Re: Programing Question
On Sat, Mar 29, 2003 at 12:09:44AM -0600, Harley D. Eades III wrote: Hello, This might be off topic, but I hope someone can answer a question for me. Is there any documentation on parport.h parport_pc.h. I seem to get err's when I try to use either of them. Do you have to link to a lib? I am not shere I am trying to learn more about devices and such. Any help will be appreciated. What do you mean, use? news:comp.os.linux.development.sys or news:comp.os.linux.development.apps would probably be better places to ask such questions. -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [ot] Linux stdio question, howto find fopened files
On Fri, Mar 21, 2003 at 03:03:33PM -0600, Michael Heironimus wrote: Well, __iob is reasonably portable because it looks like it's a standard part of a System V libc. In theory, I think glibc is supposed to support the System V ABI, but it doesn't seem to have an __iob[]. I don't think __iob is specified in any other standard, and different versions of the System V ABI standard don't even define it in quite the same way. info libc ; i SVID SVID (The System V Interface Description) - ... The GNU C library defines most of the facilities required by the SVID that are not also required by the ISO C or POSIX standards, for compatibility with System V Unix and other Unix systems (such as SunOS) which include these facilities. However, many of the more obscure and less generally useful facilities required by the SVID are not included. (In fact, Unix System V itself does not provide them all.) ... So, the long and the short of it is GNU libc doesn't claim to fully support System V. I looked for __iob in the SUS3, for instance, and couldn't find neither hide nor hair... -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mutt : how do I send multipart/alternative ?
On Fri, Mar 21, 2003 at 03:12:17PM +0100, Frank Gevaerts wrote: Hi, I'd like to send MIME multipart/alternative mails with mutt. The only things I find about mutt and multipart/alternative are about how to display such mails. Is there any documentation on this available somewhere ? (For those who are wondering : I do not intend to send html mail, but I want to reply to html mail with a multipart tex/postscript mail) Evil. I like it. Don't forget a troff version... -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse On Fri, Mar 21, 2003 at 03:12:17PM +0100, Frank Gevaerts wrote: Hi, I'd like to send MIME multipart/alternative mails with mutt. The only things I find about mutt and multipart/alternative are about how to display such mails. Is there any documentation on this available somewhere ? (For those who are wondering : I do not intend to send html mail, but I want to reply to html mail with a multipart tex/postscript mail) Evil. I like it. Don't forget a troff version... -- echo ">[EMAIL PROTECTED]< eryyvZ .T pveR" | rot13 | reverse
Re: [ot] Linux stdio question, howto find fopened files
On Wed, Mar 19, 2003 at 01:32:26PM -0500, Walter Tautz wrote: please retain the CC to rbutterworth Subject: Linux stdio question. On non-linux unix systems, one can reference __iob[] to find all currently fopen()ed files (e.g. when forking a new process one would generally want to flush their buffers first, or perhaps close most of them). Linux's stdio.h doesn't provide such an array of open FILE pointers, or at least if it does I can't find it. This seems to be a glibc question. Folks on news:comp.os.linux.development.[apps|sys] might know better. Referencing internal data structures is inherently non-portable. You might look in /proc/pid/fd/ which seems to have file descriptor numbers represented as symbolic links to the file(s) opened. fflush (NULL); flushes all fopen'ed files, but that doesn't mean writes are actually committed (you need sync/fsync for that). Any idea what they call it, or how one can find all currently open FILEs? Perhaps there is a better way? A general guide to porting underlinux /debian would be appreciated. Linux Standards Base? -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: odd compiler behaviour?
On Sun, Mar 16, 2003 at 10:06:01PM +, Colin Watson wrote: I stand corrected, then. Although 'info libc' does say that (void) is OK for main() in ISO C. Correct. Two ISO C forms of main are: int main (void) { ... } /* identical to: int main () { ... } */ or int main (int argc, char *argv[]) { ... } /* identical to: int main (int argc, char **argv) { ... } */ However: int main (); and int main (void); are not identical *declarations*. But only crazy people who like to call main recursively ever write declarations for it... In non-conforming mode (the default), gcc will also accept void main(...) and the three argument form that adds char **envp... And then, there is the KR form with implicit int return: main (argc, argv) int argc; char *argv[]; {...} KR is accepted but deprecated in all versions of ISO/ANSI C except the implicit int return type is an error in C99. (Now, back to your regularly scheduled program ...) -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: odd compiler behaviour?
On Sun, Mar 16, 2003 at 12:40:52PM +, Colin Watson wrote: On Sun, Mar 16, 2003 at 04:56:53PM +1100, Rob Weir wrote: On Fri, Mar 14, 2003 at 04:00:22PM +, Bruynooghe Floris wrote: int main() An addition to what everyone else said, this really should be int main(int argc, char** argv) to truly satisfy the pedant within :) Doesn't matter if you aren't using them. '()' in C means unspecified arguments, as opposed to no arguments which is '(void)'. In a definition, () and (void) are identical. It's only in the declaration where () means a fixed but unspecified number of arguments and only crazy people write declarations for main. -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: What he heck is FIFO?
On Tue, Mar 04, 2003 at 10:23:40PM -0300, Henrique de Moraes Holschuh wrote: On Tue, 04 Mar 2003, Greg Madden wrote: the remote I am asked for an IP and port. Grrr. For the local I am asked for a FIFO. Please tell me what to do to dial. By the way, I am in Gnome. BTW, install dictd, dict, and dict-vera. Then you can: $ dict FIFO which answers with: From V.E.R.A. -- Virtual Entity of Relevant Acronyms June 2002 [vera]: FIFO First In First Out (CPU) It won't help you dial your ISP, but now you know what the heck is a FIFO :P That's not an especially helpful definition given the context. $ info mkfifo ... A FIFO is a special file type that permits independent processes to communicate. One process opens the FIFO file for writing, and another for reading, after which data can flow as with the usual anonymous pipe in shells or elsewhere. ... An example of a FIFO is /dev/xconsole, as in: $ xconsole -file /dev/xconsole Messages get written to the special file and xconsole reads them into a window (something like tail -f /dev/syslog). -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X Windows has strange color scheme
On Thu, Feb 20, 2003 at 11:25:32PM +0100, Robert Ewald wrote: It might be just a wild guess, but could it be that you hit CTRL-ALT-+ at some point and switched the color depth to 8? Then strange colors like this, which turn normal on clicking in the window would be explainable. just hit CTRL-ALT-+ 3 times to get into 16 bit color depth again. ctrl-alt-[+|-] cycles resolution, *not* color depth. AFAIK, there is no way to change color depth in X without a restart. -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mozilla Adobe's SVG viewer
On Thu, Feb 20, 2003 at 08:47:12AM +0700, Oki DZ wrote: Hi, Yesterday I downloaded the Adobe SVG viewer plugin. Currently, I'm using Sid's Mozilla snapshot. The plugin doesn't work. Question is: is it really not working for mozilla-snapshot? I'm just want to make sure. Last I knew, the Adobe plugin was only for Windows and only worked with MS IE. Has that changed? -- echo [EMAIL PROTECTED] eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Newbie Installation Problems
On Fri, Feb 14, 2003 at 09:47:04PM -0800, Paul Johnson wrote: Fun fact: Today is Oregon Day. On this day in 1859, Oregon became the US's 33rd state, the result of the Vote at Champoeg, in which two Canadians wanted dead or alive tipped the vote from 49/50 to 51/50 in favor of becoming a US state instead of a Canadian province. Retrospectively, this is considered the worst thing to happen to Oregon as it leaves the international boundary on the wrong side of the state, something that has left us completely defenseless against political and migratory abuse by California (everybody stop moving here, Oregon's full now, not that anybody was welcome to begin with). Please take your tripe elsewhere. In the U.S. people are free to move from state to state without restriction. Get over it. Replace Californian with nigger/jew/whop/chink/irish/injun/wetback/etc. and see how palatable your xenophobia becomes. More people move to California in any given year than Oregon and California is similarly defenseless against these migrations. Since this is largely an economic decision, the solution lies in: (1) make Oregon a prohibitively expensive place to live; and (2) make Oregon a prohibitively expensive place to do business. If you succeed in both of those, I'm sure migrations will dwindle to a trickle. Of course, it might make Oregon a less desirable place to live for the natives as well. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ftpd-ssl
On Sun, Feb 16, 2003 at 01:52:17AM +, mjoyce wrote: I have ftpd-ssl running, it seems to work very well. As far as I can tell it just uses port 22, neat, this seems to make the problems of ftp, port, firewalls, passive clients etc, go away, just open and forward port 22. Is this right ? if so, i'm surprised more people don;t use it. The FTP protocol uses two ports and FTP-SSL uses two more... $ grep 'ftp' /etc/services ftp-data20/tcp ftp 21/tcp ftps-data 989/tcp # FTP over SSL (data) ftps990/tcp # FTP over SSL Probably more people would use it if more ftp clients supported it... I don't know how well it's supported in MS and Apple worlds... Most anonymous access should probably continue to use normal ftp. SSL does have overhead... -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Setting progs at a runlevel
On Sun, Feb 16, 2003 at 12:45:10AM +0100, Jeff Elkins wrote: As a RH refugee, I'm used to running the 'setup' program to enable/disable program startups at runtime. For Debian, am I correct in cd'ing to the /etc/rcX.d dir and moving SXXprogram to KXXprogram to disable it? If so, is there a better way? That works. There's also update-rc.d (though it was really written for scripts). I think some of the system management front-end tools advertise the ability to mess up the SYSV links as well (linuxconf, etc...). -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My cup runneth over!
