Re: OT: Why is C so popular?
On Thu, Aug 28, 2003 at 06:46:09AM -0500, Ron Johnson wrote: > Lets take my MUA, Evolution, for example. It's not processor > intensive. Why couldn't it be written in Python? One of the main reasons is that Python leaves a loot of the resolving to runtime, that means that the code actually has to be run before you can see tha actuall typo found at compile time in languanges as C. > So many potential bugs related to pointers and null-terminated > strings would be eliminated. The SLOC count would be lowered, and Yes that kind of problem would be removed, but Python code has other parts that may as well fail. Many being simples problems of types that simply isn't resolved untill runtime and as all code isn't run all the time it may burry it self long down and hidden. I think that is bad for large projects. > For the necessary GUI stuff, use the existing wrappers python-gtk2 > and python-gnome2. wxPython would be a better choise in my mind. Python is good for a loot many project but it does however don't solve problem with bad coding. If you have potestion null pointer probelsm in C then you are coding in a bad style for that language. That will give you problems in any language. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Wed, Aug 27, 2003 at 11:55:23PM -0700, Steve Lamb wrote: > On Wed, 27 Aug 2003 23:52:35 -0600 > Jacob Anawalt <[EMAIL PROTECTED]> wrote: > > It _seems_ to work for me to convert someone elses sytle (or lack of it) > > in coding C into a format that I like (K&R). > > It boils down to this. In Python if I want something to be in a block, I > smack tab and don't even need to worry about my editor doing the right thing. > I tell it to set tab stops to 4, save all spaces and how the code is written The biggest problem working with python ident sysntax is when one comes over someones elses code and it;s intended like shit. If this happen in C, :0=G and it's possible to read, if we have the same situation in Pyhton I have to think. Also my editor doesn't support % for python, it can't autofold python and so on a loot of good things that i have in most languanges are just not there in Python. > it is a matter of block delimiter placement. Some people prefer to have the > block delimiters separate of the line that starts the block. Some prefer the There is a nice program to fix this it's called indent... > 3 different styles. But remove the damnedable braces and what are you > left with? > > if cond > block > else > block > > Hence it is not other people's style I dislike, it is the freakin' braces. So: if test: do first else: do other isn't good any more, there is still differences and different ideas... It's just not the same ones. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Thu, Aug 28, 2003 at 01:32:07PM -0500, Ron Johnson wrote: > On Thu, 2003-08-28 at 11:06, Alan Shutko wrote: > > Anders Arnholm <[EMAIL PROTECTED]> writes: > > So, basically, you don't like Python because your text editor is > > junk. Fix it or go find a real editor! > Heck, vim in default mode (no syntax coloring) is Good Enough. Thats about what I'm using, but vim i cmode is aloot more powerful tool to C programing than vim in Python mode. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Thu, Aug 28, 2003 at 02:26:11PM -0700, Steve Lamb wrote: > On Thu, 28 Aug 2003 13:20:07 -0700 > Erik Steffl <[EMAIL PROTECTED]> wrote: > Indention isn't magically lost and you're speaking of copies. The problem > in all those cases lays in the transport or in the person who doesn't know > what he is doing, not the code. 2+ years of working with Python and your > scenerio has come up exactly 0 times in my experience. So you don't copy from example web-pages and so on when trying to learn about a new feature, area and so on. I do that aloot and thats is the biggest problem with Python. The other parts of the languange however sometimes for some works make this a minor problem that I can live with. As I can live with a some of the problems in Perl, C, C++, Java, bla. and so on. No languange are perfect, not even Python, many have problems with the witespace in languanges that use them, thats is a fact. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Thu, Aug 28, 2003 at 12:55:56PM -0400, Mark Roach wrote: > On Thu, 2003-08-28 at 09:10, Anders Arnholm wrote: > > On Thu, Aug 28, 2003 at 06:46:09AM -0500, Ron Johnson wrote: > > > Lets take my MUA, Evolution, for example. It's not processor > > > intensive. Why couldn't it be written in Python? > > One of the main reasons is that Python leaves a loot of the resolving to > > runtime, that means that the code actually has to be run before you can > > see tha actuall typo found at compile time in languanges as C. > > This is such a fallacy. It's very easy to make code that compiles but > crashes. Compilers just compile code, they do not give it a magic "This > program is well written" certificate. Use pychecker, be happy. or better > yet, test your code. PyChecker, I'll have a look at that. But that still doesn't change than main think if you say test you code that applies to any other langunage too, and we ahve lint for C. In the real world testing to often gets pushed to far away in the developemnt, I know what I'm talking about I have been working almost all the time since 1998. Mostly developing test enviroment for different products. There is no such magic that makes it possible to test everything. PyUnits is a good test engine but I leaves some parts that stilla are hard to test, that could contain runtime errors. Testing also is a hard bussinis to figure out whar the heck is going to happen in runtime. The tools at the meoment for C is far more advances that the availible tools for Python. > > Yes that kind of problem would be removed, but Python code has other > > parts that may as well fail. Many being simples problems of types that > > simply isn't resolved untill runtime and as all code isn't run all the > > time it may burry it self long down and hidden. I think that is bad for > > large projects. > > how is that different from some C code casting inappropriately deep in > the bowels of your app? In python, you at least get a useful exception, > in C you just get a nasty segfault and fun with the debugger. Well the usefullness of this messange (or the more commonly seen exception in Java or C++) could be discussed. The bottom line is that in all languages good disipline and good rutines for testing is the most importand. You will always have a loot of trobles, you will always leave someting that you didn't expect to happen. All probrams in all languanges have the problems, changing langunage will not with any magic make this go away. So making a office/mail suit using Python insteda of C wount make it any better or less likely to crash for the user becaues the problems are not in the language it in something else. There bu not saying that Python might not be a good languange for the work or that C might not be a langes for this kind of work. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Thu, Aug 28, 2003 at 12:46:43PM -0700, Steve Lamb wrote: > On Thu, 28 Aug 2003 15:10:33 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > One of the main reasons is that Python leaves a loot of the resolving to > > runtime, that means that the code actually has to be run before you can > > see tha actuall typo found at compile time in languanges as C. > > This is offset by the fact that the code-compile-run cycle is reduced to > code-run. Furthermore a lot of prototyping of code can be done interactively. > When I run into a situation where I'm not sure exactly how I want to do > something I open another window, fire up Python, import the appropriate > libraries and twiddle with the code until I get a form that works. Then I > take that and put it back into the main code. The difference in work cycle isn't that big, in Python I usally do "./verifyProgram" and in c mostly "make test". If I need to test something in C usally just fire up a new file and type in the needed code and run, or use my nice c alias that takes c code from stdin and runs the resulting binary. (And no I have never liked typing in code interactivly, I store everything in a file. But thats is an area of personaly liking.) / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 02:36:26AM -0700, Steve Lamb wrote: > On Fri, 29 Aug 2003 10:41:45 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > So you don't copy from example web-pages and so on when trying to learn > > about a new feature, area and so on. I do that aloot and thats is the > > biggest problem with Python. > > I do and it hasn't caused me any problems. Might I suggest you learn how Then tell me how you do it, if it's that easy that I shuld have know aboutis, but I still use vim as vi almost only using the vi compatible functions with some small additions that happens to make it more usefull. There might be a simple concept not documented of described in a obvius way that I have missed, but for me it gets into a painfull reindening line by line (or sometimes marking the block and using the shift operators to get it right, sometimes it's hard to know how many steps you should shift the code so I still think it's a loot harder than just =%, and using = is realy bad on python code, then you have changed the syntax of the file). > to use vim before citing problems based on your own ignorance[1]. Furthermore > until you figure out how not to send mail both to me and the list when I am > clearly reading and replying here correspondence between you and I is at an > end. I got a message in my inbox, and as it looked more personal I replyed to it. This message is only going to the list, > [1] Prime example being vim not able to fold Python code when I've seen it > done. I don't yet (or at least inte version of vim that I are useing had the autofold function for Python code). Checking the currect documentation show that now I can do it. At least on this computer. SO I stand corrected that has been added and is working. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 08:35:25AM -0500, Ron Johnson wrote: > On Fri, 2003-08-29 at 03:35, Anders Arnholm wrote: > > On Thu, Aug 28, 2003 at 01:32:07PM -0500, Ron Johnson wrote: > > > On Thu, 2003-08-28 at 11:06, Alan Shutko wrote: > > Thats about what I'm using, but vim i cmode is aloot more powerful tool > > to C programing than vim in Python mode. > Yeah, for things like braces-matching. But the need for that is > eliminated in Python... So if not using braces matching, how does one quickly jump to the end/begining of a block. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 08:33:40AM -0500, Ron Johnson wrote: > On Thu, 2003-08-28 at 14:07, Pigeon wrote: > set tabstop=4 So thats why all code form other Python programers look like shit and dont line up, not that a tab is 8 spaces wide and will stay that way for ever everything else is PLAIN fucking evil!!! Or as the vim help file puts it. Note: Setting 'tabstop' to any other value than 8 can make your file appear wrong in many places (e.g., when printing it). / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 02:06:53PM -0500, Ron Johnson wrote: > On Fri, 2003-08-29 at 11:27, Anders Arnholm wrote: > > On Fri, Aug 29, 2003 at 08:35:25AM -0500, Ron Johnson wrote: > > > On Fri, 2003-08-29 at 03:35, Anders Arnholm wrote: > > So if not using braces matching, how does one quickly jump to the > > end/begining of a block. > Well, "PageDown" and the down-arrow usually work for me... It doesn't work for me, it only jumps to the end of the page and or a line down. When programing in C I'm used to when being at the beginin of a class/function/block being able to get to the end of this block by pressing % (e.g. matching brace) and continue from there. That makes navigation in the file fast and I need to keep track of a loot less bookmarks. I what a such functionality when editing Python code too. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 01:46:37PM -0500, Ron Johnson wrote: > On Fri, 2003-08-29 at 11:32, Anders Arnholm wrote: > > On Fri, Aug 29, 2003 at 08:33:40AM -0500, Ron Johnson wrote: > > > On Thu, 2003-08-28 at 14:07, Pigeon wrote: > > > set tabstop=4 > > > > So thats why all code form other Python programers look like shit and > > dont line up, not that a tab is 8 spaces wide and will stay that way for > > ever everything else is PLAIN fucking evil!!! Or as the vim help file > > puts it. > > > > Note: Setting 'tabstop' to any other value than 8 can make your file > > appear wrong in many places (e.g., when printing it). > You conveniently ignored the fact that I also mentioned > set expandtab > There are no ^I characters when that is used. So why change tabstop? Is there any reason for that what so ever and I know that the file you send out didn't have any tabs, but the only reason what so ever for setting tabstop to 4 I could fingure out what that other programers working with the same files as you, did do this wrong! And as I interpited this as a reconedation, something in the enviroment whas wrong. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Fri, Aug 29, 2003 at 12:30:24PM -0700, Steve Lamb wrote: > On Fri, 29 Aug 2003 18:32:08 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > On Fri, Aug 29, 2003 at 08:33:40AM -0500, Ron Johnson wrote: > > > On Thu, 2003-08-28 at 14:07, Pigeon wrote: > > > set tabstop=4 > > So thats why all code form other Python programers look like shit and > > dont line up, not that a tab is 8 spaces wide and will stay that way for > Nice try. You trimmed "expandtab" from the quote. Yes but why does you need it if you don't get bad files from other programers using tabs for spaces wides as indention? I saw and knew what expandtabs does, I don't have it my self most of the time, but in some projects it's nice however in most enviroment all progamers know NOT to change tabstop and then tabstop is always 8 wide in all files. