Zasinger wrote: > Hello beginners, > Welcome...
> I'm a bit lost: I just installed perl yesterday and so far > it seems to have massive and appealing potential, I can > hardly belive what I am seeing. > > I'm very, very rusty in any sort of coding having been away > from it for over 15 years. I've been working at learning all > about 3d model making in that time. > > I want to use Blitz3D to display some 3d representations of > weather. It can't get the data I seek as far as I can tell, > but perl can it seems, and from what I've seen in 48 hours > it's very good at it indeed. > > I'm now stuck: > > I've so far got perl to use Net::FTP > Ok. Good choice of modules it is the defacto standard for doing FTP from Perl. I assume you have read the documentation for the module, available at: http://search.cpan.org/~gbarr/libnet-1.19/Net/FTP.pm > I can get perl to log in to weather.noaa.gov and drill down > to /data/observations/metar/decoded > Nice start. > I can get single files from there if I know their names, but > I'd like to write a bit of code (which I am calling > freshen.plx at this point) > > The idea is that it would freshen my local copy of that > directory, but only for files starting with "EG" for > example. > > I'm thinking I need some way to do the equivalent of mget (I > think that is the regular ftp type command) and I'd like to > mget "EG*.TXT" as far as I can figure out. > To my knowledge 'mget' is usually just a command implemented by the client and has no true FTP relation. Which is likely why you need to emulate it directly with your own code. > I've no idea how I'd do the wild card bit, so far it hasn't > worked out as I'd hoped it might. > > It seems that mget isn't implemented so I next thought that > a directory listing of that directory as a file might be a > good way to try next. > > How might I do that? > > I guess I could then pick lines off that file one line at a > time to get these files one at a time? Although I'm not at > all clear how I'd manage that bit just yet! ;O) > This sounds like a pretty good approach, so we need a directory listing and way to loop over it checking filenames.... from the Net::FTP docs we learn that the way to get a directory listing is, my @file_list = $ftp->ls; Now we need to loop over those checking for needed files, foreach my $file (@file_list) { if ($file =~ /^EG.*\.TXT$/) { # found an EG file, get it } } In the above example I use a Regular Expression to test each filename to see if the filename starts with 'EG' followed by zero or more characters, and ending with '.TXT'. For more reading on regular expressions, perldoc perlretut perldoc perlre That is the verbose way to do it. Some might suggest, foreach my $file (grep /^EG.*\.TXT$/, $ftp->ls) { # found an EG file, get it } 'grep' is a built-in that allows us to loop over a list pulling out those items we want. perldoc -f grep Based on these examples and the docs I leave it as an exercise to make this more efficient by pulling only those files that need to be updated. The 'dir' method, or some combination of 'mdtm' and/or 'size' should help, along with 'stat' on the local side. perldoc -f stat > If anyone can help me figure this out I'd be really glad. > > I've been working through some examples in "Beginning Perl" > and it's fairly slow going and rather dry reading in places > and some of it is a bit out of date too it would seem. I > think this is as a result of me having .8 and it being > written with .6 as the latest and greatest at the time, I > get the feeling some stuff has changed to this books > detriment perhaps. Mostly by scanning through some remarks > in the latest release notes (I think that was where I saw > them). I've no idea yet if it's trivial to patch up older > code so it would work once again. > Haven't read it. If you find it unsatisfactory I (and most others here) would recommend "Learning Perl, 4th Edition" from O'Reilly (aka "The Llama"). The 4th is a brand new release. There should be very few things in a beginner's book that would have changed from 5.6 to 5.8 (really *any* release in the 5 series). > Also I got that impatience thing going on too as I've waited > a couple of years to get at this thing, I'd started to feel > it was never going to happen but then perl happened to me, > making my weekend into a minor bedlam!! ;O) > The archives of this list are also an excellent resource, be sure to Google them before asking a question, there is a *good* chance it will have already been answered (more than once). Good luck, welcome to the world of Perl, http://danconia.org -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] <http://learn.perl.org/> <http://learn.perl.org/first-response>
