 0x0044a58b in invoke_subshell (command=0x0, how=1, new_dir=0x22fd40) at subshell.c:609 #8 0x0040a251 in view_other_cmd () at cmd.c:930 #9 0x0042dd8e in midnight_callback (h=0xa0168d0, id=15, msg=11) at main.c:1921 #10 0x0041212a in dlg_key_event (h=0xa0168d0, d_key=15) at dlg.c:707 #11 0x00412364 in dlg_process_event (h=0xa0168d0, key=15, event=0x22fe14) at dlg.c:827 #12 0x00412a9f in frontend_run_dlg (h=0xa0168d0) at dlg.c:859 #13 0x0041238d in run_dlg (h=0xa0168d0) at dlg.c:871 #14 0x0042e063 in setup_panels_and_run_mc () at main.c:1992 #15 0x0042e29e in do_nc () at main.c:2056 #16 0x004300e3 in main (argc=1, argv=0x61562fa8) at main.c:2710 #17 0x61003f8e in _libkernel32_a_iname () #18 0x61004282 in _libkernel32_a_iname () #19 0x610042c1 in _libkernel32_a_iname () #20 0x004a6edf in cygwin_crt0 () Attaching to hanging bash didn't give me any useful information since my bash doesn't have debug info. My bash is 2.05a-2 and cygwin.dll is 1.3.9. I noticed that ps marks the hanging bash with "S" in front of the record: S 496 1572 496 1600 1 500 22:44:19 /usr/bin/bash Any ideas what can be wrong? Regards, Alex From despair at sama.ru Sun Mar 3 17:47:40 2002 From: despair at sama.ru (Walery Studennikov) Date: Sun, 3 Mar 2002 12:47:40 -0500 Subject: Problems with vfs / ftpfs In-Reply-To: <20020228181953.A531@it.efp.com.ua> References: <20020227202729.A6604@prog1> <3C7E0132.7010603@it.efp.com.ua> <20020228174342.B3062@prog1> <3C7E4A58.8090800@cmail.ru> <20020228200000.A1027@prog1> <20020228181953.A531@it.efp.com.ua> Message-ID: <20020303124740.A8261@localhost.localdomain> On Thu, Feb 28, 2002 at 06:19:53PM +0200, Andrew V. Samoilov wrote: > > > >>>Cannot chmod target file "/mnt/d/file.txt" > > > >>>Operation not permitted (1) > > > So, how to solve / workaround this problem? > > Can you test patch below? It works, thank you. But what to do with the first problem: downloading the file from ftp before uploading it? -- Regards, Walery From pawel at sakowski.eu.org Sun Mar 3 14:47:59 2002 From: pawel at sakowski.eu.org (=?iso-8859-2?Q?Pawe=B3?= Sakowski) Date: 03 Mar 2002 15:47:59 +0100 Subject: Reget on file copying misreports progress Message-ID: <1015166879.31772.10.camel@saq> As in the subject. Let's: 1. Create /tmp/a with size of 200 MB (e.g. dd if=/dev/zero). 2. Create /tmp/b, 180 MB. 3. Copy (F5) /tmp/a => /tmp/b. 4. File exists, let's choose "Reget". Now the progress bar runs from 0% to 20%, then the copying ends. It seems that the progress is determined by dividing the amount of data transferred by the file size, which doesn't work properly when regetting. The algorithm should be changed in either of two ways: - data_transferred/(a_size-b_size), which gives progress between 0%-100% - (b_size+data_transferred)/a_size, which gives progress between 80%-100% I'm not sure which one is better. According to FAQ #9.3 I'm mentioning I'm not on the mailing list :) -- +--------------------------------------------------------------------+ | Avoid the Gates of Hell. : Pawel Sakowski | | Use Linux. : | +--------------------------------------------------------------------+ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 254 bytes Desc: not available URL: From sav at it.efp.com.ua Mon Mar 4 08:54:06 2002 From: sav at it.efp.com.ua (Andrew V. Samoilov) Date: Mon, 04 Mar 2002 10:54:06 +0200 Subject: Problems with vfs / ftpfs References: <20020227202729.A6604@prog1> <3C7E0132.7010603@it.efp.com.ua> <20020228174342.B3062@prog1> <3C7E4A58.8090800@cmail.ru> <20020228200000.A1027@prog1> <20020228181953.A531@it.efp.com.ua> <20020303124740.A8261@localhost.localdomain> Message-ID: <3C83362E.5080507@it.efp.com.ua> Walery Studennikov wrote: > On Thu, Feb 28, 2002 at 06:19:53PM +0200, Andrew V. Samoilov wrote: > >>>>>>>Cannot chmod target file "/mnt/d/file.txt" >>>>>>>Operation not permitted (1) >>>>>>> >>>So, how to solve / workaround this problem? >>> >> >>Can you test patch below? >> > > It works, thank you. > > > > But what to do with the first problem: > downloading the file from ftp before uploading it? > > I will try to fix this behavior some later. -- Regards, Andrew V. Samoilov. From mj at polcard.com.pl Mon Mar 4 17:07:07 2002 From: mj at polcard.com.pl (Mariusz Jedrzejewski) Date: Mon, 4 Mar 2002 18:07:07 +0100 Subject: Possible mistake, but maybe not... Message-ID: <20020304180706.A807@polcard.com.pl> Hello. I'm newbie at this list, and I've seen at archive, but doesn't find any answer for my problem. The problem is below: 1) I have any archive, eg. zip archive, archive contains one file: a.zip contains a.txt 2) When I explore this a.zip in MC, I see a file a.txt in panel 3) I'm looking at this a.txt with F3, I see anything. 