Re: mac X+ perl where is the web folder
I am working on MAC OS X and Perl + Apache2.0. I used to use Asp + IIS. Would you please help me. 1)Where is web folder of apache2.0 under os X? Apache 2.0? Has Jaguar moved that far ahead? It seems related with /usr/local/apache2.0 /usr/local? Is Apple using that now? I'm thinking I'm remembering apache (1.?) in one of those Library directories, so you could look at the executable with the GUI, even though you can't look at the configuration with the GUI. ,but I cannot copy a file into it or construct a folder using mkdir. Well, that's probably because you are smart and not loggin in as root. Now, all you need to do is enable root login just long enough to edit the sudoers file (Wasn't that in /etc somewhere?) and that kind of thing. 2)Should I copy .pl file into /usr/local/apache2.0/ cgi-bin in order to use cgi in perl? I have the idea that the default httpd.conf actually allows you to have cgi-bins for each user. Am I off base on that? 3) Can you give me some documents or hypertext link about web-publishing using apache and perl ? It's been a while since I looked that sort of stuff up. Try google: http://www.google.com Maybe plug in something like web-publishing apache perl example. -- Joel Rees [EMAIL PROTECTED]
Re: Opening file with application
John Labovitz [EMAIL PROTECTED] contributed On 9/8/02 3:58 PM, John Delacour [EMAIL PROTECTED] wrote: So at last Application names are cached somewhere with their paths. And about time too. Where? Maybe here: /Library/Caches/com.apple.LaunchServices.LocalCache.csstore but it's binary, and I don't know how it's formatted. -- John Labovitz [EMAIL PROTECTED] www.johnlabovitz.com Obvious thought, but have you tried plist? -- Joel Rees [EMAIL PROTECTED]
Re: Opening file with application
John Labovitz [EMAIL PROTECTED] responded: On 9/8/02 8:01 PM, Joel Rees [EMAIL PROTECTED] wrote: Obvious thought, but have you tried plist? It doesn't seem to be that -- I tried both pl and plutil, and neither wanted to read it. But I have no experience with plists, so maybe I'm not running these correctly... Hmm -- http://www.macaddict.com/osx/hacks/pledit.html talks about PropertyListEditor or something like that. (picked up by searching with google for plist.) They don't let me use my iBook at work, so I can't check what it was I used last, but I keep trying to convince myself that some editor allowed me to load and save compressed plist files. I'm probably confused, though, seems like I spend most of my time confused. Anyway, the file info app (cmd-i) allows you to set the default application for a single file or a file's type, as I recall, which is probably more to the point of the original question, now that I think about it. Plugging that path you mentioned: /Library/Caches/com.apple.LaunchServices.LocalCache.csstore into google produced three moderately interesting results, too. (Hope you don't mind if I bounce this back to the list. -- Joel Rees [EMAIL PROTECTED]
Re: Opening file with application
Methinks we shouldn't be going down this path, really. Apple has a defined way to get to this stuff: http://developer.apple.com/technotes/tn/tn2017.html ...which is probably exactly what the 'open' command uses to figure out which application to launch. Oh, darn. No excuse to write arcane code today. :) Methinks youthinks right. Thanks for the link. -- Joel Rees [EMAIL PROTECTED]
[OT] Re: OS Poll
Sure, the thread is somewhat OT, but we all need to think about what tools we'll use in the future, when the glitz has worn off information technology. I thought one of the real core differences was that it was built with GCC 3.1 instead of the 2.95 branch. From what I've been reading on openBSD's misc list (among other places) GCC 3.xx is not handling low-level optimizations very well. GCC 3.1 might, in fact, take part of the blame for the kernel panics. Trade-offs. You sometimes have to take a step back to move ahead. As a developer I'm quite happy to have paid for the new updated tools to be so deeply integrated. I kinda wish they'd gotten perl 5.8 under the wire, but that wasn't a big deal to install. There's alot under the hood that really makes it worth the $$$. IMNSHO Knowing what has had to be done to get Mac OS X running, I am simply in awe. Okay, not like what I feel towards God or the Grand Canyon, but as much awe as I can feel towards any of the work of humans. In my POV, 10.2 confirms that Apple recognizes that the corners they cut to get Mac OS X out the door (just barely) in the market window have to be filled in. Anyway, I'm buying Jaguar this week or next month, as my budget will handle it. .mac doesn't fit my budget. I've been half expecting .mac to go for-pay, so I never used the .mac addresses for more than spam buffering. (Accessed the google newsgroups from the .mac address, and I was spending too much time on that anyway.) What would induce me to keep the .mac account? If my budget were in better condition, I'd probably keep it, although I don't really use anything but the e-mail. (Could they separate the e-mail?) I already have a provider, and their rates are kept reasonable. The provider serves my regular personal mail address, so I need something to motivate a switch. If .mac were JPY 10,000 a year, including a 56K modem point of access (15+ hours, in Japan), I'd consider switching. If the 10 yen for three minutes telephone connect time were also included, or if it were JPY 20,000 for ADSL, I'd seriously consider it. I do think they ought to have a free minimal .mac account, e-mail and a small iDisk, one per valid serial number, with all new machines, valid for the duration of the warranty/service agreement. That would probably sell more AppleCare, too. But what I would really like, and if Apple were to include it with their basic .mac, it would be a huge selling point, is a spam-trapping news-group-view mail address server/mail browser -- an address for public use with a browser that would do things like showing the send/reply sequence in tree form the way it shows on newsgroups, would sort mail by sender and subject line, would download an index list of topic lines and senders on demand (so I could delete without having to wait for the whole body to download), and where the server would send an automatic response to any mail that did not match the accept filter. Anybody know of work on mail server scripts that would support this kind of thing? Of course, where this all has to eventually head is that your phone will become your mail server. -- Joel Rees [EMAIL PROTECTED]
Re: question on ssh and peeve on editors
Puneet opined the truth is that it is _primarily_ a wysiwig html editor... but if you were to think of it as a web application development IDE... then it would make sense to offer some basic scripting support (which it does, except not for perl), or a darn good integration with a (or several) external editors. I don't know much about templates, and I like GoLive, in concept, especially now that 300 MHz processors are considered slow. (First experience with GoLive was the pre-Adobe days on an, erm, LC 630. Speaking of which, anyone know a place in the Hanshin (Osaka -- Kobe) area to get a non-buggy 68040 cheap? I still have one of those boxes, and I want to dual-boot openBSD on it.) But I found that what I usually did was design the page, grab the design elements, and paste them into the real page. That is, I left GoLive behind and used a straight text editor once the design work was done. (Used CodeWarrior's editor since it was pretty stable with shift-JIS and handled the various flavors of line-endings for me. I like BBEdit, too.) General-purpose WYSIWYG editors for XML are still a little bit ahead of us on the software technology curve. Say, my company is training a large group of us on Struts (Java -- Jakarta). Anyone know of a comparable project with Perl, i. e., a framework that would allow separating the business model and control logic out of the view? -- Joel Rees [EMAIL PROTECTED]
OT Java pseudo-benchmark (was Re: What up with the mac)
Okay, here's the Java program I was talking about, since someone might
want it and I'm going to be off-list for a while:
-begin code
/**
* Let's try the Factorial in BigInteger
*
* @author Joel Rees, Altech Corporation, Esaka, Japan
* Copyright September 2002
* May be copied, modified, and/or used freely.
* No warranty. Use at your own risk.
*
* @version 0.1
*/
import java.lang.Class;
import java.math.BigInteger;
public class BigFactorial
{
public static void main( String[] args )
{ if ( ( args.length 1 ) || ( args[ 0 ].charAt( 0 ) == '-' ) )
{ System.out.println( Usage:
/* Okay, this is ridiculous. */
+ BigFactorial.class.getName()
+ integer {, integer } );
}
else
{ for ( int i = 0; i args.length; ++i )
{ BigInteger input = new BigInteger( args[ i ] );
System.out.println(
(
+ input.toString()
+ )! == + factorial( input ).toString() );
}
}
}
/* Let's not try to blow the stack with the old
* forced example of recursion, at any rate.
*/
public static BigInteger factorial( BigInteger n )
{ if ( n.compareTo( BigInteger.ZERO ) 0 )
{ return new BigInteger( 0 );
}
BigInteger result = new BigInteger( 1 );
while ( n.compareTo( BigInteger.ONE ) 0 )
{ result = result.multiply( n );
n = n.subtract( BigInteger.ONE );
}
return result;
}
}
--end code-
Should be easy to re-write in Perl.
--
Joel Rees [EMAIL PROTECTED]
Re: CPAN Jaguar
mrsparkle ellem ~ $ perl -v This is perl, v5.6.0 built for darwin I have loaded FINK CPAN is acting funny. Actually it isn't doing anything. I get a lot of this: You have run CPAN before, right? cpan install Mail::Sendmail Please, install Net::FTP as soon as possible. CPAN.pm installs it for you if you just type install Bundle::libnet Please, install Net::FTP as soon as possible. CPAN.pm installs it for you if you just type install Bundle::libnet Please, install Net::FTP as soon as possible. CPAN.pm installs it for you if you just type install Bundle::libnet +++ cpan install Bundle::libnet Can't install Bundle::libnet, don't have an associated bundle file. :-( at /System/Library/Perl/CPAN.pm line 1806 any ideas? Looks to me like FINK walked on wherever you put your configuration info for CPAN. You'll probably need to re-configure CPAN, at minimum. Anyone successfully using CPAN and FINK together? -- Joel Rees [EMAIL PROTECTED]
Re: [MacPerl] Re: problem with Japanese text
Not sure if my comments are relevant, just feeling inclined to expose my ignorance -- Character set difficulties are still a real problem, but so is dynamic text. Damian Conway's paper An Algorithmic Approach to English Pluralization http://www.csse.monash.edu.au/~damian/papers/HTML/Plurals.html contains some fairly complicated tools for generating dynamically- pluralized English. Now generalize that tool set for multiple languages and/or more complex variations. Right. Japanese is one of those languages that has relatively few specifically plural forms. To get the pluralizations right in Japanese, the program would have to consult a dictionary. In my current work, I am generating user-specific explanations for the permission and ownership information in (roughly) an ls -al listing. That is, the user gets three paragraphs, saying (a) what the effect of these permissions is on the user, (b) how this was derived, and (c) what the item's permissions are, as a whole. I see the reason for the interest in automatic pluralization there. Pluralization could probably be ignored for this purpose for Japanese, but, if the purpose is to produce text that the technically un-inclined can parse reasonably effortlessly, there are all sorts of other context related issues, most of which would require not just vocabulary dictionaries, but idiom dictionaries as well. And your locale machinery would have to include some sensitivity to dialect issues and social status issues, to make the generated text natural and non-offending. Japanese is becoming more egalitarian, more homogenized, and less colorful, so those who work on such things are aiming at a moving target. Thinking about the recognizer side, did anyone mention that Japanese text does not use word delimiters? Space has a somewhat different meaning for Japanese. -- Joel Rees [EMAIL PROTECTED]
[OT] Re: [MacPerl] Re: problem with Japanese text
... In English the singular nominative pronoun is nothing but I, no matter how old or young you are or whether you are a boy or a girl (or a computer). But in Japanese it can be Watashi or Boku or Ore Boku. Hmm. What would it mean if an educational program intended for kindergarten use said Boku no shippai desu ka? ... Well, the total richness of Japanese is I believe is increasing but ironically per-capita richness might not be. But I believe this phenomenon is not unique to Japanese; could be even more prevalent in English. If you don't believe me just compare the Two Bushes in White House :) Heh. West Texas is a country and a language to itself. -- Joel Rees [EMAIL PROTECTED] (Amusing myself by imagining the elder bush saying Kwansai and the younger saying Kansai. Not sure I could imagine the old guy saying tefu-tefu, though.)