On Sat, Feb 15, 2003 at 11:11:51PM -0600, will trillich wrote: On Thu, Feb 13, 2003 at 06:46:03PM -0500, David Turetsky wrote: You sure there ain't some way to prevent a slew of xterm windows from opening up each time I reboot into gdm? Earlier today, if there was one, there was 60.70.80 Seems like each time I reboot, the number doubles!!! the SAME EXACT THING was happening to me. i'm not all that adept at configuring this X stuff, so if you ask me there's something behind-the-scenes that appears to have a bit of a hole in it. (i.e. instead of clobbering the saved session with the current session, it appears to APPEND the current session to the previously stored session, hence doubling every instance of every window for the next time you log in.) and i'm on woody/stable. my solution? apt-get install kde Well, whatever problem you folks are experiencing is almost certainly not GDM's fault. I'd wager a gnome-session manager problem. I'll bet if you log into a Debian Session from GDM, you don't see the brain damage (unless your ~/.xsession script is horked). -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Console fonts in rxvt
On Tue, Feb 11, 2003 at 10:06:06PM -0500, George Georgalis wrote: Hi, another question on the same topic... On Tue, Feb 11, 2003 at 02:32:49PM -0700, Cameron Matheson wrote: What are you using as a font name? You just use any font that X recognizes (use xfontsel to get the font-name). With aterm, i use the How does one specify the font: -jmk-neep alt-medium-r-normal-*-*-140-*-*-c-*-iso8859-1 there is a space in the name and I can't get it to work with rxvt... Wrap it in quotes. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: shuttle disaster
On Sun, Feb 09, 2003 at 12:30:57AM -0800, Paul Johnson wrote: On Sun, Feb 09, 2003 at 01:24:22AM -0500, Geordie Birch wrote: I thought so too, but no bank in Arcata would touch it if I didn't have an account. This was in 1995. What do you expect? It's a state that's been 0wnz3d by Wells Fargo, a bank known for screwing people attempting to use it's services in the proper coin of the realm. It doesn't help that all the other banks down there seem to be trying to play catchup on how badly they can screw thier customers to compete with Wells Fargo. I bank with Washington Mutual[1] (formerly Fred Meyer Saving and Loan), which offers unlimited free checking, free checks, free check card, free online banking, no charge for talking to a human teller, they'll let you overdraft up to $100 for free (more if your credit's good) and reasonable exchange rates. They also allow non-account-holders to do currency-to-currency transactions with a teller and use thier ATMs for free, and they won't charge account holders for using another bank's ATM (though the other bank might). Last time I was in Arcata it had at least three or four banks that weren't Wells Fargo and one is a Washington Mutual. I don't recall if it was there in 1995 (I think not). That none would exchange currency with folks who aren't account holders isn't surprising. Arcata is behind the Redwood Curtain, 300 miles from anywhere and has a population less than 20,000 (and only that large due to the university). It's not exactly a hub of international commerce. The tellers could probably count the number of currency exchanges in any given year on one hand. When I was in some boony town halfway between Montreal and Quebec, the only restaurant wouldn't take my U.S. currency. Imagine that! And the dang menu was in French to boot. But, I didn't blame the proprietor for my lack of planning. I defy you to find something topical in this message or any of its predecessors in this thread. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: shuttle disaster
On Sat, Feb 08, 2003 at 02:57:29AM -0600, Gary Turner wrote: Paul Johnson wrote: Considering the military accounts for over 40% of government spending, I think we found a place to start trimming fat heavily. Especially if our politicians are going to keep claiming we're a peace-loving nation. Peace-loving nations don't spend damn near half thier money on a military. Uh, that's not exactly factual. The 40% you refer must be the amount spent on militant oldsters for Social Security (22%), Medicare (11%), and Medicaid (7%). The portion of the budget spent on defense is 17%.[1] The amazing tales of the amounts supposedly being spent on the military seem to be urban legends that just keep getting repeated. It's about 40% of *discretionary* spending. So, you're both right. Since Social Security, et al. are mandatory programs with separate payroll deductions, they are often excluded from budget discussions. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problem with xlibs?
On Mon, Feb 03, 2003 at 09:39:20AM +0200, Egor Tur wrote: Hi. I have this when compile some programme: /usr/X11R6/lib/libX11.a(XlibInt.o): In function `_XEventsQueued': You're trying the link a *static* library that depends on libpthread. If you link against the shared library, then that should not be an issue. The difference is between: $ gcc foo.c -L/usr/X11R6/lib -lX11 # link against shared library and $ gcc foo.c /usr/X11R6/lib/libX11.a # *incorporate* object archive Thanks. This work. But I don't understand why. I have 2 debian machine and install the same program on them. First: dpkg -l xlibs-dev ii xlibs-dev 4.1.0-17 dpkg -l libc6 ii libc6 2.2.5-14 Second: dpkg -l xlibs-dev ii xlibs-dev 4.2.1-4 dpkg -l libc6 ii libc6 2.3.1-9 On first computer programme build without problem but on second I have mistake with libX11 and pthread. Thanks. Do you have the shared library on the second computer? (/usr/X11R6/lib/libX11.so.6.2 and links libX11.so.6 libX11.so) -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problem with xlibs?
On Sun, Feb 02, 2003 at 11:34:32PM +0200, Egor Tur wrote: Hi folk. I have this when compile some programme: /usr/X11R6/lib/libX11.a(XlibInt.o): In function `_XEventsQueued': XlibInt.o(.text+0x76c): undefined reference to `pthread_equal' XlibInt.o(.text+0x786): undefined reference to `pthread_equal' XlibInt.o(.text+0x7a1): undefined reference to `pthread_equal' XlibInt.o(.text+0x7ba): undefined reference to `pthread_equal' XlibInt.o(.text+0x7d7): undefined reference to `pthread_equal' /usr/X11R6/lib/libX11.a(XlibInt.o)(.text+0xb1c): more undefined references to `pthread_equal' follow collect2: ld returned 1 exit status make[1]: *** [ds9] Error 1 make[1]: Leaving directory `/home/iraf/saoimagenew/saods9/ds9' make: *** [ds9] Error 2 What is this? problem with libX11? What do? Upgrade xlibs-dev? dpkg -l xlibs-dev ii xlibs-dev 4.2.1-4X Window System client library development Whatever you're doing, you apparently need to add -lpthread to the linker arguments. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: invalid date from date -d 1969-12-31
On Tue, Jan 28, 2003 at 10:32:45AM -0500, George Georgalis wrote: On Mon, Jan 27, 2003 at 11:36:36PM -0800, Eric G. Miller wrote: On Mon, Jan 27, 2003 at 10:59:47PM -0500, Stan Heckman wrote: On my system, date -d returns invalid date for dates before 1970. It is possible that this began when I upgraded libc6. Any suggestions? 1970-01-01 is time zero for *nixen. You're asking about what happened before the big bang! Guess date is not as generally useful for reformatting dates as it could be. However, its primary function is to set/print the current date/time which is always more recent than 1970. Guess again. It works fine here... debian 3.0r1 $ date -d 1/15/1905 Sun Jan 15 00:00:00 EST 1905 $ date -d 1/15/1905 +%s -2049994800 $ export LANG=C $ date -d 1/15/1905 date: invalid date `1/15/1905' Mmm, I guess it's a bug in sid's date... -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: XPDF can't find fonts
On Mon, Jan 27, 2003 at 01:58:36AM -0800, Joris Huizer wrote: Hello everybody, Sorry for double posting but it seems nobody has answered on this yet. When I try to read a pdf using xpdf it opens but it shows junk. I get these messages: Error: Couldn't create FreeType font from '/tmp/BvV0Wg' Error: Couldn't create FreeType font from '/tmp/E0vgYp' Error: Couldn't create FreeType font from '/tmp/SdqUZq' Error: Couldn't create FreeType font from '/tmp/eBh9br' Error: Couldn't create FreeType font from '/tmp/WTCNBu' What should I do to get xpdf to work ? Does this happen on all PDF's or just one in particular? Can you read it with GV? -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: invalid date from date -d 1969-12-31
On Mon, Jan 27, 2003 at 10:59:47PM -0500, Stan Heckman wrote: On my system, date -d returns invalid date for dates before 1970. It is possible that this began when I upgraded libc6. Any suggestions? 1970-01-01 is time zero for *nixen. You're asking about what happened before the big bang! Guess date is not as generally useful for reformatting dates as it could be. However, its primary function is to set/print the current date/time which is always more recent than 1970. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: getting time from timeserver on a dial-up
On Fri, Jan 24, 2003 at 12:33:45PM +, [EMAIL PROTECTED] wrote: Hi, I'm using Debian 3.0r1 stable on a dialup in the UK. I'm using PAP authentication that CHAT's to the modem then launches pppd and away we go. What I'd like is for the machine to check and set (if necessary) the system clock based on the time from one of the many timeservers on the internet. What package do I need to do this? How do I make it autorun once pppd has got the ppp connection up and running (I use the pon script) I think you'll like chrony for this task. Out of the box support for ppp connections... -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: functional languages
On Mon, Jan 20, 2003 at 12:07:31PM -0500, David Z Maze wrote: [EMAIL PROTECTED] (Robert Land) writes: On Fri, Dec 13, 2002 at 11:23:43AM -0500, Derrick 'dman' Hudson wrote: imperative and procedural are the same thing, and C is a prime example. It is such because the structure of a C program is a collection of procedures which start with main. Each procedure is a linear list of statements to be executed in order. Could you specify a linear list more clearly? - the contrary would be a nonlinear list which on the first view seems to be self-contradictory. For example, in C: int fib(int n) { if (n 2) return n; return fib(n-2) + fib(n-1); } the rules require that fib(n-2) is called before fib(n-1). In Scheme: This is incorrect. There is no sequence point in the statement: return fib(n-2) + fib(n-1); So, it's up to the implementation which side of the '+' is evaluated first. (define (fib n) (if ( n 2) n (+ (fib (- n 2)) (fib (- n 1) The C and Scheme functions are essentially identical. In C, statements are executed in order. I'm not too up on functional languages, but I seem to recall they need special syntax to execute statements sequentially. That fib function is painfully slow via recursion and susceptible to uncaught overflow in the C version (Scheme has big numbers built in). #include stdio.h unsigned long long get_nth_fib (unsigned long long count) { unsigned long long last2 = 0ULL; unsigned long long last = 1ULL; unsigned long long cur = count; unsigned long long i; if (count 1) { for (i = 1; i count; ++i) { cur = last + last2; if (cur last) { fprintf (stderr, get_nth_fib(%llu) overflows!\n, count); cur = 0UL; break; } last2 = last; last = cur; } } return cur; } Less elegant, but much faster and more robust than the recursive C version. A table look-up would be even faster up to some 'n' (say, 93 in this case). -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Compiler error: C compiler cannot create executables
On Sat, Jan 18, 2003 at 12:07:24PM +0100, Achton N. Netherclift wrote: [snip] According to packages.debian.org, the file that is missing according to the config.logs (crt1.o) is contained in the libc6-dev package. The file is missing on my system, so I attempt to install it: AFAIK crt1.o would be a temporary file created from the combination of conftest.c and some gcc parts when compiling the C file directly to an executable. Your installation is incomplete because you don't have libc6-dev installed. All that stuff about gcc not being able to produce an executable... GCC needs some of the things provided by libc6-dev. Isildur:/# apt-get install libc6-dev ... Sorry, but the following packages have unmet dependencies: libc6-dev: Depends: libc6 (= 2.2.5-11.2) but 2.3.1-9 is to be installed E: Sorry, broken packages The above message confuses me, though. Does that mean that I have 2.2.5 installed and need 2.3.1, or is it the other way around? And why can't I find the file crt1.o? It means, it sees a libc6-dev that depends on libc6 v. 2.2.5-11.2, but apt has v. 2.3.1-9 scheduled for install. Try running apt-get update again before proceeding. -- echo gra.fcw@2ztr eryyvZ .T pveR | rot13 | reverse -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Timing a program run?