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 01:49:23AM -0700, Steve Lamb wrote: > On Mon, 1 Sep 2003 10:23:38 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > So why change tabstop? > So that when we hit tab it goes to the next multiple of... 4? Then why not learn the editor :^) Whan hitting tab MY vim with tabstop of eight jumps to the next indention level, e.g. what I have in the shiftwidth variable that you also changed to four. If you have indent on that is what controls the behaviure of the editor when pressing tab. I can see no reason setting sw if not usint cindent an then the value of sw is used when you press tab. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 01:56:39AM -0700, Steve Lamb wrote: > On Mon, 1 Sep 2003 10:26:56 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > Yes but why does you need it if you don't get bad files from other > > programers using tabs for spaces wides as indention? > > Because new people to Python haven't yet learned about no tabs? And most of the use 4 spaces wide tabs? Or just use one tab as indention level? In every case that still explans why python code found on the net other looks realy bad indented. Don't take it personal, it just explans for me why it look so bad. > > I saw and knew what expandtabs does, I don't have it my self most of the > > time, but in some projects it's nice however in most enviroment all > > progamers know NOT to change tabstop and then tabstop is always 8 wide in > > all files. > Since we're talking Python here and since vim does autodetect quite nicely > between Python and C (and Perl and.. and... and...) and one can set the > tabstop, expandtabs and the like based on the file edited. What exactly is > your point? If you feel that it's necessary to changes tabstop to four to get the code readable, then only reason to have tabstop set to four, then that explans why code found on the ner look bad. It's also stated in the Python language definition that tabstop is at 8 spaces, and changing this might actually changes the syntax of a program, given that the actually is a mix of spaces and tab in the source file. (And there is a recomendation not to have tabs for crossplatform compability.) / Anders -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 02:51:20AM -0700, Steve Lamb wrote: > On Mon, 1 Sep 2003 11:02:09 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > Then why not learn the editor :^) Whan hitting tab MY vim with tabstop > > of eight jumps to the next indention level, > > I suggest you try that again VERRRY carefully. I just tried it. Entered > the editor in Python mode > :set shiftwidth=4 > if foo: Just tried, :set sw=4 class test: Four spaces no tab, ofcource I use cindent as default... > Since we're not in cindent that doesn't count and since it is easier to And why don't you use cindent? Becouse you have something religus against the c in the name? / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 03:04:28AM -0700, Steve Lamb wrote: > On Mon, 1 Sep 2003 11:18:14 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > Yes, but vim uses tabstop to determine how many spaces to put in. Hence > tabstop to 4, expandtabs on, shiftwidth to 4. Tabstops to know what to do > when we hit tab, expandtabs on so we don't send it out to other people, > shiftwidth to know what to do when we want to reindent. SO the reason is to inport bad fomrated code, and make that code better formated. For me thats dosn't make med have to change my editor. As this still needs a manual step, whan it happens I can change my tabstop to fix it in that buffer only. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 03:50:08AM -0700, Steve Lamb wrote: > On Mon, 1 Sep 2003 12:15:11 +0200 > Anders Arnholm <[EMAIL PROTECTED]> wrote: > > And why don't you use cindent? Becouse you have something religus > > against the c in the name? > > Because we're not programming in C and I'd rather not take the chance of > it doing something stupid based on the presumption we are coding in C. Then you should look into what itäs doing and learn your editor, even tough it's called cindent i more of context indent the c language indent. Some parts of it works with python but autoindening the whole file doesn't of understandable reasons. Peronaly I have gorwn so used to cindent (or the simularity in emacs) that I can't program without. Well actually I didn't do any serius programing in vi untill vim came around with just that feature. Since then I haven't used Emacs more on some minor ocations par programing with other programmers. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Mon, Sep 01, 2003 at 11:24:41AM -0400, Mark Roach wrote: > On Mon, 2003-09-01 at 06:20, Anders Arnholm wrote: > > On Mon, Sep 01, 2003 at 03:04:28AM -0700, Steve Lamb wrote: > > > On Mon, 1 Sep 2003 11:18:14 +0200 > > > Anders Arnholm <[EMAIL PROTECTED]> wrote: > Here is a very easy way to achieve that which does exactly what I want > every time: :set ts=4 et > > It works. period. There is no "better way". There is only "accomplishes > that goal", or "doesn't accomplish that goal". This has nothing to do This one does one big problem IF you get a file from someone else that has mixed tabs and spaces. And is using a tabsize of the standard 8 spaces. Then in combination with Python you by editing the code you actually does changes the sysntax of the file. Then I don't see the huge difference to change into :set sts=4 et if thats what you actually whan't. > with other people's code. This is what I do for *my* code. Please > understand that just because the word "tab" is used in the options that > doesn't mean this actually has to do with the \t character The difference between ts and sts is just that ts works on \t charachters imported in the files to, and sts doesn't. Setting ts without et is definitly wrong. Setting it with et just look wrong, but doesn't hurt so many else. (Or hopefully don't...) / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Tue, Sep 02, 2003 at 09:37:27AM +0100, Colin Watson wrote: > On Mon, Sep 01, 2003 at 10:19:18PM -0400, Mark Roach wrote: > > On Mon, 2003-09-01 at 13:26, Colin Watson wrote: > Nowadays, on average I tend to use expandtab for new code, but > converting tabs to spaces is still an operation that needs to be handled > carefully with respect to revision control, so I don't apply it across > the board. Almost all decent revision control programs tend to let you ignore whitespace when doing diffs of a program. (Works fine in most languages as whitespace don't containing any information.) So for that reason it should still work to update the code to have better indention and get the full usage of the revision control. (As you don't use python.) / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: OT: Why is C so popular?
On Tue, Sep 02, 2003 at 11:12:46AM +0100, Colin Watson wrote: > On Tue, Sep 02, 2003 at 11:43:19AM +0200, Anders Arnholm wrote: > > On Tue, Sep 02, 2003 at 09:37:27AM +0100, Colin Watson wrote: > > > Nowadays, on average I tend to use expandtab for new code, but > > > converting tabs to spaces is still an operation that needs to be handled > > > carefully with respect to revision control, so I don't apply it across > > > the board. > > > > Almost all decent revision control programs tend to let you ignore > > whitespace when doing diffs of a program. > > Other people working on the project still need to use that option, and > it still makes annotate less convenient. Unnecessary cosmetic changes > are always a pain, and if done at all they should always happen in > separate commits to functional changes. (Wearing my code reviewer hat, I > will jump up and down on your head if you mix cosmetic changes with > functional changes. :)) Yes, it's good idea to keep it separate check ins. But that doesn't changed the main thing i they are needed, e.g. the code is in bad indention according to coding standard or something else like this. Make the change, run the test suit, checking the changes (with comment "update to follow coding standard." or something like that) and then proceed with the other functional standards. Coding standard, tab/spaces and some on is usually something you like to do before you start a complex work. So that the the complex work gets a little easier. If you like to compare versions operated by a syntax changes that the ignore whitespace comes in handy. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp0.pgp Description: PGP signature
Re: Presort my email messages when exmh is the MUA.
>>>shaul wrote: > I am using exmh to handle my email (MUA ?). The MTA is smail (is it also the > MDA ? What are the differences between MDA and MTA ?). > I want to presort my email messages. > Can you recommend a way to do it (procmail ? exmh capabilities for this task, > if it got any ? other ?). I use, Exmh, sendmail, and procmail and I think that is a great combination, someday (soon?) I will probably change sendmail to Postfix. But for now I have been using sendmail, I have also a short while used smail, but it's hard to configure the way I like it to work. I have read everything some ware in the exmh or the nmh documentation. My entries in the procmailrs files is something like this: :0 w: debian/announce/$LOCKEXT * [EMAIL PROTECTED] | /usr/lib/mh/rcvstore +debian/announce -create :0 w: inbox/$LOCKEXT | /usr/lib/mh/rcvstore +inbox -create Anv most important I have in the beginning of my procmailrc file: SHELL=/bin/sh PATH=/usr/local/lib:/usr/local/bin:/usr/bin:/bin MAILDIR=$HOME/Mail/ LOGFILE=$HOME/Mail/procmail-log LOCKEXT=.lock # This is to make sure I only get mail one. :0 Wh : msgid.lock | formail -D 8192 .msgid.cache # And rewrite to fit in mh folders. :0 Whf | formail -z -R 'From ' X-Envelope-From: / Anders
Re: 2.2 ready ?