Re: OT: looking for the relevant list
Suppose one has a question about compiling some (perl-unrelated) c code (originally developed on linux) on os x, is there a list where it would be appropriate to pose this kind of question? If you can subscribe to newsgroups, or if you can bear the time delay on groups.google.com, you might try comp.lang.c or comp.lang.c.moderated Google requires you to register, but it's free. The big drag is that in the three or more hours it takes to see your question posted, you usually either solve the problem yourself or lose interest. But it's a high-volume list, so, if you subscribe, you'll want a newsreader, or you'll want to set filters for you mail reader. -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese
I ask only because it came up here before and I can find nothing online about it... I'm trying to do some things in japanese on my OSX box, unfortuanately my japanese isn't terribly good so any help info on my computer is minimally helpful. My biggest question is how to get garbled text like deg.TMGDBdeg.$D9$BBdeg.$D9deg.$E9deg.$D9-$DA$F5 back into the form of kanji/kana. My next question is where there's a good online FAQ site for doing japanese on OSX or finding OSX programs that accept Japenese (unicode?). I couldn't even guess where to tell you to start, except for google. (Google is a great tool.) The text there looks kind of like it might be some ASCII visible encoding of euc-JIS, but I am not going to take time to test the hypothesis. Where did you get it? If it's from a web browser, you can try selecting a different encoding. (OmniWeb 4.1 requires you change the default in preferences and then re-load the page.) I'm thinking there are good tools to work with that kind of text in Perl 5.8 (also the jcode module in previous versions), but there are people around here better qualified than I to tell you how. There were some good pages mentioned here several months ago, so you might search the archives. Or search the web for things like CJK, euc encoding, shift-JIS, tron characters, etc. -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese
I type in Japanese in textedit, mail and project builder without difficulty. I find I've had to fuss a bit with Project Builder's editor. Most of the docs, source, etc., I see are encoded in shift-JIS, but I haven't yet found a way to tell the PB editor what encoding to assume when opening a file. I have been able to figure out that it seems to pick up the byte order mark or something for UTF-16 (which surprised me). And it seems to require UTF-8 for compiling in C. (Maybe I should download the new version of the dev tools. I don't have the room on my vintage iBook's 5.6G drive, but I think Apple said the bulk of the documentation didn't have to be on the boot volume.) But I can tell Text Edit what encoding to assume and then save as UTF-8 or UTF-16, so there's a work-around. Unfortunately, that's not going to help the OP, near as I can tell. Do you recognize his deg.TMGDBdeg.$D9$BBdeg.$D9deg.$E9deg.$D9-$DA$F5? Have you set the language and the script in the international System preference? I like to have several users, each with different language and script settings. (Family account set to Japanese, of course.) I also find it convenient to have multiple scripts selectable, and the precedence defaults are pretty handy, as well. -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese
Editors I use a lot. Jedit, Java editor. I've got to try that some time. www.jedit.org It is extremely good at setting default encodings, changing file encoding (batch mode, too) on the fly, et cetera. Mi, great text editor from Japanese author http://www.asahi-net.or.jp/~gf6d-kmym/en/mimi/index.html Mac OS X. Looks interesting. http://www.asahi-net.or.jp/~gf6d-kmym/mimi/index.html VIM...well, not great at Japanese. But an lovely editor. Just had to add it here. Works great in X11 on OS X, too! ;) Use it in freeBSD, trying to get it set up for openBSD, having trouble with Wnn and onew, because the groff (ja-groff) port is not where the obsd port seems to think it should be. While I'm definitely glad to have it for fBSD and oBSD, I have trouble motivating myself to use it on Mac OS X. I'm a little spoiled, perhaps. (Hmm. Might be interesting to try MIFES under emulation on obsd/fbsd. It isn't free, of course. Come to think of it, I should get Java up under emulation and try Jedit first.) I haven't tried the editor in the latest Metrowerks Codewarrior, but it's always been natural for me. No character set tools, however. The color-coding would get out of sync in shift-JIS strings (for reasons that those who work regularly with shift-JIS know and appreciate). BBEdit is what spoiled me on the Mac, by the way. But none of that addresses the OP's garbled string of something I was thinking looked like euc-JIS transmorgrified into some visible hexadecimal display form -- deg.TMGDBdeg.$D9$BBdeg.$D9deg.$E9deg.$D9-$DA$F5 Where have I seen that before? It just doesn't make any sense at all as any JIS in a visible hexadecimal form. Maybe it's just raw, untouched, straight JIS, with no escapes. -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese
deg.TMGDBdeg.$D9$BBdeg.$D9deg.$E9deg.$D9-$DA$F5 Where have I seen that before? It just doesn't make any sense at all as any JIS in a visible hexadecimal form. Maybe it's just raw, untouched, straight JIS, with no escapes. Nope. Not even straight JIS with the escapes being munged to periods. (Tried opening it with an editor that understands straight JIS.) -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese
(Replying to myself again, but just for the record, ...) VIM...well, not great at Japanese. But an lovely editor. Just had to add it here. Works great in X11 on OS X, too! ;) Use it in freeBSD, trying to get it set up for openBSD, jvim. It would not make sense to use vim with Wnn and onew. With freeBSD, jvim/Wnn does fairly well, behaves like I would expect vi to behave with Japanese. It does flake out at times, however. I'll know more about how they work on openBSD today, I suppose. One of these days, I'll have to talk my employer into letting me use a Mac at work. -- Joel Rees [EMAIL PROTECTED]
Re: [OT] Japanese (correction about Mail)
I took another look at some garbled spam I seem to be picking up
regularly, which I had mistakenly assumed to be from a Korean source,
and it looks like Apple's mail app in 10.2.4 is _not_ handling 7-bit JIS
correctly. More later.
But, while I was checking that, I checked the following:
I managed to find the kanji I asked the person about with the charecter palette
description she gave, but it was or could be described otherwise as: Unicode
5782, JIS(X0213) 1-31-66, Shift JIS(X0208) 9082:
tarasu/tareru (hang down)
and it was mojibake'ed as
($BEZ(B)
That kind of looks like seven-bit JIS. The $B is a piece of a control
sequence when mixing 7-bit JIS with 7-bit ANSI.
EZ is the 7-bit JIS for tsuchi (earth, dirt). And BE is 7-bit JIS for
the da in datou (valid). Nope. Something else happened to that.
Some other codes she sent, and hence probably in the same encoding,
were $B7V(B and $Bj%(B both for hotaru.
7V is the 7-bit JIS for hotaru (firefly). j% is 7-bit JIS for a more
traditional rendering of hotaru.
Here's the meat of the source of a C tool I wrote to check:
-
for ( i = 0; i kTermWidth - 1; i += 2 )
{ unsigned long byte1 = (unsigned char) buf[ i ] - 0x21; /* kuten */
unsigned long byte2 = (unsigned char) buf[ i + 1 ] - 0x21; /*
kuten */
if ( byte1 == '\0' )
break;
byte2 += 0x40;
if ( ( byte1 1 ) == 1 )
byte2 += 94;
if ( byte2 0x7e )
++byte2;
byte1 = 1;
byte1 += 0x81;
if ( byte1 0x9f )
byte1 += 0x40;
buf[ i ] = (char) byte1;
buf[ i + 1 ] = (char) byte2;
}
buf[ kTermWidth ] = '\0'; /* training wheels */
-
(Yeah, C comes more natural to me than perl. Especially for this kind of
stuff. So shoot me.) It's missing the escape sequence and end-of-line
handling, among other things, but may be amusing to those interested in
the relationship between 7-bit JIS and shift-JIS.
some other strings are:
a$EAaD (this is the one I could decode)
Weird. All I can read out of that is kilogram told hits. Or, maybe just
the character hayai (early)?
$B0T$B0U$B0Gndc
Who's meaningful dark? Or perhaps the saba fish in the crucible?
Anyway, they _look_ sort of like 7-bit JIS, and the two you came up with
for hotaru are, in fact, 7-bit JIS for hotaru.
--
Joel Rees, programmer, Kansai Systems Group
Altech Corporation (Alpsgiken), Osaka, Japan
http://www.alpsgiken.co.jp
Re: [OT] Japanese
On Wednesday, Jun 18, 2003, at 20:01 Asia/Tokyo, Joel Rees wrote: (I'm still trying to decipher what they've done with the file system, and still trying to figure out how to get the terminal app to show the Japanese names for files. My brother in law has a book that shows a way that is supposed to even get it to show shift-JIS file names correctly in the terminal app, but I haven't got it to work on my iBook yet.) See Japanese in OS X 10.2 Terminal (in Japanese) http://member.nifty.ne.jp/poseidon/osx2t.html Good information there. Thanks. I'd tried using escapes to type in file names, but I did something wrong. File names in OS X are encoded in UTF-8 decomposition form. I.e. U+30C0: KATAKANA LETTER DA is represented as 0xE382BF (U+30BF) followed by 0xE38299 (U+3099). Decomposed. ta+dakuten. Okay, knowing Apple is de-composing the kana will be useful. UTF-8 aware tcsh is available as ftp://ftp.tba.org.tohoku.ac.jp/pub/tcsh-6.12-bin.tgz If you need install instructions in English, please see my posting to another list. http://listserv.dartmouth.edu/scripts/ wa.exe?A2=ind0306L=nisusT=0F=S=P=40940 (Of course, this is all off topic unless somebody wants to come up with some perl code for trying to undo garbled file names.) It's not a perl script, but nkf -- Network Kanji code conversion Filter -- is able to guess Japanese encodings and to restore broken JIS-Kanji. Well, I knew nkf was good for conversions, but I've never tried using it to to fix really broken text. ;) Actually, I was thinking more in terms of some code snippets that could be useful in recovering text that had suffered serious damage in a broken conversion pipeline. The garbled text in the original poster's message does not seem to be Japanese though. It looks like it didn't survive e-mailing. It looks like 7bit JIS, but converting down doesn't produce much that makes sense, as you say. The conversion that worked for hotaru is bad news, because it tends to indicate some serious non-deterministic behavior in the broken pipe. nkf is available as a part of jx package -- Japanese aware Unix tools for OS X. http://www.fan.gr.jp/~sakai/jx.html BTW there are free text editors which autodetect Japanese encodings properly in most cases. CocoaEditorJ (seems to be discontinued) http://cocoedit.hp.infoseek.co.jp/ http://cocoedit.hp.infoseek.co.jp/CocoaEditorJ.dmg (binary) http://cocoedit.hp.infoseek.co.jp/CocoaEditorJSource.dmg (source) KEdit (syntax colouring for perl, php, html) http://www.drycarbon.com/macosx/kedit/ http://www.drycarbon.com/macosx/archive/kedit010-20030619.zip (binary and source) Dunno if they would help you in making money though ;-) Heh. -- Joel Rees, programmer, Kansai Systems Group Altech Corporation (Alpsgiken), Osaka, Japan http://www.alpsgiken.co.jp
Re: Japanese + Encode::Guess
Thanks for your help Dan, but I'm mo further forward, the answer is apparently 'ascii', which is puzzling, because but the content is not ASCII - it is still legible in a web browser as it was written originally so the data is still intact. I'm guessing that Encode::Guess tests the beginning of the file to see what it contains, which being a HTML doc would have characters within the ASCII range? Thus, UTF-8, shift-JIS, or euc-JIS? Even 7-bit JIS apparently tends to be mixed with ASCII, so if your first n characters are nothing but ASCII, the guess is ASCII? Is there a parameter to force the sample length? (For five brief seconds, I was thinking about the value of randomizing the starting point for samples. :*/ ) On Wednesday, June 25, 2003, at 02:04 am, Dan Kogai wrote: print $enc-name; And it was a good thing he responded, because I was going to take a closer look at this tomorrow. (Thanks, Dan!) -- Joel Rees, programmer, Kansai Systems Group Altech Corporation (Alpsgiken), Osaka, Japan http://www.alpsgiken.co.jp
Re: [OT] Japanese (correction about Mail)
I took another look at some garbled spam I seem to be picking up regularly, which I had mistakenly assumed to be from a Korean source, and it looks like Apple's mail app in 10.2.4 is _not_ handling 7-bit JIS correctly. More later. rant Crud. I have some resume pages that I _know_ are shift JIS (looked at the byte values with hexdump), but Metrowerks Codewarrior 5 editor is won't play fair with them. Apparently, the pages I'm working with are ones which I pulled back off the web (stripping the resource fork) and edited with something that did not leave a traditional resource fork, but whatever the Mac OS X file system is using instead of a resource fork. Haven't had time to pin things down, but Mac OS X's Text Edit utility and the OS are playing strange guessing games on me, and the result is that, even with Classic booted and CW 5 set to the Osaka font, what it's showing me is as if the text were Mac (Latin) 8 bit. Pages which do not have the incontinuity of going to the web and back again are showing just fine. And Text Edit, when I save, adds another 8K to the file, just because I changed the encoding on the save. Four changes and 900 bytes of text is now 40K. Whoever convinced the architects for the Mac OS X utilities and OS to insist on converting the encoding when I just want to change the interpretation, I'd like to have a little heart-to-heart with. Converting should be NOT be under the Format menu in the editor in the dev tools. If Save As is not enough, we need another menu. What's under Format should only change the interpretion of the byte values. I'll post an update on the interactions in a day or two, unless someone throws a red flag about topicality. Maybe I'll be a bit more rational, too. /rant -- Joel, muttering Japanese expletives while wandering over to the Mac OS X bug reporting pages, if they are still there.
[OT] backslash when working with Japanese (Re: Pantherbites)
* input method is a mess! I like the old one ("disintegrated") better.
(B
(BYeah, I think the IM is going backwards as much as forwards, myself.
(B
(B * took 15 minutes to find how to enter '\' (backslash) instead of $B!W!"(B
(B (yen) ONCE Kotoeri is enabled. Once Kotoeri is enabled, '\' key
(B refuses to backslash and yens yens yens! Solution: Add "US Extended"
(B via [Input Menu] tab of [International] Preference Panel and use it.
(B
(BI was able to get the backslash in the Project Builder editor in Jaguar
(Busing the option key on the yen key or the forward slash key, I don't
(Bremember which and I didn't bring my iBook to work today. Do any of
(Bthose key combinations work? (I'll try to remember to check that tonight.)
(B
(B--
(BJoel Rees, programmer, Kansai Systems Group
(BAltech Corporation (Alpsgiken), Osaka, Japan
(Bhttp://www.alpsgiken.co.jp
Re: japanese in regular expressions possible?
This script then behaves as I would expect:
#!/usr/bin/perl
no warnings ;
$f = $ENV{HOME}/Desktop/biao.txt ; # file saved as UTF-8
open F, $f or die $! ;
$/ = \015 ; # only if the file has Mac line endings !!!
print \x{8868}\n\n; # prints the character as utf-8
for (F) {
print 1. $_ if /Ë°®/ ;
print 2. $_ if /\xE8\xA1\xA8/ ;
}
Note that if you write use utf8; it will NOT work
Thanks, John
I'll give it a try, soon as I can get 5.8 installed on my iBook.
(Running short of disk space, so I've been putting it off.)
Joel (at home)
testing my rules, please wink at this and move on
Now why'd you look?