On Thu, Jan 02, 2003 at 04:31:39AM -0600, Gerald Livingston wrote: On Thu, 2 Jan 2003 02:07:03 -0800 Paul Johnson [EMAIL PROTECTED] wrote: On Thu, Jan 02, 2003 at 03:57:10AM -0600, Gerald Livingston wrote: How the heck do I time how long it takes a certain script to run? This isn't shell specific. And you're probably going to have to get a surgeon to remove your hand from your forhead from hitting it so hard. 8:o) time command Where the heck is that documented, and where is the time command *AT*? I typed 'time' at the prompt and got a syntax error near unexpected token `newline' I did a locate n/time looking for time in a *bin/ directory -- not there. I searched man bash-builtins -- not there. AHHH -- there it is, in man bash buried in SHELL GRAMMAR -- Pipelines, where it doesn't stand out at all. No. You probably want the time program in package time. It'll live at /usr/bin/time. -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: kernel panic: i have no root and i want to scream
On Sat, Dec 28, 2002 at 04:18:37PM +0900, Elijah wrote: Hello, got this error upon booting to debian, here's some details on the boot [snip] Kernel panic: I have no root and I want to scream --- here's my grub menu.lst title Debian root (hd0,1) kernel /vmlinuz ro root=LABEL=/ vga=788 I've reintalled Debian woody 3 times, and one with the default configs and still got the error. This series of errors occured right after a fresh repartition and reformat of my hd ... Help! AFAIK, change to kernel /vmlinuz ro root=/dev/hda2 vga=788 I'm not familiar with the root=LABEL=/ syntax. The root=foo should translate foo into a device/partition name per Linux naming. -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Blackbox/Fluxbox, no EXIT command
On Sun, Dec 22, 2002 at 11:34:00AM +0800, Robert Storey wrote: Hi all, Seems that the Debian default installation of both Blackbox and Fluxbox doesn't include an Exit command. The only way I can get out of X is to hit ctrl-alt-backspace, which is not very elegant. Does anyone know what I should stick into my .blackbox or .fluxbox directory that can cure this little problem? On the Desktop: Right_Click-WindowManagers-Exit -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: This is incorrect advice (Re: ntpdate from cron -- DON'T DO THAT!)
On Sat, Dec 21, 2002 at 10:12:56PM -0600, John Hasler wrote: Michael D. Schleif writes: granted, wherever feasible, ntpd is technically the best . . . Do you know of some benchmarks comparing ntp and chrony? I thought chronyd did not implement all of the time protocol RFC and had fewer device interfaces for time sources. So, maybe in that sense ntpd is technically the best? -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ps/gs issue ; individual pages look different
See -sPAPERSIZE=letter (or whatever) and -dFIXEDMEDIA. /usr/share/doc/gs-aladdin/Use.htm (if using gs-aladdin) -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: vi alternate problem
On Tue, Dec 17, 2002 at 08:12:29PM -0500, Tom Allison wrote: One day I installed 'vim'. Now I have nothing that is recognized as 'vi' What do I actually install in order to regain this tidy little editor? nvi maybe? Vim has a compatibility mode (vim -v) to more like the original vi... -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: Politics of Java
On Sun, Dec 15, 2002 at 09:09:48AM -0800, Craig Dickson wrote: Eric G. Miller wrote: On Sat, Dec 14, 2002 at 08:46:43PM -0800, Craig Dickson wrote: Yes, but the question is, how usable is it in practice? [snip] People use tasks in Ada on a regular basis. So, it must be usable, neigh? By the same reasoning, concurrency support in C and C++ must be really great, because there are lots of multithreaded programs written in those languages. I've written many myself. Yet, in fact, there is no concurrency support at all in those languages. Yes, it isn't much of an argument for. As I inidicated originally, I just was pointing out the capability but don't have enough experience to argue the merits of one approach vs. another. I've looked at what is involved for C to do threading vs. Ada, and the Ada approach is at least superficially much cleaner (as it is part of the language design). Also, the tasking mechanism is supposed to be available even when the underlying OS doesn't have direct support (say DOS). It is up to the compiler/run-time to handle the platform details... 3 * a call on a protected subprogram of a protected object, providing exclusive read-write access, or concurrent read-only access to shared data; What makes a subprogram protected? Is this a keyword or something that the programmer has to put in the right place, like Java's synchronized? I don't know Java, but protected is a keyword. But, you don't just stick it anywhere anymore than you would stick class wherever. 8 In addition, tasks can communicate indirectly by reading and updating (unprotected) shared variables, presuming the access is properly synchronized through some other kind of task interaction. This sounds like thread synchronization is the programmer's problem, at least for the case of unprotected variables. AFAIK, yes. This all sounds considerably more portable than C threading (which is dependent either on calling OS services directly, or using a third-party library that may not be available on all platforms, or may have licensing issues), but not necessarily all that much better. The crucial point, to me, is that for at least the most common kinds of thread communication and data sharing, it should not be necessary for the programmer to remember to do anything extra (like mark functions or variables with a special thread-safety keyword). Erlang meets this requirement in spades; I don't know of another language that does. Ada is really fairly low-level like C or C++. I have not investigated Eiffel much, but people seem to ooze affection when talking about it. So, I'd be interested to see if/how it approaches the matter. Anyway, different languages meet different needs and none will probably ever be a magic bullet. -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Default fonts for applications?