4) At the same moment, in $TMP we have uncompressed a.txt with various name, like ext??????? 5) Now, the archive include CHANGED a.txt, but name of archive is still a.zip 6) I'm explore a.zip, see a.txt 7) Looking at a.txt, see *previous* version of file, readed from $TMP/ext?????? 8) But file was changed! I want to see actual version of file, nor unpacked previously to $TMP! 9) Also, when I delete a file from $TMP, and wants to see a.txt in archive, I see beautiful "input/output error", because in $TMP we haven't a ext?????? 10) After restart MC, everythink is ok. So, why MC (or something in it) don't read actual version of archive? It may change, and I'm in trouble thinking that view of file is correct. -- Greetingz from Poland, Mariusz Registered Linux User #206046 PS. I'm sorry for this ugly english, it isn't my native language... I hope you understand me... Any polish developer may contact me in polish language :-) From sav at it.efp.com.ua Tue Mar 5 08:47:20 2002 From: sav at it.efp.com.ua (Andrew V. Samoilov) Date: Tue, 05 Mar 2002 10:47:20 +0200 Subject: Possible mistake, but maybe not... References: <20020304180706.A807@polcard.com.pl> Message-ID: <3C848618.7030600@it.efp.com.ua> Mariusz Jedrzejewski wrote: > Hello. > I'm newbie at this list, and I've seen at archive, but doesn't find any answer > for my problem. > > The problem is below: > 1) I have any archive, eg. zip archive, archive contains one file: a.zip > contains a.txt > 2) When I explore this a.zip in MC, I see a file a.txt in panel > 3) I'm looking at this a.txt with F3, I see anything. > 4) At the same moment, in $TMP we have uncompressed a.txt with various name, > like ext??????? > 5) Now, the archive include CHANGED a.txt, but name of archive is still > a.zip > 6) I'm explore a.zip, see a.txt > 7) Looking at a.txt, see *previous* version of file, readed from > $TMP/ext?????? > 8) But file was changed! I want to see actual version of file, nor unpacked > previously to $TMP! > 9) Also, when I delete a file from $TMP, and wants to see a.txt in archive, I > see beautiful "input/output error", because in $TMP we haven't a ext?????? > 10) After restart MC, everythink is ok. > > So, why MC (or something in it) don't read actual version of archive? It may > change, and I'm in trouble thinking that view of file is correct. > > Please, read "Virtual FS" section in mc manual. -- Regards, Andrew V. Samoilov. From kai at cmail.ru Wed Mar 6 09:02:40 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Wed, 6 Mar 2002 11:02:40 +0200 Subject: [patch] Fix ftpfs uploading behavior. In-Reply-To: <20010908001446.A7576@localhost.localdomain> References: <20010908001446.A7576@localhost.localdomain> Message-ID: <20020306110240.A485@it.efp.com.ua> Hi! This patch fixes ftpfs uploading behavior. Now if file is opened for writting only it will be not retrieved to temporary file before uploading. This is first approximation. Test it carefully please. Regards, Andrew V. Samoilov. ChangeLog: * ftpfs.c (ftpfs_fh_open): Open data connection and remove localfile if file is opened for writing only. (ftpfs_fh_close): New function to close data connection opened in ftpfs_fh_open(). --- vfs/ftpfs.c Fri Feb 8 11:42:38 2002 +++ vfs/ftpfs.c Wed Mar 6 10:05:00 2002 @@ -1683,6 +1686,35 @@ static void my_forget (char *file) static int ftpfs_fh_open (vfs *me, vfs_s_fh *fh, int flags, int mode) { + if (((flags & O_WRONLY) == O_WRONLY) && !(flags & (O_RDONLY|O_RDWR))){ +#ifdef HAVE_STRUCT_LINGER + struct linger li; +#else + int li = 1; +#endif + char *name = vfs_s_fullpath (me, fh->ino); + + if (!name) + return -1; + fh->handle = open_data_connection(me, fh->ino->super, + (flags & O_APPEND) ? "APPE" : "STOR", name, TYPE_BINARY, 0); + g_free (name); + + if (fh->handle < 0) + return -1; +#ifdef HAVE_STRUCT_LINGER + li.l_onoff = 1; + li.l_linger = 120; +#endif + setsockopt(fh->handle, SOL_SOCKET, SO_LINGER, &li, sizeof(li)); + + if (fh->ino->localname){ + unlink (fh->ino->localname); + g_free (fh->ino->localname); + fh->ino->localname = NULL; + } + return 0; + } if (!fh->ino->localname) if (vfs_s_retrieve_file (me, fh->ino)==-1) return -1; @@ -1691,6 +1723,18 @@ static int ftpfs_fh_open (vfs *me, vfs_s return 0; } +static int ftpfs_fh_close (vfs *me, vfs_s_fh *fh) +{ + if (fh->handle != -1){ + close (fh->handle); + fh->handle = -1; + fh->changed = 0; + if (get_reply (me, fh->ino->SUP.sock, NULL, 0) != COMPLETE) + ERRNOR (EIO, -1); + } + return 0; +} + static struct vfs_s_data ftp_data = { NULL, 0, @@ -1707,7 +1751,7 @@ static struct vfs_s_data ftp_data = { free_archive, ftpfs_fh_open, /* fh_open */ - NULL, /* fh_close */ + ftpfs_fh_close, /* fh_close */ vfs_s_find_entry_linear, dir_load, From despair at sama.ru Wed Mar 6 14:09:37 2002 From: despair at sama.ru (Walery Studennikov) Date: Wed, 6 Mar 2002 18:09:37 +0400 Subject: [patch] Fix ftpfs uploading behavior. In-Reply-To: <20020306110240.A485@it.efp.com.ua> References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> Message-ID: <20020306180937.A9431@prog1> On Wed, Mar 06, 2002 at 11:02:40AM +0200, Andrew V. Samoilov wrote: > Hi! > > This patch fixes ftpfs uploading behavior. > Now if file is opened for writting only it will be not retrieved > to temporary file before uploading. > > This is first approximation. > Test it carefully please. It works for me. Thank you. Will those two patches be applied to CVS soon? -- Regards, Walery From despair at sama.ru Thu Mar 7 13:45:08 2002 From: despair at sama.ru (Walery Studennikov) Date: Thu, 7 Mar 2002 17:45:08 +0400 Subject: [patch] Fix ftpfs uploading behavior. In-Reply-To: <20020306110240.A485@it.efp.com.ua> References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> Message-ID: <20020307174508.A6019@prog1> On Wed, Mar 06, 2002 at 11:02:40AM +0200, Andrew V. Samoilov wrote: > Hi! > > This patch fixes ftpfs uploading behavior. > Now if file is opened for writting only it will be not retrieved > to temporary file before uploading. > > This is first approximation. > Test it carefully please. Probably, I've found a bug: now I can't open file for edit via ftp - MC hangs after downloading the file for editing. -- Regards, Walery From kai at cmail.ru Mon Mar 11 08:28:19 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Mon, 11 Mar 2002 10:28:19 +0200 Subject: [patch] Fix ftpfs