>>>Shaleh wrote: > On 08-Jan-99 Bert Barbe wrote: > > Is there a distribution that is kernel v. (pre-)2.2 ready (= having > > the right versions of neccessary tools to run linux (pre-)2.2) > potato definately, slink possibly Slink works, over here (I think I have mostly slink installed now, SANE is the only packet I know is potato, and a loot of other package are still hamm. Apt-tells me that it is holding back a bit over 100 packets.) / Anders
Re: Secure Mailer
On Tue, 12 Jan 1999, Nathan E Norman wrote: > A quick glance at the web site doesn't indicate that Secure Mailer > (Wietse Venema's software) has been packaged yet. Anyone know if it has There are a unstable package for postfix-19981230-1, I have briefly watchd a test installation of the MTA, and the installer scripts are broken. I have been activly using (my own compiled postfix) for some days now at my site and I think it works very well. My mail server is a 386sx-16, 4Mb memory and the enormus disk of 200Mb (just upgraded from my old 50Mb disk, and Debian 1.3), the installation on my domain is mainly for tesing and eveluation the software. I have found things that I like to look closer at. 1) Postfix reports EXPN balp 502 Error: command not implemented I don't know if this is configuration or time, but I like to EXPN mail adresses. 2) Is it possible to get postfix to do ident lookup on incomming smtp sessions. And then there has to be some serious perforamce testing :) / Anders -- o_ Anders Arnholm, Work : http://www.opensoftware.se/ o/ /\* MOVING * Phone : +46-31-*SOON* /|_, \\* MOVING * Celluar: +46-703-160969 / Sweden http://www.acc.umu.se/~balp `
Re: Partition Type A0 ??
>>>Will Lowe wrote: > partition, and that's how it did that crazy install-with-no-disk thing. > Does anybody know if that's possible? Or know what partition type A0 > really is? I don't know about how AST makes it, but I have seen several Laptop's from Toshiba that uses an extra partition in the end that is used for power saving only. I don't have access to that machine anymore so I can't check the partition type. But I'm sure that you can place the Win95 cab-files, on the "primary" fat partionan and make it bootable from there. (As it was on my Dell mashine.) / Anders
Re: remote execution
>>>Robert Vollmert wrote: > Currently, I'm using ssh with a /root/.shosts file on B, but this > seems a little dangerous. Any better suggestions? ssh using RSA authentication to a "dummy" user on the remote host, use sudo to run that script as root. This will only make that command possible from outside. / Anders
Re: [DEBIAN] Dell Latitude CPi A366XT
>>>[EMAIL PROTECTED] wrote: Sorry for repying on the answer, missed (probably deleted) the original question. > > is there anybody who has experience with the > > Dell Latitude CPi A366XT portable. > > Is the Neomagic 2200 graphics board supported? The Label in front of me, well actually Fn-Setup, tells me that this machine is a Dell Latitude CPi A366XT, X works very nice with XFree 3.3.3.1 from potato, NeoMagic chips are supported straight of (as XF86_SVGA). I have two problems left with the machine, the sound (NeoMagic 256 Audio Chip, NeoMagic says that it's up to Redhat, and Redhat only to write the drivers.) And Suspend Resume when the pcmcia drivers for my 3com card are loaded (I usually work around this by issuing ifconfig eth0 down ; /etc/init.d/pcmcia stop before closing the lid.) / Balp
Re: gui progs as root in x under normal user?