On Sun, Dec 15, 2002 at 04:44:09PM -0500, Travis Crump wrote: Lloyd Zusman wrote: Perhaps I didn't make my original question clear. Many X apps, including mozilla (which is NOT a gtk app) Mozilla has its own widget layer[xul] to allow for easy cross-platform development, but the xul widgets themselves are implemented with gtk for the default X builds[there used to be a qt and a lower level X implementation, but I am not sure if they are still functional] so at its core mozilla is a gtk app. Yes, gnome/gtk (v. 1.x) font settings affect Mozilla menu fonts. -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: Politics of Java
On Sat, Dec 14, 2002 at 08:46:43PM -0800, Craig Dickson wrote: Yes, but the question is, how usable is it in practice? [snip] People use tasks in Ada on a regular basis. So, it must be usable, neigh? I have never learned Ada, partly because I've never needed to and partially because it's never attracted my interest, in part due to its reputation as the PL/I of the '80s (translation: a vastly overcomplex, I've been recently attracted to Ada due to its ability to catch many more errors at compile time than C (meaning, less bug hunting). I was shocked when I first started using it how many things the compiler would complain about. One tricky bit is the visibility rules, esp. in relationship to operators. Anyway, I think it is much maligned simply as a reaction to its origin w/ the US DOD or because of its Pascal'ish syntax. From chapter 9 of the Ada95 Reference Manual w/ Tech. Corrigendum 1: (http://www.adaic.com/standards/95lrm/html/RM-TTL.html) Section 9: Tasks and Synchronization 1 The execution of an Ada program consists of the execution of one or more tasks. Each task represents a separate thread of control that proceeds independently and concurrently between the points where it interacts with other tasks. The various forms of task interaction are described in this section, and include: 2 * the activation and termination of a task; 3 * a call on a protected subprogram of a protected object, providing exclusive read-write access, or concurrent read-only access to shared data; 4 * a call on an entry, either of another task, allowing for synchronous communication with that task, or of a protected object, allowing for asynchronous communication with one or more other tasks using that same protected object; 5 * a timed operation, including a simple delay statement, a timed entry call or accept, or a timed asynchronous select statement (see next item); 6 * an asynchronous transfer of control as part of an asynchronous select statement, where a task stops what it is doing and begins execution at a different point in response to the completion of an entry call or the expiration of a delay; 7 * an abort statement, allowing one task to cause the termination of another task. 8 In addition, tasks can communicate indirectly by reading and updating (unprotected) shared variables, presuming the access is properly synchronized through some other kind of task interaction. Static Semantics 9 The properties of a task are defined by a corresponding task declaration and task_body, which together define a program unit called a task unit. Dynamic Semantics 10 Over time, tasks proceed through various states. A task is initially inactive; upon activation, and prior to its termination it is either blocked (as part of some task interaction) or ready to run. While ready, a task competes for the available execution resources that it requires to run. NOTES 11 1 Concurrent task execution may be implemented on multicomputers, multiprocessors, or with interleaved execution on a single physical processor. On the other hand, whenever an implementation can determine that the required semantic effects can be achieved when parts of the execution of a given task are performed by different physical processors acting in parallel, it may choose to perform them in this way. -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: OT: Politics of Java
On Thu, Dec 12, 2002 at 10:11:26PM -0800, Craig Dickson wrote: [snip] It really is a cool language; the only one I know of with a really usable concurrency model. (C/C++ have no concurrency model; Java's requires programmers to stick the synchronized keyword in all the right places; Haskell's use of lazy evaluation to model coroutines is cute, but not at all the same thing, just as their two-line qsort demo isn't really a qsort if you look at it closely; etc.) Ada tasks provide concurrency. I'm not enough of a language expert to discuss the merits, but folks seem to use them... And that quicksort looked kinda like a merge sort to me ;-) -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Exchange Calendar client?
On Fri, Dec 13, 2002 at 07:15:25PM -0600, Michael Heironimus wrote: OK, I have a serious question here. I've heard the same type of comment before. And I used to work at a company that used Outlook/Exchange worldwide, including all the shared calendaring and a global address book with the entire company in it (and yes, they got hit VERY hard by the first few big Outlook mail worms). People used it for basic e-mail. People used the address book. But the extent of the group calendaring was that some people would send meeting requests out. I think one person had a public calendar, but most of us hadn't bothered to learn how to access it. And most of the other groups were less technical than the one I was in. Is this pretty typical? Or do other places actually make real use of the group calendars? Or is it that the only people who really know how to use all the features are the people who aren't doing real work (like the phone system in some places)? I'd say that's pretty typical. Sometimes bosses delegate to clericals to arrange meetings and it's nice when you can look up everyone's availabilty. However, that falls apart as soon as you add an outside party to the meeting. Many of the people I work with still use Franklin planners to keep track of meetings (it's difficult to check your electronic calendar just anywhere; and synchronizing with little handhelds all the time is a pain...). -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: gcc complains there is now ada compiler installed
On Wed, Dec 11, 2002 at 10:34:43PM +0100, Martin Hermanowski wrote: Hello, I want to compile an ada program, it works if I use `gnatmake', but if I want to use a Makefile or cook, gcc -c won't compile: , | (22:29:58)(#1,x0)martin@pegasus:~/priv/dev/ada (626)cook | /* /home/martin/crypto/priv/dev/ada/Howto.list */ | cook: gcc -c main.adb | gcc: main.adb: Ada compiler not installed on this system | cook: command gcc: exit status 1 | cook: main: not done because of errors ` How can I get the correct behaviour? Use gnatmake in your makefiles. AFAIK, gcc can't make an executable from Ada source w/o some help anyway... But, are you sure your gcc points at an Ada capable version? -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GSview on Debian anybody?
On Tue, Dec 10, 2002 at 01:24:18PM +0100, Volodya Mumariv wrote: Stephen Patterson wrote: On 9 Dec 2002 12:34:10 -0800, Volodya Mumariv wrote: Did anybody succeed to pack GSview for Debian or install it in any ather way? Yes, there's a gv package (from the package info) Description: A PostScript and PDF viewer for X using 3d Athena Widgets gv' is a comfortable viewer of PostScript and PDF files for the X Window System. . It uses the ghostscript' PostScript(tm) interpreter and is based on the classic X front-end for gs', ghostview'. It is more comfortable and more powerful than ghostview'. Sorry, I was not exact. I need GSview because of its functionality (through gs of course) to select and convert certain part of ps file into high resolution raster graphics. Is gv able to save ps as high resolution jpg, tiff or png for example? If not, what are debian people using for that purpose (except CL driven gs)? There's no handy interface I'm aware of like w/ GSview, but all the actual work is done by gs. Look up device selections, resolution selections and -dTextAlpha and -dGraphicsAlpha. This is clearly a case where a nicer GUI would be useful... -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: acroread and anti-aliased text
On Tue, Dec 10, 2002 at 08:59:43PM -0600, Gary Turner wrote: Alan Shutko wrote: Gary Turner [EMAIL PROTECTED] writes: Not sure that's the only cause. Documents created by LaTeX and converted to PDF have the same problem. True, but the problem with (naively created) TeX documents is that dvips traditionally puts bitmapped fonts into its ps files (as PS Type 3 fonts). Certainly a naif here. What is the more sophisticated approach? I've had good luck with dvipdfm. Understands hyperref, no messy conversions of eps files (use graphicx), no pdflatex headaches... -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Compiling gsview
On Wed, Dec 11, 2002 at 07:31:49AM +0530, Sridhar M.A. wrote: Hi, I am trying to run gsview 4.3. When I try opening a ps file, it says that it cannot find libgs.so. A quick search on debian's packages page did not yield any package that contained the above mentioned file. Is there any package/metapackage that provides libgs.so so that I can use/compile gsview? It appears the package gs-aladdin just ships as a single executable. You may have to build from source to get a shared library. How did you install gsview (it doesn't appear to be packaged for Debian)? -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: acroread and anti-aliased text
On Tue, Dec 10, 2002 at 11:25:51PM -0600, Gary Turner wrote: Eric G. Miller wrote: I've had good luck with dvipdfm. Understands hyperref, no messy conversions of eps files (use graphicx), no pdflatex headaches... Looking at the man page on this is encouraging. It does, however, throw a spotlight onto my vast ignorance. Can you suggest a default set of arguments for the '-f' option? (the equivalent of Computer Modern?) If I read the man right, this option should do the job of putting some type1 fonts into the pdf file. It's been a little while since I've done any texing, but I never messed with the fontmap argument. Just like magic it uses the Type1 CMR fonts (or Times, Helvetica, Palatino, etc..) Maybe, see /etc/texmf/dvipdfm/config ...? -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to escape find result
On Fri, Nov 29, 2002 at 08:52:36PM -0500, Try KDE wrote: Hi, I'm trying to run the following script: for f in $(find . -name *.txt); do cmd1;cmd2;cmd3; done ,where cmd1, cmd2 and cmd3 are arbiturary commands. The problem is, if the found file names contain a space , for example part1 part2.txt will be interpreted as part1 and part2.txt. I'm thinkin an esacping tool may be able turn it into part1\ part2.txt before passing it to for loop. Any thought? Does: IFS=^M for f in $(find . -name *.txt); do cmd1; cmd2; cmd3; done work? (replace ^M with real newline) There's also the trick of: for f in *.txt; do echo $f ; done -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Stripping EOL feeds...
On Sun, Nov 24, 2002 at 10:32:14PM -0600, ZephyrQ wrote: I'm trying to format the debian install manual (text version) for printing and I'm trying to save a couple of trees. Is there an easy way to strip the line breaks so the text will come out unformatted? This way I can reduce the font and print whole pages of itty bitty debian install text which, in my own sick way, helps me find info I need quicker... An alternative approach. Use something like enscript to format it into postscript with two pages per physical page. Then print duplex... -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: (fwd) unsubscribe
On Sat, Nov 23, 2002 at 05:31:08PM -0500, Darryl L. Pierce wrote: On 2002.11.23 16:30 Pigeon wrote: Is he silly? Another jackass ... is it against the law to hit stupid people? Oliver See Romans 12:19 Then try Isaiah 45:7 and Psalm 137:9. Forgive them first ... and then kill them -- Dobbs, 1954, radio broadcast on capital punishment, WFMU, East Orange, New Jersey Stang, Rev. Ivan (Ed.). _Revelation X: the Bob Apochrypon_ 1994. Fireside. New York. p. 47 -- ...the plural of anecdote is [not?] data. - attrib. to George Stigler -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ld and g2c
On Mon, Nov 18, 2002 at 02:48:39PM +0200, Jerome BENOIT wrote: Hi All, Whereas `g++' can find the library `g2c', ld cannot: why ? did I miss something ? It's not in the usual places ld looks, but g++ has magic for C++ (/usr/lib/gcc-lib/arch/version/libg2c.a). -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Setting default width of xman
On Fri, Nov 15, 2002 at 08:38:33PM +0100, Michael Naumann wrote: More generally, how can I find the available resources for a particular X-client? This is something I'd also like to know. I too often end up with experimenting. In the case of xman, the widget hirarchy can be obtained with man xman You can play with Editres ; figuring out valid values for resources is another matter... -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: filtering unicode files
On Wed, Nov 13, 2002 at 02:49:29PM -0300, Eduardo Gargiulo wrote: Hi all. I´m trying to use grep to filter a string on a unicode file. If I make a cat on the file, i can see the contents, but if i pipe a grep after the cat, the string (exists in file) didn't match. Is it possible to use grep on unicode files? How? Is there any other tool to do that? What do you mean by unicode? The man page of grep says it looks to LC_CTYPE (among others) to determine how to interpret the file. This could make a difference. -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Modem connection.