Re: [OT] OSX privileges question
So I've been put in charge of setting up and maintaining our department's new dispatch/switchboard computer. In trying to keep it clean and in order, I was hoping, if possible, to be able to give users read/write access to information in files themselves, but to block them from renaming the files or moving them. I tried giving r-x access to a folder and rwx access to the file inside. This lets them open the file and prohibits them from moving/renaming it, but prohibits them from saving any changes (because they can't write to the folder). That sure isn't the way I understand it. I'll check when I get a chance whether that's the effect on Mac OS X. Are you trying to use the GUI File Info interface from Finder, instead of chmod/chown, perhaps? I've found it handy to set up dummy users and empty groups. Some careful thought will often yield the functionality you need. For instance, the first thing I might try (if I understand your question correctly) is to set up an empty group called deptA with netinfo, dummy user called deptA, complete with a file heirarchy under /Users, BTW, /etc/groups is ignored after Mac OS X goes multi. You'll need to add users to groups other than their primary group through the Netinfo Manager utility. (GUI access in the utilities folder.) -- Joel Rees, programmer, Systems Group Altech Corporation (Alpsgiken), Osaka, Japan http://www.alpsgiken.co.jp -- When software is patentable, anything is patentable. (http://swpat.ffii.org)
Re: Weird math...
Fixed-point math (generally with 4 decimal places and an implied decimal point on a 32 or 64 bit integer) is more appropriate in those cases. Math::BigFloat, perhaps?
Re: Looking for old laptop/notebook
drieux, (Sorry about taking so long to respond -- further follow-ups off list, of course.) On 2003.11.9, at 05:15 AM, drieux wrote: Begin forwarded message: does anybody know how to make Hypercard run on System 9 or X? I've been running a few hypercard apps on Mac OS 9 and X without much problem. Of course, they aren't complicated apps, and I had to disable an XCMD on one to run it. does anyone here have a good pointer on how to get old style Hypercard stuff running on OSX? a friend of mine in the UK was wondering. The stuff works under the older Apple 8.5 but apparently not in 9 or OSX. If there are problems, my first guess would be that the apps require XCMDs and/or XFCNs (or whatever those were) that were written to an API that no longer exists. If that's the case, he'll probably either need to get the source code for those and port them, or simply run them on an older system. Depending on how vital the old hypercard apps are, and the difficulty of getting the source code or porting them, or of getting old hardware that is fast enough and dependable enough, he might want to look into Mac-on- Linux: Buy a new Mac, wipe Mac OS X, install Linux, install MOL on that, install the older system on top of that. It should work. (YMMV) Since the emulation is API-level only, they should run at native speed. It would be nice if the Linux API emulation layer ran on Darwin, or if MOL ran native on Darwin, but I don't believe either of those have been done yet. One other thing to look into if he finds himself getting into porting (besides Perl, of course) -- Runtime Revolution (which has bought the Metacard engine technology) maintains a high level of compatibility with Hypercard. Metacard/RunRev has an additional advantage of being cross-platform. I think I've also heard that Supercard has been updated to Mac OS X, but I'm not sure. HTH Joel
Re: searching complex datastructures
Can anyone offer an elegant solution for a data structure that maintains sorted order as well as access to data for a (primary) key? Is everyone thinking too hard or am I not thinking hard enough? If you have a database and you need to search it on an alternate key, you either linear search on the alternate key or define the alternate key formally in the database as an alternate key. Defining an alternate key is just setting up a reverse-lookup index table. (Conceptually.) So, if you don't want to linear search, and you don't want to build a reverse-lookup hash, your usual option would be to build something that works as an alternative way to reverse-lookup, say a binary tree or a trie or even a linear sorted array to do a binary search on. Right? Deciding between the various alternatives is just weighing the cost of each approach, and deciding which cost you can afford best. Right? I guess I just don't understand the thread. reiisi
Re: get contents from clipboard
On 2003.12.15, at 07:10 AM, John Delacour wrote: At 5:01 pm +0900 11/12/03, Robin wrote: late in on this one but you can treat the clipboard as a filehandle if you pipe to pbpaste and pbcopy : open (FROM_CLIPBOARD, pbpaste|); open (TO_CLIPBOARD, |pbcopy); you can then do as you normally would for moving data to and from fle handles. See the typically useful-in-a-real-world-situation example script below So far as I can see there is no way to use this useful technique with a clipboard containing Unicode text, for example Chinese or Greek text copied from TextEdit. Am I right? Why do you think it won't work? Is there some irreversible munge applied to the pipes or the clipboard? Or is it the guessing game about being sure it was Unicode text? reiisi
Re: DBD mysql unable to install since Panther
Thanks everyone. The developer tools I installed is the developer tools from Jaguar. I don't have the Panther developer tools. But I will get it. Is there an easy place to buy the tools from? Apple seems to want to sign me up as a developer, etc. If you don't mind the agreement they make you sign, it's definitely worth signing up. You pretty much have to sign that thing anyway, whether you download the developer tools or install from the system. If you own Panther, you should have them already. It's an optional install. Run your installer and customize the install. You'll have to go to the applications folder and dig into a folder called something like installers. Then go to Apple's site and get the XCode update 1.1. Maybe it's also accessible via Software Update. It is.
Re: Soliciting opinions from Applescript refugees
On 2004/02/28, at 0:08, Chap Harrison wrote: ... and I wonder why some people swear by Applescript. I think it may be from inexperience. Yours or theirs? (heh.) As has been pointed out, FileMaker is still more of a RAD tool than a solutions tool, and AppleScript also. Precision in a language is a requirement in contradiction with ambiguity, ergo, flexibility. I think a large part of Perl's success is the ability to go from a relatively flexible syntax to a relatively precise syntax within the same language. (As opposed to Java, for example, where precision can't be escaped, but is relatively uncluttered by language artifacts, or SQL, where you have to escape the language to get precision in all but a few business contexts.) I think I agree with John and Doug, your best approach is to expect to use multiple tools, which is why mailing lists are useful.
Re: Web servers with cable DSL
On 2004/03/16, at 11:13, Bill Stephenson wrote: I was wondering if anyone here is using a MacOS X box with a fixed IP cable DSL account Cable with fixed IP? Does it exist? My current US Cable provider told me I could do static IP if I put my own router between the cable modem and my internal network, but the sales rep was apparently talking about the internal-only ranges. Their license proscribes published static IP. For similar pricing, I could have got full support for published static IP through phone company (A)DSL. In Japan, I have ADSL (1M) broadband over telephone, renting a full configurable modem/router, for just under half of what the stateside cable company is asking for a similar setup. as a commercial grade web server? Don't know, but, if you really can get your cable company to let you use fixed IP, such a setup should be about as good as a similarly outfitted Linux or FreeBSD system. Well, plus or minus a bit, depending on what server software and modules you're using, of course. If you go ADSL, your visitors' download rate is limited by your upload rate, of course. Is this a reasonable alternative to using a hosting company like Verio? I haven't tried it yet, I'm kind of planning on something like this working for me. I have heard some success stories with openBSD, but I haven't been hanging around where I'd hear about such things on Mac OS X. I could sure save some cash by switching to this set-up but I have concerns about performance and reliability. So does your ISP, of course. That's why they want to sell you something quite a bit more expensive, instead. Will DSL provide enough bandwidth to 2-5000 visitors a day for web sites that serve standard HTML and web graphics? (ie. no broadband media like video, mp3, or other streaming media formats) When's rush hour? Any help and advice will be much appreciated. I'd like to hear someone else's experience, too. (Before I try it myself, I mean. First, I'm going to try dynamic DNS, I think.) Joel
Re: Web servers with cable DSL
On 2004/03/16, at 12:29, Chris Devers wrote: On Tue, 16 Mar 2004, Morbus Iff wrote: Please don't make the web a world of Geocities. On the other hand, it has always kind of bugged me that having a fully functional web server out of the box isn't seen as a normal part of having interenet access, or more simply, a network connection. One day, when Microsoft quits sucking the money and bandwidth out of the internet, phone service will come with hosted web sites, with the option of hosting our own, in much the same way that we currently have answering services and answering machines. I guess that 'll make it a sort of world of geocities minus the ads, lusers, l33tz, and exploits. And google will run the yellow pages. WIBG
[OT] Re: Web servers with cable DSL
On 2004/03/17, at 11:29, Bill Stephenson wrote: Well, I think that Kevin (morbus) really did a good job of pointing out why I can't entirely do this yet. Some of the sites I host are critical to the businesses that use them and Verio has always provided a great service. Because they host on FreeBSD, developing on the Mac and porting to Verio is almost seamless even though Verio has never done anything special to accommodate this. However, the fact that so many on this list are hosting sites with cable DSL indicates that I can possibly move some of the sites I host to a home office based server and still save a little money. I'll spend some time reviewing the sites and costs and see how the numbers crunch. (I noticed with a certain amount of surprise that Verio seems to have merged with NTT Communications, which is not Nippon Telephone and Telegraph, but used to be a daughter company of the Japanese national telco before Japan decided to pursue the competition path like the US has.) I was surprised recently to notice that my local telco is now offering a really stripped-down 256K line with no-frills (no mail, no web space) ISP for about $22 a month, and static IP addresses for fairly reasonable. What about using http://directv.direcway.com/ to host servers? Anyone doing that? Says you need a clear southern exposure. I'd wonder about trees and cloudy days, too. And I didn't see anything about the download/upload differential, but my memory is its about the same as cable.
[OT] slice vs. splice
slice syntax isn't deprecated or anything is it? Don't see it mentioned in O'Reilly's Nutshell or in the Cookbook's section on arrays. (Sorry about the not-really-topical noise.)
Suggested version for Mac OS X.2?
Gah! That's what I get for using comments! ;-) Darn right, stop that! Code is hard to write, it should be hard to read, too! ;-) 8-) That's what I like about perl. Anyway, I'm aware that slice is not a function, just surprised that neither the concept nor the syntax seems to get any treatment in either Nutshell 1st Ed. or Cookbook 2nd Ed.. (I don't own the Camel, just went straight from the Llama to Nutshell. Thought I was saving money at the time.) Well, according to the blding edge Perl 5.8.4-rc2 docs: http://search.cpan.org/~nwclark/perl-5.8.4-RC2/pod/ perldata.pod#Slices Thanks for the pointer, merlyn. So, I'm wondering about that version number. 5.8.1 is still the latest stable Perl, right? Joel
Re: Suggested version for Mac OS X.2?
So, I'm wondering about that version number. 5.8.1 is still the latest stable Perl, right? No, 5.8.3 is the latest. And 5.8.4 will likely be out within a week. So, given an iBook that is going to host my personal site, should I load 5.8.3 parallel to the the 5.6 in Mac OS 10.2, or should I stick with 5.8.1? (Purpose of the parallel load is to keep the one used by the system more-or-less pristine, of course. Purpose of 5.8 is Unicode and large character sets.) Joel
Re: Advice for enabling perl on OSX
On 2004.4.22, at 10:59 PM, Eric Curts wrote: ... I loaded a simple script to just test things out (one that just prints out environment variables) and it will not run. When I try to bring up the script I get: What command line are you using? Forbidden You don't have permission to access /hck/cgi-bin/printev.cgi on this server. What permissions do you have on the directory /hck/cgi-bin, the directory /hck, and the root directory? Also, who owns (user and group) the file and each parent directory going up? Since the message is about access, I would expect the problems to be something in permissions and user/group. However, I'll add the following comments to what the others have said: --- - Apache/1.3.26 Server at eagle Port 16080 I set the permissions to 755 for the script, I prefer to use 750, making sure the script is owned by the www group, myself. But the looser permissions should give you access, if the elements of the path up the line are accessible. and the folder is web accessible because html files open up fine from that location. As has been pointed out, script and hypertext access are set set separately. Since you call the directory /hck/cgi-something, I assume this directory is for scripts, and you are keeping the scripts and html separate. That's good. But in that case you wouldn't want html files in that directory accessible to the web. When I make a terminal connection and try to run the script from the command line, it will not work either. What's the system's complaint? I think it may be a more fundamental problem, such as needing to edit something in their httpd.conf file or needing to enable something else on the server so that perl scripts will be executed. They have never runs scripts before, so nothing has every been set up for this. My memory is that ... . Wait, I have the client version, not the server version, so I couldn't say for sure whether the primary script directory is enabled in httpd.conf in your case. In any case, /hck/cgi-bin is most likely not enabled for scripts. As mentioned, you need a section in httpd.conf that looks something like ScriptAlias /cgi-bin/ /hck/cgi-bin/ Directory /hck/cgi-bin AllowOverride None Options None Order allow,deny Allow from all /Directory to allow access from the web. You'll also may need to set up httpd.conf to allow access from the high port. I think a listen line may be enough to get you started, although you may have to do more than that before you go to production. I would appreciate any suggestions you have for this problem, especially an idea of what configurations are needed the first time to get an OSX server to execute perl scripts. Well, read the comments in httpd.conf carefully. We only have vague guesses, since we don't know your setup. In case you aren't aware, Apple leaves the on-line manual accessible where the apache foundation folks were thoughtful enough to put it, at http://localhost/manual/ Joel Rees
reinventing wheels (was Re: Image::Magick)
On 2004.4.28, at 03:58 AM, Chris Devers wrote: ... Hand-rolling popular software from source is nice and all, but how many times does the wheel need to be re-invented, ya know? :-) sherm: The first wheels were simply logs placed under a sled; ... We probably shouldn't clutter the list with module requests, so email those to me privately, and I'll summarize to the list in a few days. Like the analogy, Sherm. (Apologies to Chris Devers and to the list for cluttering it with noise, but this thread made me think of a new tag-line.) -- Joel Rees If God hadn't meant for us to tweak our source code, He'd've given us Microsoft. ;-P
Re: reinventing wheels (was Re: Image::Magick)
Joel Rees If God hadn't meant for us to tweak our source code, He'd've given us Microsoft. Joel, Don't credit God for this thing. It is of the devil. What? You thought I thought God meant for us _not_ to tweak our source code? ;- -- Joel Rees Opinions are like armpits. We all have two, they all smell, and we really don't want the other guy to get rid of his.