>>>Robert Vollmert wrote: > A safer solution would be to do (as root): > > # export DISPLAY=:0.0 > # export XAUTHORITY=/home/yourname/.Xauthority That worked until I started to work in a networked environment. Then I don't like all clients to have root access to the users home directory so the row sound be something like this. /usr/bin/X11/xauth -f /home/balp/.Xauthority extract - $DISPLAY /usr/bin/X11/xauth merge - / Balp
Re: Help with a Dell Latitude CPia
On Wed, Nov 24, 1999 at 12:03:57AM -0600, Chris Larson wrote: > I have installed corel linux onto my laptop. Everything looks great and runs > fine except for three things. > > 1. My ethernet card is not working. I can ping myself but I cannont ping > anything else. I have a Dell Latitude CPi A366XT, that I use to write this message from I use pcmcia pacage 3.1.3 that i have compiled my self, it identifies my network card as: Socket 0: product info: "3Com Corporation", "3CCFE575BT", "LAN Cardbus Card", "001" manfid: 0x0101, 0x5157 function: 6 (network) > 2. I have no sound. sndconfig cannot detect a sound card. I think these > laptops use some Neomagic sound chip. Thats correct you willl need Linux kernel 2.2.13 or newer :^) To get support for that sound card. (At least some basic support that lets med listen to mp3's while writing this...) > 3. My Gold card 56K modem will dial out but will not connect to my ISP. I > get the error that the pppd died unexpectedly. Here I can provide no help at all. -- Anders Arnholm.-. Sisjö Kullegata 8, 421 32 V. Frölunda| open software | Tel. 031-28 21 84 | Fax 031-28 51 66 `-´ Mobil 0703-16 09 69 | [EMAIL PROTECTED] | www.opensoftware.se
Re: Experience with IBM Thinkpad 300CSE
> Does anyone have experience installing Debian (Hamm/Slink) on an IBM > Thinkpad 300CSE? My father got me an early xmas present. 486-50(?), 20Mb > RAM, 350Mb HD. It has Win95 on it now. I don't really need X, just curious > as to how well Linux will like such a beast. I have installed it on my brothers 360C and it worked. I almost configured X, XFREE86Config did run but got something "abd" in the modlines that I din't had timer to check out. And I don't know if he everfixed that. The 3690 is some kind of 486, with color display. There are several good sites about IBM thinkpads linked from www.linux.org. / Anders
Re: Installation problem from floppies, RAMDISK error, please help me!!!
>>>[EMAIL PROTECTED] wrote: > I am trying to install debian from a 386SX 16mhz with 5 megs of ram from a > floppy disk drive. It has a 106 meg hard drive. 15 megs of the hard drive > are used as an MSDOS partition. The installation goes through detecting ite ms > on the system and then says the following and freezes: > > RAMDISK: Compressed image found at block 0 I have installed Debian 1.3 on a simular maskine Just that i "only" had 50 mB harddrive and 4 Mb memory, I guess that you have to use the lowmemory installation procedure (as described in the install.txt file).
Re: Kernel for AMD Thunderbird
On Thu, Aug 23, 2001 at 07:59:01PM +0300, George Karaolides wrote: > On Thu, 23 Aug 2001, Jason Majors wrote: > I read that, but I haven't gone over to 2.4 yet. I'll cross that bridge > when I get to it... I'm running two machines on 2.4.[59] with Athlon kernel no problems. One is a Duron 600 the other a Thunderbird 1200. The duron as run kenels 2.4.1 throu 2.4.5 and the thunderbird only as 2.4.9. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgp2MTnACK5uA.pgp Description: PGP signature
Re: Making /home not accessible by outsiders
On Thu, May 30, 2002 at 05:38:29AM -0400, ThanhVu Nguyen wrote: > default). Now I changed my mind, because even /root is readable by > others ... I just wonder what kind of infromatiosn does you have in /root that is secret and has any value in hiding. I only have som standard .profile, .ssh/known_hosts, but they raily gets touched. (All root work is done with su.) The secret stuff is in /home/balp/private/. / Balp -- o_ Anders Arnholm, HiQ - Consultant o/ /\[EMAIL PROTECTED] Phone : +46-703-160969 /|_, \\http://anders.arnholm.nu/ http://www.hiq.se / ` pgpd84LglrhmP.pgp Description: PGP signature