uploading behavior. In-Reply-To: <20020307174508.A6019@prog1> References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> <20020307174508.A6019@prog1> Message-ID: <20020311102819.A204@it.efp.com.ua> On Thu, Mar 07, 2002 at 05:45:08PM +0400, Walery Studennikov wrote: > On Wed, Mar 06, 2002 at 11:02:40AM +0200, Andrew V. Samoilov wrote: > > Hi! > > > > This patch fixes ftpfs uploading behavior. > > Now if file is opened for writting only it will be not retrieved > > to temporary file before uploading. > > > > This is first approximation. > > Test it carefully please. > > Probably, I've found a bug: > now I can't open file for edit via ftp - > MC hangs after downloading the file for editing. > Fixed, see patch below (for patched ftpfs.c). Second approximation of this patch will be uploaded soon to http://www.linux.zp.ua/mc/ftpfs.upload.patch Also if your vfs/direntry.c is not from CVS, you need patch for this one 9 lines below. diff -u -p -u -p -r1.89 ftpfs.c --- ftpfs.c 2002/02/01 12:21:55 1.85 +++ ftpfs.c 2002/02/06 12:08:01 @@ -1691,7 +1723,7 @@ static int ftpfs_fh_open (vfs *me, vfs_s static int ftpfs_fh_close (vfs *me, vfs_s_fh *fh) { - if (fh->handle != -1){ + if (fh->handle != -1 && !fh->ino->localname){ close (fh->handle); fh->handle = -1; fh->changed = 0; --- vfs/direntry.c Tue Feb 19 10:59:23 2002 +++ vfs/direntry.c Wed Mar 6 10:46:35 2002 @@ -832,7 +826,7 @@ vfs_s_read (void *fh, char *buffer, int if (FH->linear == LS_LINEAR_OPEN) return MEDATA->linear_read (me, FH, buffer, count); - if (FH->handle){ + if (FH->handle != -1){ n = read (FH->handle, buffer, count); if (n < 0) me->verrno = errno; @@ -852,7 +846,7 @@ vfs_s_write (void *fh, char *buffer, int vfs_die ("no writing to linear files, please"); FH->changed = 1; - if (FH->handle){ + if (FH->handle != -1){ n = write (FH->handle, buffer, count); if (n < 0) me->verrno = errno; @@ -926,7 +920,7 @@ vfs_s_close (void *fh) } vfs_s_invalidate (me, FH_SUPER); } - if (FH->handle) + if (FH->handle != -1) close (FH->handle); vfs_s_free_inode (me, FH->ino); From kai at cmail.ru Wed Mar 13 12:40:14 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Wed, 13 Mar 2002 14:40:14 +0200 Subject: New version on mc-multiscreen-patch References: <20020216212737.A513@localhost.localdomain> Message-ID: <3C8F48AE.8020703@cmail.ru> Hello, Walery Studennikov wrote: >>PATH_SEP and PATH_SEP_STR have to be used instead of '/' >>and "/". >> > >Well, it seems to be fixed too : > > * configure.in: Add --with-logfile=path_to_logfile option > I mean runtime option for mc like -l ftplog or -M madlog. > > * main.c (log_mc_handler): > Use MC_DEBUG_LOG_FILE insted of hardcoded debug filename > > * edit/editwidget.c (edit): > Use vfs_file_is_local() instead heuristics, > Use PATH_SEP and PATH_SEP_STR insted of '/' and "/" respectively > inste_a_d > >http://www.sama.ru/~despair/mc/mc-multiscreen-0.7.3.patch.tar.gz > Regards, Andrew. From kai at cmail.ru Wed Mar 13 12:57:29 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Wed, 13 Mar 2002 14:57:29 +0200 Subject: [patch] Fix ftpfs uploading behavior. References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> <20020307174508.A6019@prog1> <20020311102819.A204@it.efp.com.ua> Message-ID: <3C8F4CB9.60105@cmail.ru> Andrew V. Samoilov wrote: >On Thu, Mar 07, 2002 at 05:45:08PM +0400, Walery Studennikov wrote: > >>On Wed, Mar 06, 2002 at 11:02:40AM +0200, Andrew V. Samoilov wrote: >> >>>Hi! >>> >>>This patch fixes ftpfs uploading behavior. >>>Now if file is opened for writting only it will be not retrieved >>>to temporary file before uploading. >>> >>>This is first approximation. >>>Test it carefully please. >>> >>Probably, I've found a bug: >>now I can't open file for edit via ftp - >>MC hangs after downloading the file for editing. >> > >Fixed, see patch below (for patched ftpfs.c). >Second approximation of this patch will be uploaded soon to > >http://www.linux.zp.ua/mc/ftpfs.upload.patch > Unfortunatelly, this patch hangs mc now if file is copied over ftp to the same ftp server. Regards, Andrew. From hp at redhat.com Wed Mar 13 17:26:28 2002 From: hp at redhat.com (Havoc Pennington) Date: 13 Mar 2002 12:26:28 -0500 Subject: private maintainer contact Message-ID: Hi, Could whoever is most actively maintaining/hacking on mc mail me privately? I can't figure out who to contact. :-/ Havoc From kai at cmail.ru Fri Mar 15 15:14:58 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Fri, 15 Mar 2002 17:14:58 +0200 Subject: [patch] Fix ftpfs uploading behavior -rc1 References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> <20020307174508.A6019@prog1> <20020311102819.A204@it.efp.com.ua> <3C8F4CB9.60105@cmail.ru> Message-ID: <3C920FF2.1020408@cmail.ru> Hello, I reworked ftpfs uploading patch and now I invite you to test it hardly. There are 3 changed files and one new field (linear) in vfs_s_super structure. This flag inform ftpfs_fh_open about ordered linear transfer in this ftp connection. Now temporary file is used only if file is copied over ftpfs from ftp server to the same ftp server. I think it is not good idea to open new control connection. Also this version will be uploaded to http://www.linux.zp.ua/mc/ftpfs.upload.patch soon. In final version I think add some more comments. Your comments and suggestions are greatly appreshiated. Regards, Andrew. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ftpfs.upload.patch URL: From despair at sama.ru Sat Mar 16 02:13:43 2002 From: despair at sama.ru (Walery Studennikov) Date: Fri, 15 Mar 2002 21:13:43 -0500 Subject: mcedit bug Message-ID: <20020315211343.A852@localhost.localdomain> Hello. An old bug: $ mcedit nonexisting_file creates nonexisting_file without prompting at start, so when we mistype filename which we want to edit we pollute filesystem with empty unwanted files. But we should create a file only when SAVING it. -- Regards, Walery From binner at kde.org Sun Mar 17 13:27:51 2002 From: binner at kde.org (Stephan Binner) Date: Sun, 17 Mar 2002 14:27:51 +0100 Subject: mc 4.5.55 scrambles prompt Message-ID: <16mahT-1k7WGOC@fmrl07.sul.t-online.com> Hello, when using xterm's dynamic control sequence feature like this export PS1=$PS1"\[\e]0;\w\a\]" which sets xterm's caption to the current working directory, mc 4.5.55 shows something like "user at host:~ > 0;~" if you're in your home. This looks broken, it shouldn't strip anything. KDE 3.0 interprets "\[\e]30;\w\a\]" and KDE 3.1 will require "\[\e]31;\w\a\]" for some features (agreed by Thomas Dickey) so it would be nice if next Midnight Commander would fix it. Bye, Steve From Juraj.Kolesar at st.fmph.uniba.sk Mon Mar 18 12:29:42 2002 From: Juraj.Kolesar at st.fmph.uniba.sk (Juraj.Kolesar at st.fmph.uniba.sk) Date: Mon, 18 Mar 2002 14:29:42 +0200 Subject: Blocks in GNOME Message-ID: <00A0B211.CEF55310.10@st.fmph.uniba.sk> Hi, I real enjoy using your program. It is very intuitive to use and now it's my best friend in writing html and c code. But I have small problem, maybe it is just because of old version. In text console it's everythig ok and properly working. Im using RH7.2 and GNOME and sawfish window manager. The problems starts when I resize the terminal window, where is mc running. When I want to edit some file, there is offten left some text from the standard mc, and it is not possible to edit the text where this is shown in irregular way. Anotherone is that it is not possible select the blocks with Shift-Uparrow or Shift-DownArrow. In text console this is working properly. Afterwords it doesnt work to save the selected block into temp.file with Ctrl-Ins and then copy it in other file with Shift-Ins. By selecting the block with F3 it is not possible also in the text console but this is not important, while Shift-arrows is working. So all problems are in Xwin when mc is runnig in terminal window. I dont know if the problems are because of my configuration, there is still something to do in development. That's all Thank for any answer Koli From proski at gnu.org Mon Mar 18 22:48:38 2002 From: proski at gnu.org (Pavel Roskin) Date: Mon, 18 Mar 2002 17:48:38 -0500 (EST) Subject: Blocks in GNOME In-Reply-To: <00A0B211.CEF55310.10@st.fmph.uniba.sk> Message-ID: Hello! > I real enjoy using your program. It is very intuitive to use and now it's my > best friend in writing html and c code. But I have small problem, maybe it is > just because of old version. In text console it's everythig ok and properly > working. Im using RH7.2 and GNOME and sawfish window manager. The problems > starts when I resize the terminal window, where is mc running. When I want to > edit some file, there is offten left some text from the standard mc, and it is > not possible to edit the text where this is shown in irregular way. Please always report the version of software for which you report the problem. Don't assume that everybody knows what version of MC is included with RH 7.2 or with the unspecified version of GNOME you are running. Use "mc -V" - it will display the version and other useful information. Please consider upgrading to the latest stable version 4.5.55 or you would waste your time reporting bugs that were fixed long ago. Also please try to explain how to reproduce the problem step-by-step. Don't worry if it doesn't cover all problems you see, but if nobody can reproduce your problem then nobody will be able to fix it. > Anotherone is that it is not possible select the blocks with Shift-Uparrow or > Shift-DownArrow. In text console this is working properly. Afterwords it doesnt > work to save the selected block into temp.file with Ctrl-Ins and then copy it > in other file with Shift-Ins. By selecting the block with F3 it is not possible > also in the text console but this is not important, while Shift-arrows is > working. So all problems are in Xwin when mc is runnig in terminal window. I > dont know if the problems are because of my configuration, there is still > something to do in development. I don't understand about F3 - it works for me. Shift-Arrows don't work in anything but Linux console because there are no standard sequences for those key combinations, and "learning" those sequences is not implemented. Linux console allows direct access to the keyboard state, that's why it works. You can recompile mc and give "--with-tm-x-support" argument to "configure". This would enable direct keyboard access under X and you can make Shift-Arrows work under X if you teach MC additional escape sequences (it's non-trivial, so ask in the list if you come to this point). This would also make MC depend on X11 libraries - you won't be able to run MC without them. This hack is by no means finished and it seems that its author lost interest to doing it the right way long ago. -- Regards, Pavel Roskin From proski at gnu.org Mon Mar 18 23:01:52 2002 From: proski at gnu.org (Pavel Roskin) Date: Mon, 18 Mar 2002 18:01:52 -0500 (EST) Subject: mc 4.5.55 scrambles prompt In-Reply-To: <16mahT-1k7WGOC@fmrl07.sul.t-online.com> Message-ID: Hi, Stephan! > when using