LC_ALL for daemons
I'm sure I've seen a thread on this, but a casual search didn't turn it up. (I'm always looking in the wrong places.) When I'm logged in as a user, I can set the appropriate environmen variables, but when a daemon is running, where is it going to get them? Joel Rees
comparison always false is a problem or not?
My experience is that this kind of thing tends to lead to dead code or endless loops. Do I need to dig in and find the macro declaration and see if I can fix it? warning: comparison is always false due to limited range of data type regcomp.c:724 pp_sys.c:302 byterun.c:898 re_comp.c:724 Joel Rees
Re: comparison always false is a problem or not?
(BOn 2004.4.30, at 04:38 PM, Sherm Pendley wrote: (B (B On Apr 30, 2004, at 2:30 AM, Joel Rees wrote: (B (B My experience is that this kind of thing tends to lead to dead code (B or endless loops. Do I need to dig in and find the macro declaration (B and see if I can fix it? (B (B warning: comparison is always false due to limited range of data (B type (B (B In my own code, I compile with -Wall and try to chase down and (B eliminate all warnings. Occasionally I might use a -Wno-something to (B turn off a warning once I've determined that it's harmless, but I (B really, really prefer to fix them. (B (BMe too. (B (B I treat other people's code differently. If I'm building something (B from source and it emits warnings, I'll make a mental note of that (B fact. Then, if it's crashy, buggy, or exhibits some other odd (B behavior, I might go back and see if the warnings are relevant to the (B problems. If the app works fine though, I don't worry about the (B warnings. (B (BThat's kind of what I was figuring would do, but I also would prefer (Bnot to find my code skipping a parse because the code got thrown out. (BSo I thought I'd check. (B (B Those file names look familiar - are you building Perl? (B (BOf course! (I suppose I should have said so.) (B (B If so, and the self-tests pass, (B (BWell, since you're kind enough to ask (heh heh), here's a few more I (Bget: (B (BIn the make phase, (B (B ... (B cc -flat_namespace -L/usr/local/lib -o miniperl (J\(B (B miniperlmain.o opmini.o libperl.a -lm -lc (B ./miniperl -w -Ilib -MExporter -e '?' || make minitest (B make: [extra.pods] Error 1 (ignored) (B ./miniperl -Ilib configpm configpm.tmp (B sh mv-if-diff configpm.tmp lib/Config.pm (B ... (B Making Errno (nonxs) (B Writing Makefile for Errno (B ../../miniperl "-I../../lib" "-I../../lib" "-I../../lib" (B"-I../../lib" Errno_pm.PL Errno.pm (B cp Errno.pm ../../lib/Errno.pm (B make: [extras.make] Error 1 (ignored) (B (B Everything is up to date. Type 'make test' to run test (Bsuite. (B (Band also the range warnings mentioned already. In the make test phase, (B (B ... (B ext/DB_File/t/db-btree...# (B # This test is known to crash in Mac OS X versions 10.2 (or earlier) (B # because of the buggy Berkeley DB version included with the OS. (B # (B FAILED at test 0 (B ext/DB_File/t/db-hashok (B ext/DB_File/t/db-recno...# (B # Some older versions of Berkeley DB version 1 will fail db-recno (B # tests 61, 63, 64 and 65. (B ... (B # You can safely ignore the errors if you're never going to use the (B ... (B # (B FAILED at test 64 (B ext/Devel/DProf/t/DProf..ok (B ... (B Failed 2 test scripts out of 804, 99.75% okay. (B ### Since not all tests were successful, you may want to run some of (B ... (B ### Bourne-style shells, like bash, ksh, and zsh, respectively. (B u=25.78 s=0 cu=599.71 cs=171.77 scripts=804 tests=80310 (B make[2]: *** [_test_tty] Error 1 (B make[1]: *** [_test] Error 2 (B make: *** [test] Error 2 (B (BIt says I can ignore those, so I did a make install and got (B (B... (B Making Errno (nonxs) (B make[1]: [extras.make] Error 1 (ignored) (B (B Everything is up to date. Type 'make test' to run test (Bsuite. (B if [ -n "" ]; (J\(B (B then (J\(B (B cd utils; make compile; (J\(B (B cd ../x2p; make compile; (J\(B (B cd ../pod; make compile; (J\(B (B else :; (J\(B (B fi (B ./perl installperl --destdir= (B WARNING: You've never run 'make test' or some tests failed! (B(Installing anyway.) (B (Bwhich worried me a little. So I went ahead and ran t/harness after the (Binstall completed. (That's backwards, I suppose.): (B (B ... (B op/lex_assign..ok (B op/lfs.skipped (B all skipped: writing past 2GB failed: process limits? (B op/listok (B ... (B ../ext/DB_File/t/db-btree..# (B # This test is known to crash in Mac OS X versions 10.2 (or earlier) (B # because of the buggy Berkeley DB version included with the OS. (B # (B ../ext/DB_File/t/db-btree..dubious (B Test returned status 0 (wstat 10, 0xa) (B ../ext/DB_File/t/db-hash...ok (B ../ext/DB_File/t/db-recno..# (B # Some older versions of Berkeley DB version 1 will fail db-recno (B # tests 61, 63, 64 and 65. (B # (B # For example Mac OS X 10.2 (or earlier) has such an old (B ... (B ../ext/DB_File/t
Re: LC_ALL for daemons
On 2004.4.30, at 04:30 PM, Sherm Pendley wrote: On Apr 30, 2004, at 2:24 AM, Joel Rees wrote: When I'm logged in as a user, I can set the appropriate environmen variables, but when a daemon is running, where is it going to get them? Daemons are started from scripts found in /System/Library/StartupItems (for Apple-provided daemons), or /Library/StartupItems (for your own). Startup scripts usually include /etc/rc.common, so if you want to export an env variable for *all* daemons, that would be a good place to do it. Be aware that /etc/rc.common is a system file though; Apple-supplied OS updates might overwrite it, so keep a backup just in case. Thanks. I needed that information, because I had forgotten that Mac OS X doesn't use the rc/* convention. Don't look like there are any hooks for an rc.local. An rc script for the daemon would be handy for setting an environment variable. Hmm. There are plists in those startup items. I wonder if I can pervert those. Time to dig out my copy of in-a-nutshell and see if there are enought clues there. If you're writing a daemon, Well, I'm still vacillating between a cron job and a daemon. I'm writing my own update tool for dynamic dns for the experience. (My ISP wants $60 a month for one static IP.) It'll screen-scrape the router/modem's setup pages for some Japanese text. I'm sure I could modify somebody else's script, but I have this cowboy mentality, at least until I learn how to read Perl for meaning instead of just function. you might also want to be aware of /etc/hostconfig. Yeah, I needed to look in there, too. This sets a series of flags that indicate what daemons should be started. At system startup, *all* of the items in the StartupItems folders are run, so they check for the corresponding flag to see if they should actually start their service. Yep, that's where I can turn sendmail on, once I figure out some configuration details, and I want that so my script can report problems updating and not just log them. Well, that's for another day. Thanks, Sherm. I have to admit I get lost less with a little prompting. -- Joel Rees
questions from configure
Okay, one more that had me curious, from the configure script -- vfork() found. Perl can only use a vfork() that doesn't suffer from strict restrictions on calling functions or modifying global data in the child. For example, glibc-2.1 contains such a vfork() that is unsuitable. If your system provides a proper fork() call, chances are that you do NOT want perl to use vfork(). Do you still want to use vfork()? [y] Do I take this to mean that Mac OS X (10.2.8) still doesn't have a proper fork for Perl's purposes? (for Perl 5.8.4) And, the default configure suggesting against threads, is that because Perl's threads are still in the oven, or because Mac OS X's threads are still a little unripe, so to speak? -- Joel Rees
Re: backing up system
Perhaps it's because I'm not strong on Perl yet, but I took a bit more of a naive view here -- On 2004.5.1, at 05:22 AM, Joseph Alotta wrote: Greetings, I try to back up my system once a week. I have a firewire disk drive that I use for this purpose. I have been using the Lacie software that came with it. Before Panther, I used to be able just to plug it in and run it under my own id. Now I need to log in as root to run it. Which means I can't do anything else until it finishes, Does logging in concurrently as root not work? Not that I'd urge you to leave your root account enabled for logging in, concurrently or otherwise. (I just tried, for grins, under 10.2.8, su-ing to an admin user, then sudo-ing a sh to get a root shell without logging in as root, but open-ing /Applications/AppleWorks 6 as the root user didn't seem to do anything other than opening the /Applications directory in a GUI window. open-ing /Applications/TextEdit.app as root runs TextEdit, but the process is owned by the user I'm logged in as. sudoing the open directly from the admin user yields complaints about not being able to map display interlocks or open default connections, etc. That's not Panther, of course.) and it takes about 40 minutes. I am looking to do something more automatic. Questions: 1. Can iSync be used for backups? I'm not sure if I have iSync unless it is standard in Panther. Well, Apple's blurbs seemed to say such things, but I think, when I read the fine print, it was for backing up to your .mac account. 2. Otherwise, has someone wrote a perl program to do this that I can run in cron. Wasn't there a related thread here just this last week, including mention of either CpMac or ditto? I'll shut up now. -- Joel Rees Opinions are like armpits. We all have two, they all smell, and we really don't want the other guy to get rid of his.
Re: backing up system
I just tried it, and the application can be downloaded installed just fine, but if you try to run it you're asked for a .Mac login to procede. I don't suppose there was an option for selecting the .Mac server? I have some vague memory that Mac OS X server includes the ability to set up and provide several of the services provided on .Mac, but I might have been hallucinating. Sometimes my dreams seem pretty real when I don't get to bed until four-ish. Maybe I should see if I can find something on that. -- Joel Rees If God hadn't meant for us to tweak our source code, He'd've given us Microsoft.
[slightlyOT] reading logs with long urls
My apache log files show that I'm getting two or more of those long url attacks every day, and access_log grows to over 4Mb in just a week, in spite of the fact that there are less than ten valid accesses in any particular day. So, I'm going to write a daily script to compress-and-rotate. (4M compresses easily to less than 40K, since it's mostly those stupid attacks.) That's no big deal, I think, although I may pop up with specific questions on that later on. (I assume either the root kit is dead stupid or my ADSL modem is fooling it. I suppose I should check myself on Netcraft some time.) I also built a little C filter to get the attacks out of the way. I used C because all the vectors are around 32K, and it's easy enough to just use a 128 MB input buffer and look at the length. (It's an interactive tool, so an unexpected really long line will at most kill my shell.) I think I was told by someone that Perl's input buffer would adjust to this kind of insanely long line. Does it slow the input down much to have to re-allocate the buffer? The reason I'm asking is that, for now, my filter is just killing the long lines. I'm thinking some visible RLL could make those easier to see as well as easy to see around. (Not sure what I want to see in them.) (And sometime I'd like to build an error page script that would dump 64K from /random back at the zombie. But I have more important things to do first.) -- Joel Rees Getting involved in the neighbor's family squabbles is dangerous, but if the abusive partner has a habit of shooting through his/her roof the guy who lives upstairs is in a bit of a catch-22.
Re: [slightlyOT] reading logs with long urls
My apache log files show that I'm getting two or more of those long url attacks every day, and access_log grows to over 4Mb in just a week, in spite of the fact that there are less than ten valid accesses in any particular day. How about configuring Apache to disregard (and not log) any URL longer than a predefined length? The default settings are correct for rejecting the long URLs, and reporting the attempts is correct behavior. Handling the large logs is a time tax on using inherently incomplete technology. Also, what are those long url attacks, I haven't heard of them. See Daniel Staal's post. Fortunately, our Mac OS X boxes are at present somewhat immune to the code insertions anyway, because the code is almost always x86 code. Part of the purpose of wanting to compress out the bulk of the long url is to make it easier to tell if/when we start getting powerPC code insertion attempts. -- Joel Rees Complaining about systems that are incomplete misses the point. In this world, a system can't be perfect and useful at the same time. Of course, that's no excuse to refuse to fix problems -- we'll never run out of problems.