Fix your dang clock already! I'm quite sure this month is November not January ;-) -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GLX and undefined references
On Mon, Nov 11, 2002 at 10:38:14PM +0100, Roman Joost wrote: [snip] gcc lesson06.c -o lesson06 -L/usr/X11R6/lib -lGL -lGLU -lXxf86vm You probably need to add -lX11 for any X proggy (and maybe others). -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: pan from unstable too unstable??
On Wed, Nov 06, 2002 at 01:03:40AM +0100, daniel paranhos zitterbart wrote: Hej guys I'm actually using debian SID and pan to read my news(eg debian-user). Reading function very good, but if want to post a followup, to a mail address or to an news server, that doesnt matter, pan crashes. I think i've configured everything right. is that a known problem? anybody running pan from unstable and everything(eg posting to newsgroups via mail) is working stable?? See: /usr/share/doc/pan/README.composer -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] - Interesting politics and the GPL
On Wed, Oct 23, 2002 at 03:36:05PM -0700, Craig Dickson wrote: government funding. That's a legitimate issue. Software developed by the government, or under government contract, ideally should be considered to be in the public domain; the government, representing the people, paid for its development, so one could reasonably argue that it should be freely reusable by any American citizen for any purpose, including Title 17 U.S.C. § 105. Subject matter of copyright: United States Government works Copyright protection under this title is not available for any work of the United States Government, but the United States Government is not precluded from receiving and holding copyrights transferred to it by assignment, bequest, or otherwise. Obviously that isn't the whole story (and this only only the federal government...). -- static const char signature[] = Copyright (c) 2002 Eric G. Miller [EMAIL PROTECTED]; -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Detecting Net Apps
On Mon, Sep 30, 2002 at 10:13:25AM +, Aurelio Turco wrote: Is there some easy way to detect whether a particular binary contains code to access the network (as client or server)? $ nm path_to_binary | grep socket Course, having sockets code doesn't necessarily mean network (could be just local UNIX sockets...). -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Odd Path issue
On Thu, Sep 26, 2002 at 01:55:40PM -0500, Kent West wrote: I'm using bash. echo $PATH reports: ~/bin:/usr/local/bin:/usr/bin:/bin:/usr/bin/X11:/usr/games s/~/$HOME/ Guessing the ~ expansion in $PATH is a problem... -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ASSISTANCE/ You are free to investigate about my family background, especially my father wealth, but Treat this matter with utmost confidentiality and if for personal reasons this proposal does not meet your approval, please keep my confident and disregard this mail. I look forward to your quick response.
On Wed, Sep 25, 2002 at 05:25:37AM -0400, Mark L. Kahnt wrote: On Tue, 2002-09-24 at 22:01, MR REUBEN SAVIMBI wrote: My Dear I am MR REUBEN SAVIMBI one of the favorite sons of Mr. JONAS SAVIMBI (The Rebel leader) ... [***SNIPPAGE OF SCAMAGE***] Treat this with the utmost confidentiality? and yet it is being sent to whole mailing lists, as well as any other harvested email addresses that can be found? These scams were old when I first got one from Nigeria in 1997 - now that it is in email, it should be obvious by the third one received that this is just some template to encourage the collection of money for service and legal fees, and outright bribes. To anyone on this list - if you wondered for a moment about the legitimacy of any of these, put it out of your mind - it isn't, and at best, it is asking for money laundering assistance, which at least in North America is illegal (however, IANAL.) Just read a Secret Service guy in Sacramento, CA was just busted for perpetrating this Nigeria scam. I guess he figured he saw enough people get taken by it, he'd try running it himself... http://www.sacbee.com/content/news/story/4528593p-5548095c.html -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: hosts.deny a directory???
On Fri, Sep 20, 2002 at 08:41:16PM -0700, Kenward Vaughan wrote: In today's system upgrade I got the following immediate message: ... Preconfiguring packages ... egrep: /etc/hosts.deny: Is a directory fgrep: /etc/hosts.deny: Is a directory ... 'Tis true, and at least one other /etc file is now a directory ( /etc/passwd- ). The hosts.deny directory contains one file, %gconf.xml The passwd- directory has a single file as well, cookie I've had some problems recently with /etc getting mauled for unknown reasons, requiring pulling out backups to fix things. The file system for / has been messed up several times as well, for unknown reasons. Should I consider reinstalling the system, or can I tell how clean things are in some way to assess what the hell is going on? Sounds like the system is really munged up, possibly filesystem corruption or failing hardware? -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Just a tip for you zip100 users
On Thu, Sep 19, 2002 at 04:11:37AM -0500, John Manko wrote: I've been running RedHat for a few years now, but decided to change to Debian to test myself, :). And I've been learning a lot. [snip] Do you notice that the zip (hdd) drive is actually partitioned (hdd4)? so, instead of mounting Same with Jaz drives. From my /etc/fstab... /dev/sda4 /jaz vfat defaults,noauto,rw,user,unhide 0 0 (Guess I should delete that since the Jaz is dead...). -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Network error between linux kernel 2.4.7-10 and Sun solaris 6
On Wed, Sep 11, 2002 at 10:08:21AM +0200, Francois Chenais wrote: Hello, I have a network problem between a rh 7.2 with kernel 2.4.7-10 and a sun solaris 6. I know this error doesn't directly concern debian but it seems a Linux pb. Any experience about it ? It would help if you described the problem. However, a Debian solution might be different than a Red Hat solution. -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Writing GIFs
On Wed, Sep 11, 2002 at 09:16:02PM -0600, Bob Proulx wrote: Craig Dickson [EMAIL PROTECTED] [2002-09-11 10:55:37 -0700]: I realize I am a terrible person for wanting to violate Unisys's patents, but I'm working on a web site that I think will often be visited by people with old browsers (it's the site for the elementary school my daughter attends), so rather than use PNGs, I think I'd be better off with GIFs. I have always prefered jpegs since they load faster. I like that fact that most browsers will show you the image as it downloads and therefore it feels faster to me on slow connections. The JPEG encoding isn't very good for text and line art. If you turn compression on the JPEG to 0, it'll work but then the size would likely be greater than a comparable GIF or PNG. Don't all browsers support .jpg format? (I really don't know but had always assumed so.) Last I checked, lynx and links don't ;-) Remember those alt=BUY THIS NOW!!! attributes for our sight impaired folks as well. -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Using exim efficiently on a dialup machine
On Mon, Sep 09, 2002 at 09:49:00AM +0200, Joerg Johannes wrote: Hi List Some weeks ago I switched from kmail to mutt/exim. At first I did not realize that exim did not deliver several e-mails, some to the list, and some private ones. I saw using exim -bs -bp, that those mails are marked '*** frozen ***'. I don't know why this is the case, I'm sure I was online when sending them from within mutt. So, how can I defrost the frozen messages and get them sent? How do I configure exim so that it looks if ppp0 is up and then retries to send the messages instead of freezing them? (Is this possible at all?) Or what is the favoured way for a dialup machine to send mails from within mutt? The messages might be frozen because after an extended period of time, exim was unable to deliver them. A standard setup to use exim from a dialup is to have it send messages via your ISP's SMTP server (rather than trying to deliver them itself). If you're off-line, and send a message via mutt (or whatever) the messages are normally queued for delivery. By default, there's probably a cron script that runs exim every five minutes or so (unless it's running as a daemon). Additionally, /etc/ppp/ip-up.d/exim should be run when the connection comes up, and then flush the queue. So, long story short, either exim is misconfigured for delivering mail, or the messages couldn't be delivered (network outages, unresolved hosts, etc...). You might want the -Mt argument for thaw message(s) with ids ... The messages will be located in /var/spool/exim/. You might also check /var/log/exim for info about the delivery failures. Apparently there is also an auto_thaw configuration option... -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: consistant userid:username setup for system accounts?
On Sun, Sep 08, 2002 at 12:24:06PM -0700, nate wrote: hi I have ran accross this a few times in the past but was curious if there was any effort in Debian 3.0 that goes to assigning a uid number to the system accounts to keep them consistant accross installations. Many of the accounts already are(the core ones) but at least in my case I had some trouble migrating a cyrus installation from one system to another because on the system it came from(potato) cyrus was uid=100, on the woody system it was going to cyrus was uid=103 and sshd was uid=100. On my system at home which is woody cyrus is uid=101 and sshd is uid=102. It would really be nice if there was an assigned method(sort of like assigning ports) for keeping uid numbers consistant. This would also help a great deal with distributed authentication such as LDAP or NIS. at least it would help when working in a mostly debian networked enviornment. The Debian policy manual section 10.2 covers users and groups. Specifically, the uid/gid numbers 0-99 are explicitly assigned via base-passwd, 100-999 are dynamically assigned system uid/gid's. So, it seems that any package that doesn't need an explicit uid/gid gets the next available system uid/gid via adduser --system. I would assume that system packages are only thought to need consistent id's when they actually own files (vs. root.root config files). One hundred id's isn't very many to go around. -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Anyone know how to get *bigger* fonts on the console?