xterm's dynamic control sequence feature like this > > export PS1=$PS1"\[\e]0;\w\a\]" > > which sets xterm's caption to the current working directory, > mc 4.5.55 shows something like "user at host:~ > 0;~" if you're > in your home. This looks broken, it shouldn't strip anything. It is done on purpose in function strip_ctrl_codes(), file src/util.c. Some of the sequences can create mess on the screen if they e.g. move the cursor. I don't think that separating the sequences into "harmful" and "benign" and filtering only the former is worth the trouble. Actually, I think that the idea of loading the command prompt from the subshell is broken. It leads to many little bugs of a different kind. For example, if you run a command from mc that prints something and you press Ctrl-O before it finishes, the prompt show the output of the command. It looks very weird. It's also very easy to make MC prompt out-of-sync with the shell prompt. Just do Ctrl-O, "cd ..", Ctrl-O and look at the prompt and at the panel. The panel shows the new directory, the prompt is unchanged. I believe that this feature should be removed from MC at some point because it cannot be fixed properly - it's broken by design. In the meantime don't expect too much from prompt loading. -- Regards, Pavel Roskin From proski at gnu.org Mon Mar 18 23:40:38 2002 From: proski at gnu.org (Pavel Roskin) Date: Mon, 18 Mar 2002 18:40:38 -0500 (EST) Subject: Possible mistake, but maybe not... In-Reply-To: <20020304180706.A807@polcard.com.pl> Message-ID: Hi, Mariusz! On Mon, 4 Mar 2002, Mariusz Jedrzejewski wrote: > Hello. > I'm newbie at this list, and I've seen at archive, but doesn't find any answer > for my problem. If you want your reports to be considered seriously, please specify the version of the software for which you report the problem. Additional bonus if you are using the latest version. -- Regards, Pavel Roskin From proski at gnu.org Tue Mar 19 00:03:40 2002 From: proski at gnu.org (Pavel Roskin) Date: Mon, 18 Mar 2002 19:03:40 -0500 (EST) Subject: Russian characters in src/Makefile.am Message-ID: Hi, Andrew! Could you please eliminate Russian characters from src/Makefile.am? You are setting a bad example and I cannot leave it as is in the release. Other people will follow your steps and will add text to Makefile.am in other languages, even in 2-byte languages, such as Chinese. This is not the right way to mix different encodings in the same file. Consider adding a new keyword to mc.1 that would hold the translation of "Topics". -- Regards, Pavel Roskin From sav at it.efp.com.ua Tue Mar 19 14:15:59 2002 From: sav at it.efp.com.ua (Andrew V. Samoilov) Date: Tue, 19 Mar 2002 16:15:59 +0200 Subject: [patch] Fix ftpfs uploading behavior -rc1 References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> <20020307174508.A6019@prog1> <20020311102819.A204@it.efp.com.ua> <3C8F4CB9.60105@cmail.ru> <3C920FF2.1020408@cmail.ru> <20020318203002.GG21973@atrey.karlin.mff.cuni.cz> Message-ID: <3C97481F.9080500@it.efp.com.ua> Hello, Pavel Machek wrote: > Hi! > >>I reworked ftpfs uploading patch and now I invite you to test it >>hardly. There are 3 changed files and one new field (linear) in >>vfs_s_super structure. This flag inform ftpfs_fh_open about ordered >>linear transfer in this ftp connection. Now temporary file is used >>only if file is copied over ftpfs from ftp server to the same ftp >>server. I think it is not good idea to open new control connection. >> > > Do you think linear field in vfs_s_super is right thing? > > It seems very wrong to me. That state is certainly not > per-superblock.... > > Pavel I consider to rename this field to u.ftp.control_connection_buzy or data_transfer and call MEDATA->linear_start in vfs_s_open instead of vfs_s_read. Or even eliminate linear_* family and handle O_LINEAR flag in fh_open/fh_close. But this requires changes not only in ftpfs.c but also in fish.c. Index: direntry.c =================================================================== RCS file: /cvs/gnome/mc/vfs/direntry.c,v retrieving revision 1.40 diff -u -p -u -p -r1.40 direntry.c --- direntry.c 6 Mar 2002 08:46:35 -0000 1.40 +++ direntry.c 15 Mar 2002 14:51:18 -0000 @@ -788,6 +788,10 @@ vfs_s_open (vfs *me, char *file, int fla if (IS_LINEAR(flags)) { - fh->linear = LS_LINEAR_CLOSED; + print_vfs_message (_("Starting linear transfer...")); + if (!MEDATA->linear_start (me, FH, 0)){ + g_free(fh); + return NULL; + } } else if ((MEDATA->fh_open) && (MEDATA->fh_open (me, fh, flags, mode))){ g_free(fh); return NULL; @@ -816,8 +817,10 @@ vfs_s_read (void *fh, char *buffer, int - if (FH->linear == LS_LINEAR_CLOSED){ - print_vfs_message (_("Starting linear transfer...")); - if (!MEDATA->linear_start (me, FH, 0)) - return -1; - } - - if (FH->linear == LS_LINEAR_CLOSED) . . . -- Regards, Andrew V. Samoilov. From sav at it.efp.com.ua Tue Mar 19 14:37:06 2002 From: sav at it.efp.com.ua (Andrew V. Samoilov) Date: Tue, 19 Mar 2002 16:37:06 +0200 Subject: Russian characters in src/Makefile.am References: Message-ID: <3C974D12.1040808@it.efp.com.ua> Hi, Pavel today you wrote: > Hi, Andrew! > > Could you please eliminate Russian characters from src/Makefile.am? You > are setting a bad example and I cannot leave it as is in the release. > Other people will follow your steps and will add text to Makefile.am in > other languages, even in 2-byte languages, such as Chinese. This is not > the right way to mix different encodings in the same file. Consider > adding a new keyword to mc.1 that would hold the translation of "Topics". Well, it seems to be a good move. I will see some later. BTW, when do you plan to release next version? Another good thing may be to move rules for generating and installation of mc.hlp.?? from src/Makefile.am to doc/??