Re: [slightlyOT] reading logs with long urls
Another idea might be to sleep proportional to the length of the URL. For example, 1k request get 1 second, a 2k request gets 2 seconds. At least this will slow them down. Fortunately, these attacks are not DOS, just looking around for boxes to 0wn. So it's just a matter of rotating the logs properly, and compressing them if I want to keep the old logs around. (At this rate, a year's worth of uncompressed logs would consume my hard drive.) And, of course, the problem of viewing the logs, which is where this thread is closest to being on topic here. Writing a viewer will help me understand the perl approach to parsing characters. Not that I don't understand it, but it's hard for a guy who essentially cut his teeth porting Forth to get used to. -- Joel Rees
Re: backing up using DejaVu
Hmmm, maybe it will work... I started the following shell script #!/bin/sh i=1 while true do echo $i i=`expr $i + 1 ` done I forced the system to sleep for a few seconds, woke it up and the script kept on trucking... While you always want to avoid doing things that push the limits of the system while you're backing it up, in theory, all the timing issues should be handled by the drivers before they allow the system to go to sleep. Or, if some hardware job is in progress when the system goes to sleep, the hardware should be waking the system up to finish the process. So it _should_ keep on trucking. If it doesn't, it's a system bug, and needs to be reported. But risking dropping your backups to find Apple's bugs may or may not be what you want to do, and you can be sure that backups won't proceed while the system is asleep. I think this question has come up before, but if there is an API for the Energy Saving settings (or whatever those are in English), I think I'd want to have my backup script set the system to not sleep while the backup was in progress, and then restore the setting when done. Until that part is working, I'd change the setting by hand, of course. -- Joel Rees
sudo and cpan
Okay, I seem to have forgotten how to use CPAN. Where are the detailed instructions? (perldoc cpan only gets me a page.) And while I'm making noise, When you have perl 5.6 as the system perl (/usr/bin) and perl 5.8 as a parallel install in /usr/local/bin, you want to set your user's path to put /usr/local/bin in front of /usr/bin before you run cpan, so cpan doesn't get confused, right? What's the difference between perl -mCPAN (or whatever that was) and /usr/local/bin/cpan? Is it normal for cpan to not ask to be set up after a fresh install of 5.8.4? It prompted me to install Bundle::cpan, but didn't run through the list of ftp services and so on, etc. that I remember from previous times. Does cpan have to be run as root? What happened when I tried to install Bundle::cpan as an admin user, then ran sudo cpan? Was I supposed to do a clean or something first? It installed the second time, when I sudoed it, didn't fail any tests, but skipped some tests because I didn't tell it the location of any ftp servers to test off of and things like that. I was surprised that it didn't seem to know my local host name. Now it seems to be unable to release the locks when I quit unless I start sudoed. And I can't find the instructions about querying which modules are installed, either. Maybe it'd help if I had gotten more sleep last night. -- Joel Rees
Re: sudo and cpan
On 2004.5.6, at 08:08 PM, Joel Rees wrote:
Okay, I seem to have forgotten how to use CPAN. Where are the detailed
instructions?
(perldoc cpan only gets me a page.)
And while I'm making noise,
When you have perl 5.6 as the system perl (/usr/bin) and perl 5.8 as a
parallel install in /usr/local/bin, you want to set your user's path
to put /usr/local/bin in front of /usr/bin before you run cpan, so
cpan doesn't get confused, right?
What I'm thinking about is learning enough sh to split the path and
insert /usr/local/bin in the middle, because I really don't want to put
set path=(/bin /sbin /usr/local/bin /usr/bin /usr/sbin)
in my users's .bash_profile, etc. It's not like it's going to change or
anything, but it just feels wrong.
set path=(/usr/local/bin ${path})
also feels wrong. It avoids walking on the system provided path, but it
gives /usr/local/bin priority over /sbin. If everything in /sbin should
be called by full path anyway, why is /sbin in the path at all?
Maybe this is really a question for a security list.
What's the difference between perl -mCPAN (or whatever that was) and
/usr/local/bin/cpan?
perl -MCPAN -e shell
is interactive.
Is it normal for cpan to not ask to be set up after a fresh install of
5.8.4? It prompted me to install Bundle::cpan, but didn't run through
the list of ftp services and so on, etc. that I remember from previous
times.
Evidentally I don't run CPAN non-interactively very much.
Does cpan have to be run as root? What happened when I tried to
install Bundle::cpan as an admin user, then ran sudo cpan? Was I
supposed to do a clean or something first? It installed the second
time, when I sudoed it, didn't fail any tests, but skipped some tests
because I didn't tell it the location of any ftp servers to test off
of and things like that. I was surprised that it didn't seem to know
my local host name. Now it seems to be unable to release the locks
when I quit unless I start sudoed.
Still don't find much on this one.
And I can't find the instructions about querying which modules are
installed, either.
Okay,
perldoc perllocal
perldoc perlmodlib
Maybe it'd help if I had gotten more sleep last night.
Reading the faq helps, too.
--
Joel Rees
Re: OS X
On 2004.5.6, at 09:01 PM, Stephen Harris wrote: Hi, Decided to teach myself Perl and the got Sam's Teach Yourself Perl in 24 hrs (book CD ) from the local library. I began working with the book and found that perl 5.6.1 only runs in Classic environment, That would be macPerl 5.6.1? but that the Unix version of perl runs fine in OS X. OK, maybe, but how do you get it in there? It's in there. Try this in the terminal: ls -l /usr/bin/p* and then this: perl --version which will show that it is 5.6.0. If you want the latest perl, 5.8.4, that does have to be installed (for instance, if you are like me and need the Unicode support). (I'm running OS X 10.2.8 on a dual processor G-4 silver Power PC.) Thanks S. M. Harris
Re: sudo and cpan
On 2004.5.6, at 09:58 PM, Chris Devers wrote: ... But anyway, back to your original question. Your /usr/bin/cpan should just be a little Perl script that amounts to little more than this: $ /usr/bin/perl -MCPAN -e shell hmm. I could have sworn I'd ended up with non-interactive behavior when I just ran /usr/local/cpan. Maybe it was just that all the defaults were already set from installing 5.8.4 from source. Thanks. Any thoughts about the file locks left over when I run without sudo and quit? This doesn't happen when I'm running cpan as a non-amin user. Or maybe I should say it doesn't happen as a user other than the one I installed as, but I haven't tried running as a non-install admin user yet. I'm thinking too hard, I know. Time to install XML and get back to work. -- Joel Rees
Re: Anyone got a perl script to catch the disk: and help: uris in web browsers?
On 2004.5.26, at 10:49 PM, Chris Nandor wrote: In article [EMAIL PROTECTED], [EMAIL PROTECTED] (Joel Rees) wrote: Macintouch is showing an AS script: http://www.macintouch.com/#notesandtips I've never been comfortable with AS, so I'm thinking about installing it to see how it works, then re-writing it in perl, and maybe even further trap all attempts to climb the directory tree in the uri. You can use Mac::InternetConfig to disable the protocol handlers. But RCDefaultApp is a better solution for this. Yeah, but neither of those are nearly as interesting. Of course, since Apple has already done the easy part, it doesn't make any sense to try to build it myself. The perl script in the patch they distributed is kind of interesting. Any thoughts on the protocol registration issue or the login items issue? I'm thinking it might be useful to have a (say, perl? heh.) script clean up registered protocols and login items that were not explicitly okayed by the user, for one thing. -- Joel Rees If God had meant for us to not tweak our source code, He'd've given us Microsoft.
Re: Whither Developer Tools?
If not, you can always copy if off your collegue's hard drive. It being the installer package, of course. It installs stuff all over the place - apps, documentation, headers, libraries, etc., etc. Trying to copy all that stuff manually would be a massive pain. I'm pretty sure that's what Chris meant, but I figured it was worth spelling it out for the sake of the archives. (As if anyone reads them...) I guess if we are going for completeness in the archives, someone should mention that, if you re-partition your hard disk and re-install from the DVDs, the dev tools/xcode installer will end up in the same place again. (Unless you do a custom install and deliberately deselect it, that is.) It's on the install DVDs (or CDs) for new machines, it's just not there as a separate package.
Re: Installing modules on osX 10.3.3
On 2004.6.15, at 06:08 AM, Bill Stephenson wrote: On Jun 14, 2004, at 3:12 PM, Sherm Pendley wrote: I'm curious - is reinstalling the OS a common troubleshooting technique for older MacOS versions? I'm a fairly recent switcher myself. I purchased my first Mac expressly to run Mac OS X DP4. Yes, it certainly was for Systems 7-8. Clean System install, And there was that sweet spot where the system only took half an hour to re-install. When we were working in that sweet spot, re-installing made eminent sense. But that sweet spot would go away. Apple would add necessary functionality, and, of course, we would add apps and extensions so that it would consume a whole half a day to re-install the system and apps. I always that those half-day re-installs were terrible, until I starting helping with MSWnt4 or OS2Warp installs at work. then install apps one at a time until a failure (conflict) occurs. The Extensions Manager helped a bit, but I always shied away from using it and learned to manage conflicts manually instead. I started using the extensions manager to move everything out of the extensions folder before installs. Then I could compare what the app installed with what had already been in place. Leaving only the most recent version of extensions and disabling as many Microsoft extensions as I could was usually pretty effective. I used to restart my 6100 and then my 8600 several times a day just to avoid System freezes. You could just feel one coming on after a bit. I sure don't miss that. System 9 was a pretty big improvement, but after several years my 8600 started acting weird and now it surely needs a clean install to make things work again. YDL? NetBSD? ;-)
Re: Location of Files to be Printed
Dunno about Jaguar earlier. [reiisi-rend:~] family% ls -la /var/spool/cups ls: cups: Permission denied heh. Permissions are correct. Anyway, cups could be got to run (with a lot of patience, as I recall) in 10.0. It ran, if not with fully satisfactory results in some cases, in 10.1. I haven't noticed any serious issues in 10.2. Obligatory Perl Reference: Although CUPS is installed and working (for both native Mac and Fink-installed Unix apps), the development headers are nowhere to be found, so Net::CUPS doesn't want to compile. I'm sure I've seen stuff for cups on Apple's developer's sites. Can't say where, or if it was enough to compile Net::CUPS.
Re: CPAN can't write makefiles after perl 5.8.1 reinstall on 10.3.4
On 2004.7.15, at 06:43 AM, Brian Dimeler wrote: Hi, I was having trouble installing LWP on the version of Perl that came with our office iMac (running OS X 10.3.4) and therefore I reinstalled Perl as per the suggestion of CPAN, using the guide at http://developer.apple.com/internet/opensource/perl.html . my, oh, my. Would someone else take a look at that page and tell me whether they agree with me that it looks like some summer intern at Apple has kind of laid a little land mine in there, maybe thinking he was bringing the page up to date with Panther? ...
Re: CPAN can't write makefiles after perl 5.8.1 reinstall on 10.3.4
(B Would someone else take a look at that page and tell me whether they (B agree with me that it looks like some summer intern at Apple has kind (B of laid a little land mine in there, maybe thinking he was bringing (B the page up to date with Panther? (B (B Absolutely. I wish Apple would take that page down - it was broken (B even for Jaguar. (B (BWhat I'm wondering about is this: (B (B--- (BA quick trip to Jaguar$B!G(Bs Terminal showed me that this version didn$B!G(Bt (Bmake it into the default install: (B (B [cpu:~] user% perl -v (B This is perl, v5.8.1-RC3 built for darwin-thread-multi-2level (B--- (B (BI'm supposed to believe that little exchange was on a default Jaguar (Binstall? (B (B The latest Perl (5.8.4) comes with a readme.macosx file. Please, read (B and follow those instructions. Apple's instructions may be of interest (B to historians, but they have no relevance to the current Mac OS X. (B (BI note the original was contributed. Maybe it's time to contribute a (Bnew one. I wonder if I kept any notes when I did the parallel install (Bof 5.8.4 on this Jaguar box. (B (B-- (BJoel Rees (B Opinions are like armpits. (B We all have two, and they all smell, (B but we really don't want the other guy to get rid of his.
Re: Forking Signals
Now if we take that same simple program and either
don't define $SIG{'TERM'} or set it to 'DEFAULT' we
get END when the parent dies, but when we kill the
child cleanup isn't run (duh) but neither is END. Is
that standard behaviour? I would've thought it'd try
to do END if at all possible to clean up after itself.
Lessee, I think it's a kill 9 that can't be caught. And maybe kill 15,
but I've never played with that. Other signals can be caught, but I've
only done that in C, so my memory may be faded.
Re: Download images/movies
On 2004.8.22, at 04:07 AM, Chris Devers wrote:
On Sat, 21 Aug 2004, Mark Wheeler wrote:
I have a picture gallery I building for my family. When a movie or
picture is displayed, I want them to be able to save it. But... if I
just provide a link in the coding to the actual file, it will open up
in the browser window and be displayed. Is there a way to have
download, either automatically or by a Save As... dialog box, the
file rather then displaying it? I hope that was clear. :)
This is untested, but I'm guessing that you could write a simple CGI
script that takes the URL for an image as an argument -- maybe just
using $ENV{'HTTP_QUERY_STRING'} so that the url can be simple like --
http://site/images/fetch.pl?path/to/image/file.jpg
-- and then have your script find path/to/image/file.jpg and spool
it back to the client with a Content-type of
application/octet-stream instead of image/jpeg.
This can probably be done with about half a dozen lines of code, and
if the browser is well behaved -- that'll be the part that's a pain to
verify -- the alternate content type should force the right behavior.
Worked on a project where we wanted pdfs to automatically open in Adobe
Acrobat. Various things we tried worked with some browsers and not with
others, and I had to point out that some people would have their
browsers set to save everything to disk. Boss decided that anyone who
knew enough to set the browser to save everything to disk would be
assumed to be unlikely to be surprised at the results.
You should have less problems because the save-to-disk is what you're
going after, but some browsers in some settings will attempt an ascii
dump to the browser.
I have a script that builds a page that spreads a set of pictures out
in a table with links to several resolutions (no, not filtering the
resolution at run-time). I did not bother with setting the Content-type
in the download links because I knew two things -- those relatives who
would not know to right-click would also not know what to do with the
files when downloaded, and I was pretty sure the one person I wanted to
be able to download would not have to be told to right-click or
control-click.
Sorry to say, I won that bet with myself, so I didn't bother building
in the Content-type machinery which is most interesting in this thread.
If anyone's interested in what I did put together, I could post it.
--
Joel Rees
Complaining about systems that are incomplete misses the point.
In this world, a system can't be perfect and useful at the same
time.
Of course, there's no excuse for refusing to fix problems --
we'll never run out of problems.
Re: Download images/movies
Thinking twice, I'll be out of pocket for a while, so I'll just go ahead and post this just in case anyone is interested. http://reiisi.homedns.org/~joel/cs/shared_code/showpics.pl.text Contains some shift-jis, which can be stripped out with no ill effects. -- Joel Rees It's not the Here's a button, click it! attitude, It's Bill saying he has to be free to invent our technological future. (But I'm just as glad it's not Steve's company with the 95%, either.)