On Sat, Sep 07, 2002 at 03:08:33PM -0700, Wade Richards wrote: On approximately Fri, Sep 06, 2002 at 06:43:26PM -0700, Eric G. Miller wrote: Some of the fonts in /usr/share/consolefonts are quite interesting... I've been playing a with those fonts, but they appear to all be 8 pixels wide, and between 8 and 19 pixels high. I'm looking for BIG fonts, like 16 pixels wide, and 25 or 30 pixels high. 1) Is it even possible to have wider than 8 pixel fonts in the console? 2) Where I can find such large fonts? 3) What font-editing software can you recommend to help me create such fonts? I don't know if there are any huge console fonts packaged for Debian. There is a console font editor called fonty thay may help you make your own... -- begin 664 .signature MF5L;E-(Y'(-ID4@/G1E;BYSI`,FUG93P)`@(`@(`@(`@(`@( M`@(`@(`@(`@(`@(`A%FEC($N($UI;QEB`\96=M,D!J',N;F5T /@H` ` end -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Screen blanking on logout
On Fri, Sep 06, 2002 at 02:43:06PM +0200, Ian Johannesen wrote: On Fri, Sep 06, 2002 at 02:14:02PM +0200, Raffaele Sandrini wrote: Hi Raffaele, [...] I did not find an option wich lets getty blank the screen so wich are my options to get that feature? write a script wich firstly runs clear and the getty and take that one in the inittab or do i have to use mingetty to get the screen blanked? Since I'm using bash2 I normally do echo 'clear' /etc/skel/.bash_logout But if you only want it for one local user you could just replace the above with: echo 'clear' ~/.bash_logout .bash_logout is executed when logout is issued. Also, something like linuxlogo resets the console. -- Eric G. Miller [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Logout script
On Sun, Jun 30, 2002 at 09:36:12PM -0400, Kapil Khosla wrote: Hi, In my current setting if I press ctrl-alt-del, my system shuts down. I saw something in /etc/inittab like # What to do when CTRL-ALT-DEL is pressed. ca:12345:ctrlaltdel:/sbin/shutdown -t1 -a -r now I want to logout and restart my KDM and window manager by pressing CTRL-ALT-DEL, Can anyone explain how can this be done, The window manager I use is sawfish and I use KDM to login, Note: Ctrl-Alt-Backspace will kill the X session under the default setup. After which, KDM should respawn. However, this is rarely what you want to do, since it can muck up your session management. What you probably want to do is map a hot key sequence in X, that'll perform a logout. I'm remembering that Sawfish has an easy way to setup such key sequences... You almost certainly don't want to mess with inittab for such functionality... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Question regarding DL of Debian
On Fri, Jun 28, 2002 at 12:14:30AM -0400, Ian D. Stewart wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thursday 27 June 2002 23:07, Paul Johnson wrote: On Thu, Jun 27, 2002 at 06:29:15PM -0400, David J. Weaver wrote: I am in the US, but it sounds like (on the website) that I have to download Non-US files to get US encryption? Is that right? Yes. They do this to get around the fact that the US has some rather braindead cryptography laws... Granted. But they are decidedly less braindead than they were ten years ago... s/laws/regulations/ AFAICT, it was/is the implementing regulations rather than the actual law where the cryptography exports were/are restricted. Though regulation /may/ actually be harder to change than the enabling legislation unless the changes are ramrodded from the top down or a court invalidates the regs. There's apparently a treaty relating to this matter called the Wassenaar Arrangement and some Executive Orders thrown in for good measure... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: stack and heap size documentation
On Fri, Jun 28, 2002 at 05:57:36PM -0700, Paul Scott,,, wrote: I have been searching for documentation on default stack and heap sizes with g++/gcc and how to change them if necessary. I haven't had any luck so far. I have been getting segmenatation faults possibly related to my use of the new operator. Your probably corrupting memory somewhere. AFAIK, new should only fail if you're totally out of memory (and then do so in a predictable manner). Segfaults are rarely due to programming errors in the compiler and standard library. Start by compiling with warnings turned up and debugging symbols turned on... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Ot c++ programming in linux
On Fri, Jun 28, 2002 at 06:17:14PM -0700, faisal gillani wrote: Well i am a newbie learning c++ these days we are being thaught on turbo c 3.0 but as like other things i want to work on c++ in linux .. so i installed gcc on my linux box but i dont have any idea how to install it for example i write a program as follows in turbo c #includestudio.h typo #includeconio.h header not found void main (void) undefined behavior { printf(hello world); no return statement } how do i write the same program in gcc ? news:comp.os.linux.development.apps can help, as can a few good books. BTW, you want g++ if you really want to do C++... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: hdd change
On Fri, Jun 21, 2002 at 01:26:15PM -0500, Patrick Wiseman wrote: On Fri, 21 Jun 2002, Patrick M wrote: Is there a reason for using 'tar' instead of 'cp'? It preserves ownership and symbolic links, for one thing (or 2). The GNU cp command does those... cp -a. Other cp variants maybe don't... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: allowing telnet for only a few users?
On Sun, Jun 23, 2002 at 09:28:32PM -0400, Joey Hess wrote: Eric G. Miller wrote: On Fri, Jun 21, 2002 at 10:38:08PM -0400, Joey Hess wrote: Does anyone know how to set up telnetd so only a couple of users can use telnet to log in, and the rest must use some other, more secure method, such as ssh? I have a few secure guest accounts that I want to allow telnet for, while disabling it for everyone who can get to a shell. Looks like pam_listfiles can do this... That's exactly what I was looking for, but it seems that there is no way to make pam differentiate between login by telnet, and logins at the console, and other uses of /bin/login. Rats. I haven't looked into it, but there are examples for ftpd. So, does it not work for telnetd? That is, have a rule like: telnetd auth required pam_listfiles.so\ onerr=fail item=user sense=allow file=/etc/telnetusers Or is telnet not PAM aware outside of login?? I'm no PAM expert... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: man or info?
On Sat, Jun 22, 2002 at 12:19:16PM -0700, Osamu Aoki wrote: Another one is information on libc. apropos libc does not offer good pointer to libc in manual page. info libc gives no clue. It is info Libc. Why capitalize this?? I have no clue. info libc works fine here... (sid). I believe it should be the same in woody, don't recall about potato... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: allowing telnet for only a few users?
On Fri, Jun 21, 2002 at 10:38:08PM -0400, Joey Hess wrote: Does anyone know how to set up telnetd so only a couple of users can use telnet to log in, and the rest must use some other, more secure method, such as ssh? I have a few secure guest accounts that I want to allow telnet for, while disabling it for everyone who can get to a shell. Looks like pam_listfiles can do this... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: using old Adobe post script fonts
On Sun, Jun 16, 2002 at 07:51:44PM -0400, David Teague wrote: Hi I have 3.5 inch floppies containing installation files for Adobe postscript fonts (these were for Windows 3.1). [snip] I paid about $100 for this stuff. I would like to use them. I hope some one can tell me how to use these with Linux. My wife would be overjoyed to be able to use them with here Windows box. Mmm, are any of those fonts different from the Adobe fonts you probably already have installed? (Helvetica, Bookman, Courier, Times, Zapf Dingbats, etc...) Anyway, you can use them. You might want to see some of the Font HOWTO's for details. Briefly: Copy them to some directory. Change to that directory and run type1inst, make the directory available to X and ghostscript... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: appearance of xfce on Debian
On Fri, Jun 14, 2002 at 11:06:44PM -0700, Jeff wrote: Michael P. Soulier, 2002-Jun-14 23:19 -0400: On 14/06/02 Kevin C. Smith did speaketh: It sets the bar below the screen for some reason. edit ~/.xfce/xfwm-session for [WM_NAME] XFce Main Panel change [GEOMETRY] Mine looks like this: [GEOMETRY] 110 722 808 46 0 0 See if that works. This happened to me too, on both my desktop and my laptop. Mike I decided to check out xfce and this is happening to me too, but the suggested action above does not help. It keeps resetting back to the default, which pushed the panel off the screen. Any other suggestions? Log out from X and edit it from a console session. My guess is your editing it from within an xfce session, and your changes are getting clobbered when you end the session. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Unix-only analog to WinPopup?
On Sat, Jun 15, 2002 at 01:55:39AM -0700, nate wrote: quote who=Ron Johnson Using apt-cache search doesn't show much. (Unless I'm using the wrong keyword to search by!) while i am not certain what your end goal is, rpc.walld is a cool program, it won't do a popup(as far as i know) but it can broadcast a message out to connected servers on a network. i didn't even know it existed until i saw it in action on one of my sun boxes, a NFS server broadcasted to the clients that it was shutting down. pretty neat (to me). i am not sure how(or if) to send a message using it .. I think xmessage will work over a network via the -display option. Course, you need permission to connect to the remote X server... Missed the original post, so not sure if that's what your looking for... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: appearance of xfce on Debian
On Fri, Jun 14, 2002 at 12:58:58AM -0400, tvn1981 wrote: Hi, I apt-get install xfce and when I start it . I cannot see the xfce main bar. Why is Debian setting up xfce this way ? How can I get the normal xfce main bar ? How are you running it? See also, xfce_setup (which doesn't have a manpage, but basically creates an ~/.xsession and some other dot files). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Nasty X + Mozilla bug...