/Makefile.am and check ALL_LINGUAS for list of languages. But I need your help with automake for these changes, especially for doc/Makefile.am -- Regards, Andrew V. Samoilov. From martin at internet-treff.uni-koeln.de Tue Mar 19 20:47:54 2002 From: martin at internet-treff.uni-koeln.de (Martin Bialasinski) Date: 19 Mar 2002 21:47:54 +0100 Subject: Making the "Panel" Background White? In-Reply-To: <3C9772DB.7498@fast.net> References: <3C9772DB.7498@fast.net> Message-ID: <878z8ouvwl.fsf@haitech.martin.home> Randy Kramer wrote: Hi, > > mc -C directory=red,yellow > > > > Do you see the text or you have red on red? What terminal aree > > you using? > Well, I can see that there is text there, but it appears to be > almost the same color (red more than yellow) and very hard to read. > It appears the same for my son, but I'll check with my wife when she > comes home. FYI, this is what I get: xterm: red on yellow rxvt: red on yellow eterm: red on lightbraun gnome-terminal: red on lightbraun/orange So translation of symbolic colornames to RGB values is not consistent between terminals. > BTW, mc -C directory=yellow,red gives significantly better results. yellow is translated different if it is used as foreground color than if it is used as background color. mc -C directory=yellow,yellow has always a brighter tone as foreground color in the terminals I have installed. Just observations. I don't know how the terminals work internally, or how to change the behaviour. Ciao, Martin From pavel at suse.cz Tue Mar 19 21:26:45 2002 From: pavel at suse.cz (Pavel Machek) Date: Tue, 19 Mar 2002 22:26:45 +0100 Subject: [patch] Fix ftpfs uploading behavior -rc1 In-Reply-To: <3C97481F.9080500@it.efp.com.ua> References: <20010908001446.A7576@localhost.localdomain> <20020306110240.A485@it.efp.com.ua> <20020307174508.A6019@prog1> <20020311102819.A204@it.efp.com.ua> <3C8F4CB9.60105@cmail.ru> <3C920FF2.1020408@cmail.ru> <20020318203002.GG21973@atrey.karlin.mff.cuni.cz> <3C97481F.9080500@it.efp.com.ua> Message-ID: <20020319212645.GI12260@atrey.karlin.mff.cuni.cz> Hi! > >>I reworked ftpfs uploading patch and now I invite you to test it > >>hardly. There are 3 changed files and one new field (linear) in > >>vfs_s_super structure. This flag inform ftpfs_fh_open about ordered > >>linear transfer in this ftp connection. Now temporary file is used > >>only if file is copied over ftpfs from ftp server to the same ftp > >>server. I think it is not good idea to open new control connection. > >> > > > > Do you think linear field in vfs_s_super is right thing? > > > > It seems very wrong to me. That state is certainly not > > per-superblock.... > > I consider to rename this field to u.ftp.control_connection_buzy > or data_transfer and call MEDATA->linear_start in vfs_s_open I guess u.ftp.control_connection_busy is okay. Pavel -- Casualities in World Trade Center: ~3k dead inside the building, cryptography in U.S.A. and free speech in Czech Republic. From kai at cmail.ru Wed Mar 20 15:11:11 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Wed, 20 Mar 2002 17:11:11 +0200 Subject: regex fix for cygwin References: Message-ID: <3C98A68F.80100@cmail.ru> Pavel Roskin wrote: > Hi! > > >>I agree that we must use the second approach, i.e.: >> >>1. Use the mc's regex code, i.e. src/regex.c and src/eregex.h. >>Maybe we should rename src/regex.c to src/eregex.c just to be consistent? >> > > I would rather call them both regex.* unless it causes problems (it should > not). > > >>2. Reenable the mc's regex code for all platforms, i.e. remove those "#if >>0" from src/regex.c and src/eregex.h >> > > Take the latest regex sources from gnu. It is Ok if all code is disabled > when GNU libc is found - libc should provide the same functions, although > I'm a bit worried if those internal functions can be used - if not, then > the included regex should be used. > > Important is that either you disable all or you enable all. No mix of > libc regex and included regex. > > >>3. Replace all instances of #include "regex.h" by #include "eregex.h" >> > > Or vice versa, but make sure that the included regex.h has preference. > I use Slackware 7.1 with glibc-2.1.3. After March 18 mc compiles and use its own regex.c and ignores glibc's ones. And I am not happy with ~28k overhead and 6 warnings. -- Regards, Andrew V. Samoilov. From avarakin00 at hotmail.com Thu Mar 21 01:31:45 2002 From: avarakin00 at hotmail.com (Alexander Varakin) Date: Wed, 20 Mar 2002 20:31:45 -0500 Subject: regex fix for cygwin References: <3C98A68F.80100@cmail.ru> Message-ID: ----- Original Message ----- From: "Andrew V. Samoilov" To: "Pavel Roskin" Cc: "Alexander Varakin" ; Sent: Wednesday, March 20, 2002 10:11 AM Subject: Re: regex fix for cygwin > Pavel Roskin wrote: > > > Hi! > > > > > >>I agree that we must use the second approach, i.e.: > >> > >>1. Use the mc's regex code, i.e. src/regex.c and src/eregex.h. > >>Maybe we should rename src/regex.c to src/eregex.c just to be consistent? > >> > > > > I would rather call them both regex.