Re: Download images/movies
I think that's what I'm looking for. One question. What do you mean whitelist the filepaths. My only reference point is email. Whitelist for me means that email address on my whitelist always get through, even though the spam software might initially think it's spam. Can you clarify? If the script I posted was readable, you might have noticed that it accepts one parameter and sets the directories only if that parameter matches correctly. It looks like a waste, but it's one way of what he was calling whitelisting in a fairly strict way, but allowing the same script to be used on multiple sets of images. You do have to add a little code for each set of images, of course. That script needs some comments. -- Joel Rees Getting involved in the neighbor's family squabbles is dangerous. But if the abusive partner has a habit of shooting through his/her roof, the guy who lives upstairs is in a bit of a catch-22.
Re: Download images/movies
Just a few nosy comments --
html
head
titleUntitled Page/title
/head
body
a
href=javascript:window.location='cgi-bin/download.cgi?picname=Upload-
Background.gif'picture link/a
Not sure why you want to bother with javascript in there. ICBW, but I
don't think it buys you anything. And some of your family may decide to
turn javascript in their browsers off.
/body
/html
-
#!/usr/bin/perl -w
use strict;
use CGI ':standard';
I didn't notice that you had used anything from CGI in the script.
Might as well comment it out.
my $filename = param('picname');
Did you follow what was said about ../../ someodd with /etc at the end?
It's a good way to dump all sorts of things about your machine into
someone else's browser, including user names and ids, the entire
httpd.conf file, and so forth.
That's why I don't usually accept filenames in scripts. But if you do,
you need to check for / at the top or ../ anywhere, and balk if you get
those. It can get kind of tricky, since \/ is /.
my $path = /images/$filename;
For instance, somebody puts this in their browser:
http://your.domain.com/cgi-bin/download.cgi?picname=../etc/httpd/
httpd.conf
binmode STDOUT;
print Content-Disposition: attachment;filename=$filename\n;
print Content-Type: application/octet-stream\n\n;
If you _had_ been using CGI, the above two lines could have created
some subtle conflicts.
open (FILE, $path) || die(Can't open($filename): $!);
This is why you got the attempted download that stalled, of course.
That die statement won't do much useful. Well, if it were going out
STDOUT, it might have shown up as your downloaded file.
You'll want to look into using a logging file or the http version of
carp.
my $data = FILE;
close (FILE);
print $data;
exit;
--
Joel Rees
Nothing to say today
so I'll say nothing:
Nothing.
[OT]Re: Brand New Empty Mac
But I just thought I'd get the opinions of the list on the best way to set up such a brand-new machine -- do you partition your hard-drives? I usually do. But on a Mac, I have decided not to to be play too smart. I'll usually format a gig or two for classic and leave the rest to the boot drive. I have not satisfied myself that there is any advantage to a swap partition, and I won't mention why here. On the other hand, Apple may have decided to let fstab have some so in Panther, in which case a separate swap partition can help smooth the virtual memory system a bit. Some of your Mac OS X applications may throw a hissy if they are not on the boot partition, or if the users are not on the boot partition, so you lose most of anything you'd gain with separate /home or /usr. And the automount puts everything under /Volumes, anyway. However, if you plan to serve the web with that box, a separate partition for web stuff might give you warm fuzzies and maybe even some real protection. Just make sure you format any partition(s) for web as Unix File System, instead of HFS+. That way you should actually get the permissions bits to work right. I might also have another UFS partition for PostGreSQL and other such. I also keep a spare hugh partition if I can, for downloads and large images. If you have a bigfiles partition it should be HFS+, of course. Do you have the system on one partition and documents on another and so on? No, that just buys you heartache in Mac OS X, of present. Any issues around the installation of Perl and other things like C libraries that I should be thinking about? I like to have multiple users, one for doing serious work in, one, perhaps with limits on it, for surfing, and, of course. I also like to make two administrator accounts, just in case something goes haywire in one. Most people don't really need to bother with anything fancy, just let the install set up run. -- Joel Rees If God had meant for us to not tweak our source code, He'd've given us Microsoft.
Re: Download images/movies
Thanks for your input. In regards to filename, I'm assuming you are
talking about the filename passed within the HTML, right?
If I know what you meant be that, I'd be more able to say.
So I'll dodge and try explaining it this way: Any script that could
potentially be called by someone typing it's URL into their web browser
is subject to getting input you don't want. If you accept file names
directly in such scripts, there are several bad things that can happen:
An attacker can gain information you may not want him or her to get
about the structure of things in your site (ergo, by checking the
status bar). Worse, an attacker could pass a complex path giving access
to files you don't want him or her to be able to access. Another bad
thing is that an attacker could pass a huge string that contains lots
of escape characters to hide the complex path. (I regularly see zombies
posting 32k parameter strings to my home server.) If I tried to put one
of those strings into a limited space buffer (usually more of a problem
in C than in perl), it could blow up my program or allow a buffer
overrun attack.
I think what I will probably do is pass an ID number to the script and
then process it that way.
If the file name contains the ID number or name string, you'll still
need to clean the http parameters.
If you have a hash (not an array) that you use the ID to index into and
get the real file name, that is one good way to clean the parameter. An
if-ifels condition that compares the IDs to a (small) set of valid IDs
could be another way.
But whatever you use to map the IDs to file names, you'll want to check
for weirdness like escape characters and long strings, just to avoid
stressing your machine.
I will still check for ../ andywhere the passed ID, as well as /
at the beginning of the ID. You mentioned that V is /. Im afraid
you lost me there. Can you explain?
The backslash is an escape character and can be used to hide things.
For instance, ../ will go up one level in the directory, but so will
\.\.\/.
Have fun. I gotta go to work.
Thanks,
Mark
On Aug 25, 2004, at 8:16 AM, Joel Rees wrote:
Just a few nosy comments --
html
head
titleUntitled Page/title
/head
body
a
href=javascript:window.location='cgi-bin/
download.cgi?picname=Upload-Background.gif'picture link/a
Not sure why you want to bother with javascript in there. ICBW, but I
don't think it buys you anything. And some of your family may decide
to turn javascript in their browsers off.
/body
/html
-
#!/usr/bin/perl -w
use strict;
use CGI ':standard';
I didn't notice that you had used anything from CGI in the script.
Might as well comment it out.
my $filename = param('picname');
Did you follow what was said about ../../ someodd with /etc at the
end?
It's a good way to dump all sorts of things about your machine into
someone else's browser, including user names and ids, the entire
httpd.conf file, and so forth.
That's why I don't usually accept filenames in scripts. But if you
do, you need to check for / at the top or ../ anywhere, and balk if
you get those. It can get kind of tricky, since \/ is /.
my $path = /images/$filename;
For instance, somebody puts this in their browser:
http://your.domain.com/cgi-bin/download.cgi?picname=../etc/httpd/
httpd.conf
binmode STDOUT;
print Content-Disposition: attachment;filename=$filename\n;
print Content-Type: application/octet-stream\n\n;
If you _had_ been using CGI, the above two lines could have created
some subtle conflicts.
open (FILE, $path) || die(Can't open($filename): $!);
This is why you got the attempted download that stalled, of course.
That die statement won't do much useful. Well, if it were going out
STDOUT, it might have shown up as your downloaded file.
You'll want to look into using a logging file or the http version of
carp.
my $data = FILE;
close (FILE);
print $data;
exit;
--
Joel Rees
Nothing to say today
so I'll say nothing:
Nothing.
Re: Download images/movies
On 2004.8.26, at 04:45 AM, Bill Stephenson wrote:
On Aug 25, 2004, at 10:59 AM, Mark Wheeler wrote:
Hi Joel,
Thanks for your input. In regards to filename, I'm assuming you are
talking about the filename passed within the HTML, right? I think
what I will probably do is pass an ID number to the script and then
process it that way. I will still check for ../ andywhere the
passed ID, as well as / at the beginning of the ID. You mentioned
that V is /. Im afraid you lost me there. Can you explain?
I'm curious, I've seen the ../ thing mentioned many times over the
years but I've never successfully created a script that would open a
file that way. I use a Clean Name sub-routine (that I got from
Lincoln's CGI book) just to be safe on files I want to process or
return to a client;
sub clean_name {
unless ($selected_file =~/^[\w\._\-]+$/) {
That'll disallow some valid filename characters, particularly in Unix.
But that's okay if you never use it to validate filenames that have
those characters, and it side-steps a lot of picky details about trying
to find all the naughty stuff.
print STRONG$selected_file has naughty characters. Only ;
print alphanumerics are allowed. You can't use absolute
names./STRONG;
die Attempt to use naughty characters;
}
return $selected_file;
}
Still, I've tried scripts without it and they will never open a file
name input from a form like:
http://site.com/server.cgi?file=../../../../../../../etc/passwd
Maybe it's because I usually append the $file to a $path
How deeply nested is the path? would be one question.
or never input the right combo of ../ (path info) but I've never
seen it work. Can someone actually show me a cgi script example that
does this?
From a casual examination, I would guess that the sample Mark posted
should be one that would do this. (It occurs to me that he really might
probably prefer not to have anything he's pumping out to the web right
under root the way it looks like he's got his images directory.)
It seems to me that the file permissions for etc/passwd should
prevent this from working in the first place.
Well, open up a terminal window and look at the permissions:
% ls -la /etc
% ls -la /etc/passwd
--
Joel Rees
Re: Thunderbird
I'd like to take more time for this, but waiting doesn't produce more time this week. So I'll just toss out an idea -- You might be interested in this page, entitled Import Address Book records into to Thunderbird : http://www.macosxhints.com/article.php?story=20040905025741769 When I googled for thunderbird and address, trying to learn what thunderbird was, this was the first link that came up. Yeah, shortly after posting I did the google and found that page. Tried the script, it had some issues* but I tinkered with it until I got it to work. But even then, Firebird's import didn't seem to find anything in the file. I don't know about .vcf, but .csv is fairly easy to just look at with a text editor (formatting off, of course). The primary complications are for commas and new-lines buried in fields. Microsfot used quotes and made the whole thing a mess, but once you get your head around the mess it isn't that bad. (One of these days, we have to put ASCII behind us, but that's a topic for a rainy weekend or two.) (After you load the file there's a pop-up to link specific fields in the file to specific fields in the Thunderbird format. But neither .vcf not .csv seems to show up wih anything.) Since .csv *is* listed as a text-type it understands, I wonder if this might be a bug in the import abilities. I was hoping someone else here uses Firebird and may have dealt with the issue before. * SImpleText doesn't like the create new document stuff, so changed it to call TextEdit. And it craps out on some of the addresses for no reason i can discern, so I just commented that all out to produce null addresses. ~wren __ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail
Re: Thunderbird
On 2004.9.24, at 11:55 AM, wren argetlahm wrote: --- Chris Devers wrote: --- Joel Rees wrote: I don't know about .vcf, but .csv is fairly easy to just look at with a text editor (formatting off, of course). Yeah, they're both just text and (pretty) easily readable. The problem comes in that I don't want to stop using AddressBook and so I'm looking for a maintainable solution, where I can just hit a couple buttons or run a script rather than needing to manually enter anything. That's part of the reason i've been looking at FB's import function and Mac::Glue. I don't know for sure, but I'm thinking that FB doesn't offer any sort of scripting API (ala Mac::Glue or commandline commands) that'd let me enter the data programmatically if the Import function doesn't work. I'd love to be disproven however. Incidentally, the .vcf file generated by AB looks akin to your example but with a space between every charecter and two newlines instead of one. Is that normal, or might that be part of the reason that Firebird is having difficulty reading it? I would guess that would be the entire reason. Use the file open menu item in Text Edit and try loading the .vcf file generated by Address Book as a Unicode UTF-16 file. (You may need to customize the encoding list.) If you just double click or drag-and-drop, it will use the default encoding, which is probably UTF-8.