See http://theregister.co.uk/content/55/25689.html for a nasty X crashing (and possibly machine crashing) bug regarding huge fonts. Guess it's back to links/lynx/w3m or whatever for a little while. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ecc, error correcting code insertion
On Wed, Jun 12, 2002 at 02:00:35PM +1000, James Cameron wrote: Is there a Debian package that contains code that will add error correcting codes to a data stream (e.g. a pipe) so that loss of a segment of the stream can be recovered from? ?? How do you expect to recover from an error reading/writing a pipe? You have EAGAIN, and SIGPIPE to catch. Also, a segment written to a pipe will never got lost unless the OS is fubar'ed or one of the ends has closed the stream in which case, there's probably nothing to do about it (except catch SIGPIPE). Presumably you want a programming library interface? It's just not clear what you're really looking for (CRC32?). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Swapfile
On Mon, Jun 10, 2002 at 09:07:04PM +0200, Jens Karsten Müller wrote: Hello, today i created a swapfile instead of a partition, because i think it's more dynamic. I only see advantages using a swapfile. So first of all, I'd like to know about disadvantages, because almost everybody uses a swap partition. And finally I need to know what to change on my potato. I just wrote the following line into /etc/init.d/mountall.sh: swapon /swapfile This does work, but I'm not sure whether this is the right place for it. So is there a better place for it? Put it in /etc/fstab. The reason people tend to use a dedicated partition is it can be a little faster (even better if it's on a lesser used disk). Example fstab entry: # extra swapfile /var/local/swapfile none swap sw,pri=1 0 0 When mountall.sh runs, it'll pick up the entry... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how to also remove files that were required for some package ?
On Sun, Jun 09, 2002 at 04:54:39AM -0400, tvn1981 wrote: Hi, just wondering if apt can do something like this when I get package A, it also fetch the required packages B and C now when I want to remove A, can apt intelligently remove B an C also ? (assuming no other package require B and C) AFAIK, this isn't possible, or necessarily desirable in the general case. However, have a look at deborphan. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X nice value
On Sun, Jun 09, 2002 at 02:23:55PM +0100, Hereward Cooper wrote: Hi, When I configure/reconfigure xserver-common I give debconf my nice value for X, but when ever I 'startx' it says it is reverting back to a value of 0, no matter what value I give it. How can I force X to use my nice value? Only root can do nice 0, so I think the nice trick really only works if you run a display manager (or startx as root). Basically, as a luser, nice just allows you to lower the priority of your processes. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Force password change on first logon on Debian Box
On Sat, Jun 08, 2002 at 12:40:10AM -0300, [EMAIL PROTECTED] wrote: I need to force the users to change their password on first logon. How can I do this on a Debian Box? $ passwd -e username Seems the adduser script doesn't have a way to force this as the default behavior?? -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian GIS
On Sat, Jun 08, 2002 at 09:58:52PM -0400, Ian D. Stewart wrote: On 2002.06.05 23:50 Eric G. Miller wrote: GRASS is best for folks doing raster analysis/modeling. The vector stuff isn't quite there yet. Are there others that do better at vector-based modeling, or would I be best served waiting on GRASS to 'get there' ? I'm not aware of any free software that has what I'd consider sufficient support/functionality in the vector database and modelling area. GRASS does support vector data, it's just pretty limited in what you can do with it. GRASS 5.1 (in development) has a new data model and has seen a great deal of work in this area, but will probably be quite some time before the whole thing stabilizes (GRASS 5.0 stable is just being released ...). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: utempter.h ?
On Fri, Jun 07, 2002 at 02:39:50PM +1000, Chris Kenrick wrote: On Thu, Jun 06, 2002 at 11:13:55PM -0500, Ted Goodridge, Jr wrote: Which package is utempter.h in? X needs it to build xterm... Thanks in advance, None of them, according to http://www.debian.org/distrib/packages. At the bottom of that page, you'll find a facility that allows you to search the package contents for given files, and no joy in this case. utempter is not in Debian, and the X configure shouldn't try to use it for a Debian build, AFAICT. Ferretted that out of some mailing list threads that google brought up for utempter. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian GIS
On Wed, Jun 05, 2002 at 01:17:04PM -0300, Henrique de Moraes Holschuh wrote: On Wed, 05 Jun 2002, Ian D. Stewart wrote: Can anybody recomend a good GIS (Geographical Information System) package for debian? I did a quick search with 'apt-cache search GIS', but got a long list of unrelated results. I've found GRASS on freshmeat, but there doesn't appear to be .deb available. IMA (www.ima.sp.gov.br) is currently sponsoring the packaging of some GIS related software, but I hear MapServer and GRASS are f* painful to package, and a major mess of dependencies, too. GRASS is easy enough to build from source (probably easier than dealing with binaries). There've been some recent experimental debs, but I don't know the status of them currently. Anyway, most of the dependencies can easially be met in woody/unstable. You might want to grab libgeotiff and libgdal from remotesensing.org... It's true, GRASS is somewhat difficult to package (this is changing...). GRASS is best for folks doing raster analysis/modeling. The vector stuff isn't quite there yet. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: port forwarding
On Mon, Jun 03, 2002 at 11:21:28AM -0700, ben wrote: On Monday 03 June 2002 05:01 am, Paul Johnson wrote: hey ballo, for the last couple of days, your posts are showing up as msg.pgp attachments; i.e., the attachments have to be viewed in order to see the msg. Probably a function of the mail reader. Mutt shows them inline... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: onshore-timesheet
On Tue, May 28, 2002 at 09:55:02PM -0400, Harland Christofferson wrote: is anyone successfully using this package? i have loaded it but get the error: Sorry, I cannot complete your request: Cannot connect to backend: localhost:5432:timesheet: no connection to the server Haven't used it, but I recognize the port as being the same as PostgreSQL. So, does this package require PostgreSQL, and if so, is it running and listening/accepting TCP/IP connections? -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Color black not defined
On Mon, May 20, 2002 at 09:20:14PM -0700, John Galt wrote: Hi, When I run xemacs , i get an warning message color black not defined and in the text of emacs, my cursor is not visible which I assume is due the warning message. Do you have a customized .Xresources? I've found trailing spaces are not tolerated, so check for black . This seems to be an X thing. Not sure why xrdb doesn't strip trailing white space (bug?). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Why wdm opens a TCP port?
On Fri, May 17, 2002 at 08:54:04AM +0300, Petteri Heinonen wrote: Yes, it's not the _X_ which keeps open the port, but the _wdm_. But the port indeed changed every time I restarted the wdm. I just restarted it couple of times, and the port numbers were 1083, 1084, 1085, etc. Very odd... It's not that odd really. What address range is reported? I'm going to guess it's only listening on the local loopback. Also, try telnetting to the address:port. Does it hang and time out? Do you get connection refused? It would seem WDM sets up a TCP port to communicate between a parent and child (more than one WDM process running?) and therefore doesn't really care what port is used (lets the OS decide). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Why wdm opens a TCP port?
On Fri, May 17, 2002 at 10:45:36AM +0300, Petteri Heinonen wrote: Netstat reports that foreign address is *:*, and I indeed _can_ connect to it with telnet, altough if I send anything to it with the telnet, the connection gets closed. Further, there is only one wdm process running. Maybe I'll just trust my iptables :) But anyway it would be nice to know what is the purpose of this kind of behaviour, and how can it be turned off. Okay, well that doesn't sound very good. I'm not sure why WDM would want a TCP socket open on a random port, when a UNIX socket or even pipes could probably meet the need for what it's doing without opening a public port. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Administrator
On Fri, May 17, 2002 at 12:17:51PM +0100, Karl E. Jorgensen wrote: On Thu, May 16, 2002 at 05:46:57PM -0500, [EMAIL PROTECTED] wrote: Thank you for your post. Pastor Chick is very busy processing Emails and Internet tape orders. Please be patient as he will respond to you as soon as possible. Creation Ministries Staff Does this mean that we're looking for help from above in order to get woody out the door? Pastor Chick is somewhat famous for his simplistic christian cartoon books. There are a number of quite hilarious parodies. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Character Set trouble
On Fri, May 17, 2002 at 06:04:17PM -0700, Ross Boylan wrote: I am seeing the ? character instead of actual accented characters in mail I get, even if it has Content-Type: text/plain; charset=iso-8859-1 I am using mutt in a KDE console on woody, so I'm not even sure what component I should be looking at. When I visit my mail file in emacs the accented characters also don't appear properly, though I see several gibberish characters instead of a simple ?. I'd appreciate a pointer about how to fix this up. Thanks. man locale export LANG=en_US (or something using 8859-1). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Administrator
On Fri, May 17, 2002 at 08:47:01PM -0500, Ron Johnson wrote: On Fri, 2002-05-17 at 20:00, Eric G. Miller wrote: On Fri, May 17, 2002 at 12:17:51PM +0100, Karl E. Jorgensen wrote: On Thu, May 16, 2002 at 05:46:57PM -0500, [EMAIL PROTECTED] wrote: Thank you for your post. Pastor Chick is very busy processing Emails and Internet tape orders. Please be patient as he will respond to you as soon as possible. Creation Ministries Staff Does this mean that we're looking for help from above in order to get woody out the door? Pastor Chick is somewhat famous for his simplistic christian cartoon books. There are a number of quite hilarious parodies. I wondered if he was the Chick Tracts guy. Still, the nickname Chick isn't uncommon, so didn't want to assume... Could be I jumped the gun on that. Not sure if Jack Chick considers himself a pastor And don't know (or care) if he's connected with Creation Ministries. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Why wdm opens a TCP port?