* unless it causes problems (it should > > not). > > > > > >>2. Reenable the mc's regex code for all platforms, i.e. remove those "#if > >>0" from src/regex.c and src/eregex.h > >> > > > > Take the latest regex sources from gnu. It is Ok if all code is disabled > > when GNU libc is found - libc should provide the same functions, although > > I'm a bit worried if those internal functions can be used - if not, then > > the included regex should be used. > > > > Important is that either you disable all or you enable all. No mix of > > libc regex and included regex. > > > > > >>3. Replace all instances of #include "regex.h" by #include "eregex.h" > >> > > > > Or vice versa, but make sure that the included regex.h has preference. > > > I use Slackware 7.1 with glibc-2.1.3. After March 18 > mc compiles and use its own regex.c and ignores glibc's > ones. And I am not happy with ~28k overhead and 6 warnings. > > > -- > Regards, > Andrew V. Samoilov. > Hi Andrew, I don't understand - are you saying that after March 18 you can't compile mc? I agree that having extra 6 warnings from regex.c is not good. I took this version of regex.c from the latest stable release of glibc - 2.2.5. Maybe we should escalate this issue to glibc folks, or just fix this warnings. I don't think we should worry about ~28k overhead - people are switching from CDs to DVDs and from 32 bits to 64 bits already , whereas MC can fit on just few floppies. Alex From kai at cmail.ru Thu Mar 21 08:38:21 2002 From: kai at cmail.ru (Andrew V. Samoilov) Date: Thu, 21 Mar 2002 10:38:21 +0200 Subject: regex fix for cygwin References: <3C98A68F.80100@cmail.ru> Message-ID: <3C999BFD.3010507@cmail.ru> Alexander Varakin wrote: > ----- Original Message ----- > From: "Andrew V. Samoilov" > To: "Pavel Roskin" > Cc: "Alexander Varakin" ; > Sent: Wednesday, March 20, 2002 10:11 AM > Subject: Re: regex fix for cygwin > >>Pavel Roskin wrote: >> >> >>>Hi! >>> >>> >>> >>>>I agree that we must use the second approach, i.e.: >>>> >>>>1. Use the mc's regex code, i.e. src/regex.c and src/eregex.h. >>>>Maybe we should rename src/regex.c to src/eregex.c just to be >>>> > consistent? > >>>I would rather call them both regex.* unless it causes problems (it >>> should not). >>> >>>>2. Reenable the mc's regex code for all platforms, i.e. remove those >>>> > "#if > >>>>0" from src/regex.c and src/eregex.h >>>> >>>> >>>Take the latest regex sources from gnu. It is Ok if all code is >>> > disabled > >>>when GNU libc is found - libc should provide the same functions, >>> > although > >>>I'm a bit worried if those internal functions can be used - if not, then >>>the included regex should be used. >>> >>>Important is that either you disable all or you enable all. No mix of >>>libc regex and included regex. >>> >>> >>> >>>>3. Replace all instances of #include "regex.h" by #include "eregex.h" >>>> >>>> >>>Or vice versa, but make sure that the included regex.h has preference. >>> >>> >>I use Slackware 7.1 with glibc-2.1.3. After March 18 >>mc compiles and use its own regex.c and ignores glibc's >>ones. And I am not happy with ~28k overhead and 6 warnings. >> >> >>-- >>Regards, >>Andrew V. Samoilov. >> >> > > Hi Andrew, > > I don't understand - are you saying that after March 18 you can't compile > mc? I can't find where did you read this. > I agree that having extra 6 warnings from regex.c is not good. I took this > version of regex.c from the latest stable release of glibc - 2.2.5. Maybe we > should escalate this issue to glibc folks, or just fix this warnings. > I don't think we should worry about ~28k overhead - people are switching > from CDs to DVDs and from 32 bits to 64 bits already , whereas MC can fit on > just few floppies. But another people use 1 floppy or 1 CD distros and embedded systems. And people with already installed glibc-2.2.5+ will use regex from 2.2.5. I have not any problems with glibc-2.1.x regex implementation, but I must compile and use new regex.c. And now we will need to track and fix any bugs if they will be found in this implementation instead of enjoing all of benefits of system installed software. As far as I can see configure checks for glibc-2.1 now. And I suggest to eliminate src/regex.[ch] including/compilation for glibc-2.1+ based systems. There no need to change all of files with included eregex.h. Something like #ifdef GLIBC_2_1 #include #else can be added into our eregex.h. Regards, Andrew. From binner at kde.org Thu Mar 21 18:54:48 2002 From: binner at kde.org (Stephan Binner) Date: Thu, 21 Mar 2002 19:54:48 +0100 Subject: mc 4.5.55 scrambles prompt In-Reply-To: References: Message-ID: <16o7i0-0EM7JQC@fmrl07.sul.t-online.com> On Tuesday 19 March 2002 00:01, Pavel Roskin wrote: > It is done on purpose in function strip_ctrl_codes(), file src/util.c. > Some of the sequences can create mess on the screen if they e.g. move the > cursor. I don't think that separating the sequences into "harmful" and Please don't try to make mc "smart" so it fixes messes a user maybe causes, he should notice and fix it himself if he included such harming sequences eg by defining a "clean" prompt for Midnight Commander in ~/.mc/bashrc . The latter can now be only used to fix the cosmetic issue but I would love to see the current working directory displaying as caption. :-) Bye, Steve From proski at gnu.org Fri Mar 22 06:21:24 2002 From: proski at gnu.org (Pavel Roskin) Date: Fri, 22 Mar 2002 01:21:24 -0500 (EST) Subject: mc 4.5.55 scrambles prompt In-Reply-To: <16o7i0-0EM7JQC@fmrl07.sul.t-online.com> Message-ID: Hi, Stephan! > > It is done on purpose in function strip_ctrl_codes(), file src/util.c. > > Some of the sequences can create mess on the screen if they e.g. move the > > cursor. I