Re: Thunderbird
On 2004.9.24, at 11:34 AM, Chris Devers wrote: On Fri, 24 Sep 2004, Joel Rees wrote: I don't know about .vcf, but .csv is fairly easy to just look at with a text editor (formatting off, of course). VCF is (basically) an ascii format. You can encode binary data (e.g. photos) in it, but it's base64 encoded (just like email) so you can poke at it with a regular text editor. A typical entry might look something like this: BEGIN:VCARD VERSION:3.0 N:Meyer;Russ;;; FN:Russ Meyer EMAIL;type=INTERNET;type=HOME;type=pref:[EMAIL PROTECTED] item1.EMAIL;type=INTERNET:[EMAIL PROTECTED] item1.X-ABLabel:_$!Other!$_ TEL;type=HOME;type=pref:800 555.1212 item2.ADR;type=HOME;type=pref:;;42 Any Lane \n;Hollywood;CA;12345;United States item2.X-ABADR:us X-AIM;type=HOME;type=pref:rmvix END:VCARD Now that you mention it, I guess I have looked at those with a text editor. Etc. It's a little confusing, Not so much confusing as just got a lot of stuff in it. Looks like colons for the element labels and semicolons for the element delimiters. And I think I see a buried newline escaped with a backslash. Hmm. Who made this format up, anyway? My goodness, these things have got RFCs behind them: http://www.imc.org/pdi/ Surprised they don't mention any movement to convert these to XML. but it's mostly a regular format that isn't too hard to read or otherwise work with. Well, ... (One of these days, we have to put ASCII behind us, but that's a topic for a rainy weekend or two.) ??? Every tool has a role; ascii has lots and lots and lots of useful ones. Also roles that it's totally wrong for, but that doesn't mean that it makes sense to get rid of it altogether... Yeah, but it's time to move on. (I'm busy in my spare time trying to invent an encoding scheme that includes a variety of meta-punctuation, including meta-field separators. Of course, by this point, I'm duplicating effort by the Unicode consortium, to a certain extent.) -- Joel Rees
Re: [OT] Text Editor for OSX
On 2004.10.4, at 02:29 AM, Doug McNutt wrote: I'm not so sure about the OT designation. FORTH is on topic on a perl list? ;-) Apple's Macintosh Programmer's Workshop (MPW) is the best programming environment I have ever used. BBEdit worksheets are a start but are not nearly as flexible. emacs is another option but it still doesn't approach MPW with its window = file metaphor. MPW allows one to execute a shell command by selecting it and using the ENTER key. I've always wondered how much MPW was inspired by FORTH. Output from the command, which can be a named file or an open window, can be redirected to any other open window or to a file. ... Let me see. IIRC, trying to run perl as an interactive shell had its limits. But it should not be hard, I suppose, to feed a selection or line from a text editor to an instance of perl. I'm a little lazy right now. Was SubEthaEdit originally on open source project? (And did Wren notice BareBone's TextWrangler and decide that didn't go far enough?) One thought -- Wren, if you're going to go so far as to write YATE, I'd suggest your internal character encoding be a thirty-two bit encoding that uses the full thirty-two bits to allow you to keep track of input encoding on a character-by-character basis. While Unicode support is a must, I would not use it as an internal encoding because of the round-trip problems. But then I've only wren one text editor, and that was in FORTH, and not very comprehensive. -- The best programming tool is a soldering iron -- 8-O
Re: OT: what happened to my permissions?
On 2004.10.5, at 10:40 AM, John Horner wrote: Please forgive the OT nature, but I just know you will be able to help. I upgraded an old Mac I use as a server from 10.1 to 10.2. After the upgrade, the webserver Documents folder had all the wrong permissions. I had to log in via the terminal and CHMOD various things. All is working well now, but not for *new* files. Every new file I upload upload from where and how? has the permissions -rw-r- although the enclosing folder itself is drwxr-xr-x. I'm a bit confused about this, and the more I read about UMASK the more confused I get. Short version of question: how do I set the default permissions, permanently, recursively, for all new files uploaded to /Library/WebServer/Document/ ? Other questions: If I do a rebuild permissions, will it fix this? Will it over-ride various folders which are world-writable world writeable? whaddaya wanna doodatfer? so I have to go back and CHMOD again? Can I run rebuild permissions in Terminal or only from the GUI? And finally, if anyone's really annoyed by this being OT, where should I go to ask for this kind of help in future? http://discussions.info.apple.com/ Admitted, it doesn't exactly look obvious, but if you go looking starting here you can find a unix forum under both Panther and Jaguar. (It's a little better hidden under 10.2.) If you look closely, it's the same forum. If you have a (free) developer account already, you can use the same login name and password to post questions. Not having done what you did, I'm not going to take a stab at your other questions. But what does umask without any arguments tell you? Something like 037 or 015? -- Joel Rees even though much of what I do is not sensible it does make sense if you know why ...
Re: [OT] Black Screen of Death
I was hoping that the error might be found in one of the Console logs. It ought to be in one of the logs, yes. Sorry that I don't remember which. Check in /var/logs List by time, so you can tell which files were written most recently: ls -latT (IIRC. might want to man ls if that set of options doesn't work.) Some of the files are binary, but I'm pretty sure the most important ones are text. Oh, and run this command first: last That is not all, but it' all off the top of my head this AM. It's difficult to read the text that's scrolling on the screen because it's moving so darn fast, my old eyes aren't very good anymore, and I tend to get those pesky Optic Migraines when I stare at anything that strobes or has significant glare, ... -- Joel Rees Getting involved in the neighbor's family squabbles is dangerous. But if the abusive partner has a habit of shooting through his/her roof, the guy who lives upstairs is in a bit of a catch-22.
Re: Yet another Mac OSX and DBI question
One quick question. I am sat in a hotel room a long way from a braodband connection. this is a 300 M download. That's a lot on a dial up at hotel phone rates. Is there a less financially crippling way of just getting the essential files, or do I need the full update to XCode? I've never tried this, but the files have been named. Might be worth a try getting them from Apple's darwin source code, hand compiling them and copying them into place. I suspect you won't get the options (like dynamic library vs. static or object code, object output format, among others) correct the first try. I'd guess dynamic library and somthing other than elf, but I'd likely be wrong and incomplete. Anyone else have a guess to hazard? I'm sure you'd want to get the complete package installed once you get back where you have broadband. And maybe one of us should file a bug report or suggestion that Apple put up a download of just the relevant files? -- Joel Rees It's not the Here's a button, click it! attitude, It's Bill saying he has to be free to invent our technological future. (But I'm just as glad it's not Steve's company with the 95%, either.)
Re: Installing new(er) perl on Jaguar
Can someone disambiguate this article for me please? I'll try. I think you'll recall that this has come up before, so you may want to take a look in the archives. http://developer.apple.com/internet/opensource/perl.html Consider this article to be old enough to be reference only, by the way. I'm not up to a point-counterpoint today, I guess, so I'll just snip the rest of this and wing it. You can have more than one version of perl installed. In fact, it is generally recommended to do so. Most Unix systems make heavy use of the perl that is installed with the system, so if you upgrade perl underneath the system you risk confusing the system. And if you upgrade from 5.6 to 5.8, you'll have to make sure all the binary modules the system has installed (and all the binary modules you installed) are re-compiled. It's a lot easier to manage (and therefore a bit more secure) if you don't have to worry about effects on the system when you upgrade and add modules to the copy of perl you're using for non-system stuff. 5.8 is recommended for a number of reasons. (More complete Unicode support is a big reason in my case.) So, when the installer asks where to install it, tell it to install in /usr/local (or wherever you keep the non-system stuff you install for all users). You will have to edit the first line of _your_ scripts (not the system's) to #! /usr/local/bin/perl and remember to invoke the right CPAN, and such things, which can be a little confusing. There is a way to get a list, but I don't remember it, and javadoc seems to be interfering with my memory of perldoc right now. Maybe someone can help me here? -- Joel Rees Nothing to say today so I'll say nothing: Nothing.
Re: Installing new(er) perl on Jaguar
There is a way to get a list, but I don't remember it, and javadoc seems to be interfering with my memory of perldoc right now. Maybe someone can help me here? Umm... a list of what? Perls? whereis perl will do that. If you're in doubt about which of the installed Perls appear first in your PATH, which perl will tell you that. List of installed modules. I must be distracted. -- Joel Rees Nothing to say today so I'll say nothing: Nothing.
Re: Installing new(er) perl on Jaguar
Instead, install perl in /usr/local/, and make sure that that comes first in your path. That's fine for scripts that you run from a shell, with perl scriptname.pl. Do you run many of your scripts that way? I certainly don't. Most of my scripts depend on the first #! line to choose a Perl. Actually, there is one more reason not to put /usr/local/bin in your path ahead of /bin. It's a potential security weak spot. (Say some package you test out puts a program of some sort named ps in /usr/local/bin.) -- Joel Rees even though much of what I do is not sensible it does make sense if you know why ...
Re: MySQL
Paul DuBois's books on MySQL have been pretty good for me. I got a lot out of the one called, I think, Perl and MySQL for the Web. It's been over a year, but the MySQL lists seemed to be a lot of help: http://marc.theaimsgroup.com/?l=mysqlr=1w=2 (These days I seem to be messing around with PostGreSQL and xindice, mostly.)
Re: catnip.local (redirect)
Hmm. All this made me look at how I have this box set up and I discover that what I was about to tell you (and what I've done to a box I'm borrowing from work) was wrong. I'lll try to reconstruct things. I hope I make sense. My spouse has, at her workplace, a Mac OS X machine with web sharing turned on. This machine is, therefore, reachable on the internal company LAN as either http://catnip.local or http://catnip.company.com If my memory is right, I used to get that back in the days of 10.0 when the link went down or the ethernet cable came slightly loose or whatnot. /etc/hosts was only referenced during single user mode back then or something. I can't remember if using the machines domain in netinfo was the cause or the cure. But if you look under /machines in the Applications/Utilities/netinfo GUI widget, you'll notice that there are two entries that look very suspicious. At one time I had added an entry under /machines for the name of this box. I duplicated the localhost entry and edited it in a way that seemed appropriate. Sometime between 10.0 and 10.2.8, Apple fully restored the functionality of /etc/hosts, so I presently have a line something like 10.2.40.49 reiisi reiisi.homedns.org in /etc/hosts, instead. When she works from home, she accesses the company network via VPN. The machine is still accessible as http://catnip.company.com. Unfortunately, many of the links automatically convert too URLs beginning catnip.local. Via VPN (the way she does it), there is no catnip.local. I was going to mumble something here, but I need to hit the hay. I'm not sure I'm making any sense anymore anyway. Does anyone know where this redirection to catnip.local is stored and whether (how) she can make it stop? -r -- email: [EMAIL PROTECTED]; phone: +1 650-873-7841 http://www.cfcl.com- Canta Forda Computer Laboratory http://www.cfcl.com/Meta - The FreeBSD Browser, Meta Project, etc. http://www.ptf.com/dossier - Prime Time Freeware's DOSSIER series
[OT] Re: catnip.local (redirect)
My spouse has, at her workplace, a Mac OS X machine with web sharing turned on. This machine is, therefore, reachable on the internal company LAN as either http://catnip.local or http://catnip.company.com [] other places to look at for grins and giggles -- man hostname man domainname cat /etc/hostconfig (/etc/hostconfig is what actually tells the box to name itself automatically, which is causing the use of the local domain, but don't jump to the assumption that you should therefore change it from automatic to catnip.company.com. I don't remember why I don't do that, but I did think I had a good reason. Looking up etc/hostconfig+mac os x at your favorite search engine might provide some clues.) The biggest failing with mac os x is its biggest strength -- too much indirection and delegation. It's often really difficult to figure out where the buck stops. But it does just work for most people, and they do seem to be making headway at sorting things out so that mere mortal sysads can figure them out. Have you asked/searched on Apple's boards and mailing lists? I'm pretty the topic has floated there in the past. -- Joel Rees Nothing to say today so I'll say nothing: Nothing.
Re: installation weirdness with Mac::Glue
I thought I'd play around with Mac::Glue, so I fired up the CPAN shell to install it. The installation went, in part, like this: [...] END failed--call queue aborted, DATA line 1. *** malloc: vm_allocate(size=268435456) failed (error code=3) *** malloc[10104]: error: Can't allocate region Out of memory! [...] Kind of an odd-ball question, but you aren't short of disk space are you? what's the output of df?
Re: TextWrangler
On 2005.1.21, at 11:38 PM, William Ross wrote: On 21 Jan 2005, at 12:35, Jeff Lowrey wrote: At 07:10 AM 1/21/2005, Ken Williams wrote: See http://www.barebones.com/products/bbedit/threeway.shtml for a 3-way comparison between BBE, TW, and BBELite. While we're playing around with Editor Wars... there's no need for that sort of language... Boy,, there's nothing like a good old-fashioned editor war! But this one doesn't seem to have much punch to it. More like a dust devil than a cyclone. Visual Slick Edit v9 from http://www.slickedit.com/mac/ will run on OS X. aaargh! that was horrible. 50MB download (50MB! Quark 3.3! one floppy! etc!), only runs under X11 (so you can't even paste in the extremely long temporary license key unless you save it somewhere and open it in an xterm). its interface reminds me of Windows 1.0 and it keeps offering to bind my java. actually, to come back to the topic, it looks like it might be a pretty good IDE for compiled and linked projects but it doesn't seem to have anything to do with perl, nor much with os x either. I have to admit, I'm more likely to download that now than I would have been just from Jeff's comments. An rgh! has to rate something. Won't be until I can afford panther (or tiger?) and some more RAM, though. That, or until I can find a proper download for the libraries for the old X11 beta. Couldn't find them last I looked. Oh, and thanks for letting us know about TW going free, Chris. I've always liked BBEdit, myself, almost as much as the CodeWarrior editor. (Talk about twisted tastes.) -- Joel Rees Opinions are like armpits. We all have two, and they all smell, but we really don't want the other guy to get rid of his.
Re: Need to reinstall unix head utility any ideas?
Speaking of the case insensitivity issue, is anyone here experimenting with the case sensitive HFSx volume format?
Re: Suggestions to better ship Perl on OS X.
Yes, you're right ... I suppose that Find in the Finder doesn't find it because /System is excluded from Find, unless you specifically choose it? sudo find / -name perlfunc*
Re: Variables in external file
On 2005.2.19, at 01:39 AM, Mark Wheeler wrote:
Hi,
Just a quick question. Is it possible to have a bunch of variables in
a separate file and then require that file in the script file? Let me
give you and example.
--
Script file
--
#!/usr/local/bin/perl -w
use strict;
require variables.conf
print Content-type: text/html\n\n;
foreach (@list) {
print;
}
exit;
-
variables.conf
-
my @list;
I remember how my works in blocks, but I'm having trouble remembering
how my works in files.
And I'm having trouble remembering what to do when you actually _want_
a declaration to have global effect. Where're my books?
$list[0] = '1';
$list[1] = '2';
$list[2] = '3';
$list[3] = '4';
$list[4] = '5';
1;
-
When I try the above script, I get an error - Global variable @list
needs to be defined. What am I doing wrong, or is this even possible?