On Thu, May 16, 2002 at 03:01:23PM -0400, Noah Meyerhans wrote: On Thu, May 16, 2002 at 10:34:32AM +0300, Petteri Heinonen wrote: Does anybody know how to configure wdm so that it doesn't open TCP port 1030 when started? I guess this is not a big security threat, but I've tried to keep open ports in minimum, and I can't see why this one would be necessary either. I have to admit that I don't know the purpose of this open port, but for example gdm, which I've used in some other boxes, doesn't, at least by default, open any ports. Edit /etc/X11/wdm/Xservers and add '-nolisten tcp' to the line that looks like :0 local /usr/bin/X11/X See if that fixes things. If not, then it's a bug and I must fix it since I am the wdm maintainer. Only thing is, the X ports are 6000-7, according to /etc/services. Sure the 1030 port is not for some local use only? Is it always the same port? The ports doesn't appear to match any reserved ones in /etc/services... Don't use wdm myself, so... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Setting effective UID for a shell script
On Wed, May 15, 2002 at 04:31:09PM +0930, Tom Cook wrote: On 0, Alberto Cabello Sanchez [EMAIL PROTECTED] wrote: As far as I know, you can't. I think you have to setuid() and exec() in a, say, C compiled program. No, that's when you want to do it the other way around. No. 'Cause if you don't have the priveledge already, a call to setuid won't give it to you. But, you can give up priveledge. If you are root and want to run it as someone else, then: su -c cmd will do it without asking for a passwd. If you're root, and want to run the command as another user, you have to specify which user. That is what the OP wanted to do. If you are not root and want to run it as root, then: * Use one of the setuid script wrappers around. I don't know of any off the top of my head, but they must be around. * Write your own setuid script wrapper in C. It's not hard - it goes something like this: #include unistd.h int main( int argc, char** argv ) { execve( argv[0], argv[1] ); return 0; } chown root setuidscriptwrapper chmod u+sx setuidscriptwrapper or something of that order. Control *very* closely who has permission to run this script. Lucky for you, you're wrapper won't compile. Semantically, it'd invoke endless recursion anyway. All around, this is a bad idea. You might as well remove the root password. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Setting effective UID for a shell script
On Wed, May 15, 2002 at 05:17:41PM +0930, Tom Cook wrote: However, you are not entirely correct. This does, in fact, compile, with the exception of the type of execve for execv. It also, funnily enough, doesn't loop infinitely, which makes me think that the execv call is just failing. I am Interested, but not enough to figure out Why. $ gcc -g -Wall -o suidscript suidscript.c suidscript.c: In function `main': suidscript.c:4: warning: passing arg 2 of `execve' from incompatible pointer type suidscript.c:4: too few arguments to function `execve' $ ls suidscript ls: suidscript: No such file or directory Note: I said semantically, since you pass argv[0] as the command to execute, the program will keep executing itself (if argv[0] is fully qualified). But, if you managed to get it to compile, it no doubt is segfaulting due to noted errors above. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GRUB question
On Mon, May 13, 2002 at 11:25:03PM -0400, Allan Wind wrote: On 2002-05-13 23:11:43, Scott Henson wrote: I am trying to install grub on my system so I can try out the GNU/HURD, but I cant seem to get grub to install properly on my system. I can get grub to boot and everything. I can even boot into all my OSes, but I cant get it to show me a menu. My menu.1st file is attached. Did you actually save it as menu.1st (menu dot one st) instead of menu.lst (menu dot l s t)? It took me a couple of tries to figure out that you have to store it in /boot/grub/menu.lst. Curiously enough, with the proper invocation to install, GRUB will use /boot/grub/menu.1st. I should know ... $ ls /boot/grub device.map e2fs_stage1_5 fat_stage1_5 ffs_stage1_5 jfs_stage1_5 menu.1st minix_stage1_5 reiserfs_stage1_5 stage1 stage2 vstafs_stage1_5 xfs_stage1_5 No wonder setup never worked *g*. It's a bad choice of name, as l I and 1 are easy to misread (and most other files seem to have a 1 in the name...). Wonder why they didn't just use menu? -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Dial Up Machine Configuration
On Thu, May 09, 2002 at 02:05:13PM +0100, Dave Whiteley wrote: [snip] Consider for example a time server. I would like to have it running to provide time synchronisation for all my machines, so I would like it to start at boot. However, I do not want it to dial up regularly, which would suggest that I should start it up at dial up. Consider chrony which has ppp scripts for up/down behavior. I've been using it for years now without intervention... I am (thinking about) trying to devise scripts to control things. For services that don't have up/down scripts for ppp already, you may have to do this. Not too hard... It should be possible to bring up services at boot, so that they only service the local domain. Then, on sucessful dialup, to restart them in a mode that will make use of the link. When the dialup is broken then it reverts to the original setting. Of course, this means that everything will start up at the same time, and they will all probably monopolise the phone link. So we need to prioritise the services and so allow the user who brought the link up to get reasonable service. Of course, I would like to be able to upgrade packages without having to re-edit all the scripts. I am sure that I am not the only person to want to do something like this. Packages are supposed to ask before touching modified init scripts on update... Sometimes, your scripts may get broken by significant changes though (so be prepared). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: x-terminal-emulator xterm
On Wed, May 08, 2002 at 12:41:28AM -0400, Andy Saxena wrote: On Mon, May 06, 2002 at 12:40:26AM -0700, Vineet Kumar wrote: [snip] It is on some levels; the symlink is dereferenced to get to the actual code of the executable. A program can tell how it was called by looking at argv[0], though. You should get the same result if you copy (instead of symlinking) to a different name and running the copy. With the symlink, the execution is identical, but the environment (containing the command line) is different. Could OP explain how the environment is different? argv[0] is not the environment and shouldn't effect execution unless the program takes explicit actions based on the name it was invoked with (usually a bad idea). Very interesting. Perhaps you could also tell me a way to look up the original symlink entry? $ ls -l /usr/bin/x-terminal-emulator lrwxrwxrwx1 root root 37 Oct 1 2000 /usr/bin/x-terminal-emulator - /etc/alternatives/x-terminal-emulator $ ls -l /etc/alternatives/x-terminal-emulator lrwxrwxrwx1 root root 31 Apr 14 12:24 /etc/alternatives/x-terminal-emulator - /usr/bin/gnome-terminal.wrapper Aha, a shell script! Dang, keep meaning to change that alternative to xterm... Also, how would i get a symlink to point to xterm -mutt? The man page is quite pathetic in its documentation. You can't. You probably want a small shell script or alias. But I don't recall a -mutt option for xterm ;-) (maybe xterm -e mutt?) -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: x-terminal-emulator xterm
On Wed, May 08, 2002 at 09:44:30PM -0400, Andy Saxena wrote: I am still not certain how this applies to my original question. Even if an executable can tell whether or not it is being called by a symlink, why should the xterm binary be coded to disregard the ~/.Xresources file? IMHO, it shouldn't. That's what documented argument flags are for. But, it does seem to ignore Xterm* resources if its name isn't xterm. I couldn't find where this behavior is documented in the man page... -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sorting data points
On Sun, May 05, 2002 at 08:57:51PM +0200, DSC Siltec wrote: If I think about it, this definitely isn't the quicksort. However, it may be that it is mathematically equivalent to the heapsort. In fact, it may be an easy implementation of the heap sort. Anyhow, when I was looking back at the data, the same Nlog2N factor came out, and www.nr.com at least claims that they actually get that in their worst case. In which case, they may be mathematically equivalent. The two methods are, at least, conceptually equivalent in terms of the general method. So if you wanted to use a standard algorithm, I guess I'd advocate the heapsort. But it's not quite the same as the one I described. In an approximate way, you've described a merge sort which is generally the preferred method of sorting linked lists and runs in N Log N consistently. The main difference from your description is that the list is recursively divided in half and then merged back together in sorted order. When all the recursion returns you perform the final merge and then have a sorted list. For a linked list, it doesn't require any extra space, which is nice. For arrays, quicksort is likely to do as well, and possibly better. But for 50 items, you could use a bubble sort or insertion sort and not notice a performance hit. -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Gimp - Image resolution is out of bounds, using default resolution instead.
On Thu, May 02, 2002 at 05:30:01PM -0500, Benjamin Pharr wrote: Image resolution is out of bounds, using default resolution instead. I get this message when opening photos in Gimp. These photos in particular are ones that have been scanned by Wal-Mart when they process my 35mm film. Any ideas why I'm getting this? I'd guess GIMP thinks that something doesn't add up. What is the file format? It's probably harmless. Do the images come up okay? -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dynamically generated files
On Mon, Apr 29, 2002 at 05:29:37PM +0930, Tom Cook wrote: On 0, Vineet Kumar [EMAIL PROTECTED] wrote: * Scott Henson ([EMAIL PROTECTED]) [020428 20:03]: I need something that dynamically generates files on the file system. Much like cgi. I need it to just happen when a program accesses the file. It will only be reading said file not executing it. Anyone have any ideas on how to do this on a woody system with ext3 fs? Thanks Getting more specific would help. Probably what you want is a fifo, and a process that continually writes to it. For example: mkfifo /tmp/datefifo while : do echo `date` /tmp/datefifo done Then, see what happens when you cat /tmp/datefifo in another console. (Try it a few times.) I don't quite understand what this is doing. What mechanism is used to implement the fifos? And why does this happen? # while true ; do echo `date` /tmp/datefifo ; done # tail -f /tmp/datefifo Mon Apr 29 17:27:23 CST 2002 Mon Apr 29 17:27:26 CST 2002 Mon Apr 29 17:27:26 CST 2002 Mon Apr 29 17:27:26 CST 2002 Mon Apr 29 17:27:26 CST 2002 and that is all I get? And why does the gnome-terminal with the while loop in it crash after a few more seconds? Problem with FIFO's is they need a reader and a writer connected and block until both are true. I'm not sure why the terminal crashes, but it ain't limited to gnome-terminal. I expect it has something to do with bash handling write/close errors on the FIFO which might generate an unhandled signal (SIGPIPE, for instance). BTW: FIFO's are handled in the kernel, and typically have a fixed buffer size (say, 4096 bytes). -- Eric G. Miller egm2@jps.net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]