don't think that separating the sequences into "harmful" and > > Please don't try to make mc "smart" so it fixes messes a user maybe causes, > he should notice and fix it himself if he included such harming sequences > eg by defining a "clean" prompt for Midnight Commander in ~/.mc/bashrc . This function was introduced back in 1994, before I even heard of Midnight Commander. So I'm not trying to make mc smart. You are assuming a certain sequence of events, namely that the prompt is changed while mc is being actively used. If you consider somebody already having a fancy prompt trying mc the first time, the problem may be less trivial for that user. As I said, there is no guarantee that only the symbols from the shell prompt appear in the mc prompt. Run "find /" and press Ctrl-O before it finishes. This bug needs to be fixed, but nobody has fixed it so far. Not filtering the prompt could cause mess all over the screen, not just in the command line. > The latter can now be only used to fix the cosmetic issue but I would > love to see the current working directory displaying as caption. :-) Ideally, MC could interpret PS1 or some other variables in the way other shells do it. I never liked load_prompt(), I just have other priorities now. -- Regards, Pavel Roskin From avarakin00 at hotmail.com Fri Mar 22 18:01:47 2002 From: avarakin00 at hotmail.com (Alexander Varakin) Date: Fri, 22 Mar 2002 13:01:47 -0500 Subject: AIX port References: Message-ID: Hi, I am trying to build the latest snapshot of mc on AIX. Which compiler must be used for that? Must it be gcc or xlc ? After some code fixing I was able to build it with xlc (we don't have gcc on our Unix boxes). Now I have problem with subshell - it doesn't work at all. Does anybody have latest mc running on AIX with subshell working? Alex From christian.gennerat at vz.cit.alcatel.fr Tue Mar 26 15:47:49 2002 From: christian.gennerat at vz.cit.alcatel.fr (Christian Gennerat) Date: Tue, 26 Mar 2002 16:47:49 +0100 Subject: mc.ini and syntax files with white background Message-ID: <3CA09825.755772B5@vz.cit.alcatel.fr> The standard colour schema in mc matches with a blue background. Do some users prefer a White (or sand, papayawhip) background and have a complete set of mc.ini and syntax files for this background ? From christian.gennerat at vz.cit.alcatel.fr Wed Mar 27 11:27:17 2002 From: christian.gennerat at vz.cit.alcatel.fr (Christian Gennerat) Date: Wed, 27 Mar 2002 12:27:17 +0100 Subject: mc.ini and syntax files with white background Message-ID: <3CA1AC95.CBAED261@vz.cit.alcatel.fr> The standard colour schema in mc matches with a blue background. Do some users prefer a White (or sand, papayawhip) background and have a complete set of mc.ini and syntax files for this background ? From consultant12 at ulm.siemens.de Wed Mar 27 12:49:56 2002 From: consultant12 at ulm.siemens.de (consultant12) Date: Wed, 27 Mar 2002 13:49:56 +0100 Subject: Documentation for Norton Commander Message-ID: I have just started a contract at Siemens in Ulm, Germany. They are using Norton Commander. Unfortunately it is a German version. There is no documentation in any language. All relevant books appear to be out of print. I would be most grateful if any one could either send me some documentation in English or suggest from where I might get some. Regards Dennis Edwards From bjorn at bjornen.nu Wed Mar 27 20:23:51 2002 From: bjorn at bjornen.nu (=?iso-8859-1?Q?Bj=F6rn?= Eriksson) Date: Wed, 27 Mar 2002 21:23:51 +0100 Subject: Documentation for Norton Commander In-Reply-To: References: Message-ID: <20020327202351.GB29904@bjorn> On Wed, Mar 27, 2002 at 01:49:56PM +0100, consultant12 wrote: > I have just started a contract at Siemens in Ulm, Germany. > > They are using Norton Commander. > > Unfortunately it is a German version. > > There is no documentation in any language. > > All relevant books appear to be out of print. > > I would be most grateful if any one could either send me some documentation > in English or suggest from where I might get some. Hey! It's not April 1st yet. I started out writing a helpful reply but I had a nagging suspicion something was wrong... :-) NC5 is avaiable on a number of abandonware sites. It's not legal though. -- //Bj?rnen. bjorn at bjornen.nu | http://bjornen.nu | [aka. mdeans at algonet.se] -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 232 bytes Desc: not available URL: From christian.gennerat at vz.cit.alcatel.fr Thu Mar 28 09:31:00 2002 From: christian.gennerat at vz.cit.alcatel.fr (Christian Gennerat) Date: Thu, 28 Mar 2002 10:31:00 +0100 Subject: mc.ini and syntax files with white background Message-ID: <3CA2E2D4.9585A256@vz.cit.alcatel.fr> The standard colour schema in mc matches with a blue background. Do some users prefer a White (or sand, papayawhip) background and have a complete set of mc.ini and syntax files for this background ? From sangeeth_10 at yahoo.com Sun Mar 31 02:34:12 2002 From: sangeeth_10 at yahoo.com (sangeetha) Date: Sat, 30 Mar 2002 18:34:12 -0800 (PST) Subject: detection of devices - urgent Message-ID: <20020331023412.34116.qmail@web13306.mail.yahoo.com> hi, is there anyway to detect whether devices(keyboard and mouse) are connected to system or not.........using c or assembly code on linux platform if so plz do let me know how we can do that.thanx in advance. sangeetha __________________________________________________ Do You Yahoo!? Yahoo! Greetings - send holiday greetings for Easter, Passover http://greetings.yahoo.com/