Thanks,
Mark
Re: could not build a module
I will install the Xcode and see how it goes. You know where to find it?
Re: could not build a module
I will install the Xcode and see how it goes. You know where to find it? You need to become an Apple Developer connection Member thet ypu can dowload a lot of development tools. The Apple web site for the ADC is https://connect.apple.com And if you don't have broadband, you can almost always find it somewhere in the OS install. Of course, it may not be the absolute latest, but it will be there. I was just wondering if Ted knew.
Re: What Perl editor do you recommend?
On 2005.3.3, at 07:15 AM, John Delacour wrote: At 9:45 pm + 2/3/05, Phil Dobbin wrote: I'm thinking that if he's not comfortable with pico maybe emacs is not the best idea... I'd love to hear a convincing explanation from someone why anyone would use such tools in preference to TextWrangler, BBEdit or Affrus. I can imagine they'd make it a chore to write code in us-ascii and either a nightmare or an impossibility to deal with non-ascii, but maybe that's because I'm just an unreformed Mac user :-) Two points, or maybe three -- One, vim can be customized to handle mult-byte characters. Emacs, can, too, from what I hear. I'm personally not satisfied with the results, but it does work, even if it's rather clumsy. I have the impression that pico can also be customized, since there are a number of Japanese people who use it. The other, vi is, as has been mentioned, almost always there, and it's much easier to use than ed. Also, vi inherits a lot of powerful macro processing capabilities from ex/ed that are somewhat arcane, but still useable. If you're comfortable with vi and can keep track of the arcane syntax, it pretty much lets you do everything you can do in mpw. I personally use whatever's handy, but when I edit the files under /etc, I usually don't really want to waste the time fiddling with permissions and such. And if I have to type the file path in by hand anyway, I might as well open up a terminal and use vi.
Re: What Perl editor do you recommend?
Apologies for fanning the fires, but this hits kind of close to home ... On 2005.3.3, at 07:39 AM, David Cantrell wrote: [...] and either a nightmare or an impossibility to deal with non-ascii, but maybe that's because I'm just an unreformed Mac user :-) If you put non-ASCII in your code you're doing something wrong. Language-specific stuff - including English - belongs in a seperate resource file if you care about internationalisation. Resources have to be edited with something, and it is often useful to be able to use REs on them. Also, making systems and apps universal is trying to solve a problem that shouldn't be solved, even if the tools are useful. Even if the core engines of, say, a medical system can be universal, there are huge pieces of functionality that should _not_ be so. If you try to run a Japanese clinic the way an American hospital or clinic is run, you're not going to help very many patients. Likely to scare a number of them, in fact. The guys that build the local stuff should work in their own language as much as possible, and that includes not just comments, but, if possible, identifiers, syntax, and grammar. Otherwise, they tend less to understand what they are doing and more to think it's all just a mathematical game. And they tend not to really understand re-factoring if it doesn't work on symbols in their own language. Right now, comments are about all that can be dependably worked with in non-Latin characters, but even those, it's useful to have a full and accessible set of RE-type tools to work with.
Re: First CGI Setup
Should not try to give people advice at two in the morning. I said I've set each user's web-facing directories and files to owned by user, but group is the apache user. The directories that serve the domain root are owned by the apache user. Directory permissions are read/write/search (rwx) for owner, read/search (r-x) for group, no permissions (---) for others. And I failed to mention the permissions on the files. Putting the files in the apache user group allows you to remove the read (static html) and execute (cgi) permissions for others if you want, which shores things up a bit. I don't remember if Apple gives you an apache group, but that's easy enough to add with netinfo if they don't. And I said I personally am a bit of a bigot about file extensions. I don't use them except for perl because I don't have to, and because I prefer to have all my cgi in one place. But I should have said I don't use extensions with perl, only with php, because that's the way php is built. (But I don't use php at home, which is kind of ironic. :-/) My reason for confining executables to a specific set of directories is somewhat related to my reason for not using HFS on web-facing partitions. It allows me a greater level of confidence that I know which files the server is going to expose to the web and how. Less to keep track of. Less chance of mistakenly treating a non-cgi file as an executable, and less chance of spilling source code through some slip in the configuration. (And how's that for trying to keep this on-topic?)
Re: dealing with UTF8 text
On 2005.3.31, at 10:18 AM, Avi Rappoport wrote: Hi old friends (and new), I'm quite enjoying getting back to scripting, and like Perl a lot, especially with Affrus. While I'm probably inefficient, it's nice to have a language actually designed for text processing (search engine logs, in my case). However, I've got some Unicode issues and that seems to be platform-specific, so thought I'd ask here. Have you done perldoc perlunicode and used that as a lullaby for several afternoon naps in a row? Used the stuff referred there for a few more afternoon naps? (perldoc always seems to put me to sleep, but if I don't open it up and stare at it in spite of the soporific effect, nothing seeps in at all.) Have you gone to unicode.org and scanned what they have to offer relevant to the character ranges (languages) you need to be parsing? Have you looked up the traditional encodings for your language/locale, particularly the microsoft (bleaugh) code pages? (Google or your other favorite search engines can help.) I've done enough research to know that I should avoid hardcoded counting with positions and use the perl functions which will automatically handle utf8 characters properly. That's cool. I'm pretty sure I'm reading in utf8 and comparisons seem to work. Comparisons can seem to work when the encoding is all off, as long as the input is being munged the same way in all inputs. That doesn't mean it will work for all valid input, however. What I can't do is generate readable cross-platform output to show my clients. Nothing necessarily surprising there. It takes quite a bit of tuning your brain to get the code right. (I speak from experience with Japanese encodings. ;) Even opening the output in BBEdit as UTF8 doesn't convert the codes into properly rendered extended characters, and by the time it gets into Excel on their Windows workstation, all hope is pretty much gone. BBEdit, IIRC, handles some of the traditional encodings fairly well. (Does quite well with the Japanese encodings, at any rate.) So if you are opening UTF-8 and it isn't looking right, your output is probably not UTF-8. If you check the options in the file opening dialogs, you may find a way to convert from the actual encoding you're writing out. And/or you should be able to adjust your perl, but we can't help you with that unless we see some code and have some idea what encoding/language/locale you're trying to write out. Incidentally, in many of the traditional encodings, the basic Latin will be in the some positions (same code points) as UTF-8 Unicode basic Latin. The stuff that looks like HTML entities is fine when viewed in a browser: #1575;#1604;#1578;#1593;#1575;#1585;#1601; s#305;emens And if necessary, I can deliver in HTML. But my logs have characters like this in them: (from BBEdit as UTF8:) atualizao carreo (from BBEdit as Mac Roman) atualizao torunn tmmervold lschen I can tell they mean something, but I can't figure out how to make them readable. Help? TIA, Avi
Re: Tiger version
As far as perl goes, I always install a parallel perl for my dev stuff, so I don't have to mess with the system install. If you are anxious for Tiger, you can pay USD 500 for the privilege of testing pre-release versions. If you do lot of development for the Mac, it's worth the price. I've done it once, just haven't been able to get involved enough in Mac development to pay for the next one. Found myself wondering what to do with all the CDs, also found myself wishing I could afford more hardware for running the pre-release software on.
Re: keychain
On 2005.4.21, at 09:15 PM, Ken Williams wrote: Hi Joseph, In my address book, I've got several of those too. I believe they're certificates from people who have signed their messages. If you don't know them, they're probably on a list you're on. That's definitely a possibility. It bugs me that Apple lumps things together like this because there's another possibility as well. If spam comes with a certificate, what do you suppose might happen? It's a bit of a pain, but I would prefer the keychain, in the default settings, prompted the user before storing any certificates. I'd also like to be able to set it to prompt before storing addresses, as well, but that's just something I can live with. When it stores certificates I don't know anything about, the chain of trust tends to have even less to do with me. Some things simply can't be mechanized. -Ken On Apr 19, 2005, at 11:31 PM, Joseph Alotta wrote: Hi Everyone, I looked at man security, just for kicks and I dumped the keychains. I was suprised to find email addresses for people who I do not know. I am a single user powerbook with dial up 56k access. Is this normal to have email keychain data for people I do not know? I could post those emails, but in case they're legitimate, I don't want them to get spammed. Any suggestions? Joe. On Apr 19, 2005, at 7:47 PM, Ken Williams wrote: Yeah, check out the 'security' command-line program. I use it in conjunction with Module::Release so that I don't have to type my PAUSE password every time I upload something to CPAN - it just fetches the password from my keychain. -Ken On Apr 19, 2005, at 5:01 PM, Larry Landrum wrote: I need to authenticate users in a perl CGI and was hoping to use the Keychain but can't find a perl way to do that. Has anybody done that before?
Macperl list false advertising?
Anyone know why http://www.perl.org/community.html describes the macperl list as Mac Perl - OS 7-9 and X discussion
Re: Macperl list false advertising?
On 2005.5.3, at 11:04 AM, Marc Manthey wrote: On May 3, 2005, at 2:17 AM, Joel Rees wrote: Mac Perl - OS 7-9 and X discussion joel, have you heard from apple ? its a small company from cupertino that build some cool mahines. They started from system 7.0 to 10.4 and there was a major change from 9 to osx;) Thank you, marc. I do appreciate it when people let me know I have assumed something is more obvious than it is: http://www.perl.org/community.html == ... a href=http://lists.perl.org/showlist.cgi?name=macperl;Mac Perl/a - OS 7-9 and X discussion ... == The macperl Mailing List Name:macperl Summary:Main discussion list for MacPerl, the port of perl on Mac OS (Classic) The reason this comes up is a recent post to the macperl-forum and macperl-modules lists looking for some help compiling mp3::info, or for someone willing to share a pre-compiled module. I may be wrong, but I'm pretty sure he wanted this list instead of that. -- Joel Rees even though much of what I do is not sensible it does make sense if you know why ...
Re: Installing WebService::GoogleHack
On 2005.5.18, at 09:53 PM, Lola Lee wrote: [...] Now, $! . . . what does this do? I looked in perldebtut and it says that ! means, redo a previous command, but what is the purpose of $? And, where should I be putting this in, again? Just so this doesn't get lost in the wash, $! is a special variable. (Has nothing to do with the ! command explained in perldebtut.) The contents of the special variable $! is the text version of the error message of the last error which occurred. There's a whole swarm of these special variables that use the $ sign with punctuation, including $_ . There are even some that are hashes or arrays, rather than scalars, and begin with % or @ instead of $. For more information, type perldoc perlvar at the command line.
Re: CamelBones on Intel? Maybe not.
I know what you mean, Sherm. Wish I could send you something to push into the iNTEL Mac world with, but I'm in the same position as you. Hope you can find a place that can see the value in understanding perl from the inside. If Perl 6 moves ahead, perl might go into the embedded world the way java hasn't yet really gone. For me, the computer industry just lost its last little bit of shine. I'm looking for a new career. Any general purpose computers I buy will run AMD since I doubt I'll be able to afford PPC hardware, and I'll be scratching Mac OS X from this old iBook this weekend. Not sure if I'll load Linux or openBSD on it, since it's my server. Jobs is insane. -- Joel Rees Nothing to say today so I'll say nothing: Nothing.
Re: CamelBones on Intel? Maybe not.
On 2005.6.7, at 11:13 PM, Robert wrote: Wiggins d'Anconia [EMAIL PROTECTED] wrote in message news:[EMAIL PROTECTED] Ian Ragsdale wrote: On Jun 6, 2005, at 5:18 PM, Joel Rees wrote: Jobs is insane. I'm not so sure about that. IBM seems unwilling or unable to produce mobile G5s, which is a market that Apple considers very important. They also are 2 years behind schedule on 3.0Ghz G5s, and appear to be focusing on video game processors instead of desktop and mobile processors. Apple might be OK in a speed comparison right now (on desktops, they are clearly losing in laptop comparisons), but how about in two years? Perhaps IBM has told Apple that they won't attempt a laptop chip, since the volume is way higher for video game consoles? What should Apple do? They should have released Mac OS X for Intel as soon as they had it ready. Why wait? It seems Apple is too caught up in their own keynotes to understand volume sales. One thing M$ was definitely *always* better at. IBM will probably laugh this one to the bank, not exactly going to put a dent in that $99 billion in revenue... Because it wasn't ready Five years and it still isn't ready? That's exactly why they shouldn't have kept it hidden in the lab if they were going to be doing it. and obviously after watching the keynote they are still working on some things. They are trying (and it looks good so far) to make the transition as painless as possible. I think it is a good move. If they were just saying, okay, we have had so many people begging for Mac OS X on iNTEL, we're going to give it to them and charge them double for running it on non-Apple hardware, that would be a good move. Moving everything to the monoculture is not a good move. Personally, it looks like it will be a bit painful for a few years, but a far better move in the long run. Unless they become just another cheap clone maker with a pretty software interface. (Did I hear someone say Sun?) Apple is not Sun in any sane comparison. You think? Ian http://danconia.org
Re: CGI script running as a given user?
On 2005.6.7, at 11:51 PM, Rich Morin wrote: I've got a Perl CGI script that acts as a system browser. For example, it can look at files and directories and say interesting things about them. This works fine, as long as the files are world-readable, but fails (because Apache runs as 'www') as soon as the user wanders into private areas. One answer to this is to launch a small-footprint web server that runs as the current user. The CGI would run under that server and all would be nifty and cool (well, not really, but OK :-). I'm wondering if I've overlooked a way to get Personal Web Sharing (aka Apache) to handle this for me. Something like have the user authenticate via https, then launch a given CGI script with that user's uid. There's an apache module that does exactly that. I think it's called suexec or something like that. But you want to read the documentation carefully, because it has a lot of security issues that you have to understand.
