And now that I've said that, I'd personally recommend you do not learn perl 5. Barring a couple highly targeted bouts of insanity, Perl 6 is going to make it look like TCL.

First of all, the project is nothing nerfarious. I kept the first note short to save y'all time.

I inherited the code from someone else who is not so motivated to help me understand it. So, my main job is to just get it up and running and to understand it is further down the list. This would seem to preclude going to Perl 6. I did notice a .NET extension of perl out there. I have not gotten to the point of looking into that. But, I have other code coming to evaluated which is .NET and so this intrigued me. Any comments on that (if you know what it is).

The code I got is a web crawler (but only with permisson!). As I mentioned, I am a long time C/C++ programmer and I have seen these kinds of things before. I have just mostly done them on Windows and not Linux or other flavors of unix, so some of the configuration stuff has got me bogged down (although I know unix shell OK). The "documentation" says it's using

• CGI
• CGI::Cookie, (for session tracking)
• URI
• DBI
• POSIX (for strftime)

And that is about the extent of the docs :-). I seem to have gotten all the libs installed. I found inside and started to figure that out. I am putting this all on my Windows XP machine (where I have been using .NET) and IIS. This might be a problem. Aside from the biases you might hold near and dear, perl doesn't seem to like Win32 too much, me thinks. I think i may be better off with Apache, for example.

I have been having trouble now with some missing modules which seem to be installed but are not found (more on that later). I found DBI on CPAN (MYSQL Drivers) but it wanted ppm to install and I cannot seem to get ppm to work. Ugh. Installs and tests etc. seem to go OK for the most part, although its hard to decide if a "skipped test" is critical. I must admit to having given up rather easily on this problem and decided to go back to the books since I have no clue as to what the monster is I have started to create, which has caused me to seek advice.

So, that seems to be all the details worth posting. I am now getting "insert user failed ($dbh doesn't appear to be defined)" on one of the apps. which appears to be related to my poor MySQL driver installation if you care to comment. Inside doesn't seem to show it but the CPAN download seems to indicate it's in there. But, I haven't gotten to the point of quite figuring that all out. I am trying to understand the architecture of the whole thing now since it's not going to just work with one pull on the starter cord :-)

One observation: it seems that there are dozens (lots of, many, more than one) of implementations out there (in CPAN, at least). This seems like a good thing but is rather daunting at first. What other tools do you recommend for manging experimenting with all these things? (other than inside which I already found).

Thanks much, cheers

Hello Again,

Well it certainly sounds as if you have taken on quite a task, but one which I am confident that you'll be able to manage from the sounds of things. Perl, while quite large architecturally, is easier to learn because it is intuitive; much more so than C(++), so you'll quickly be able to comprehend its intricacies.

OK, after that ego booster, let me say that I would IMO move to a *nix based box for development. Apache/unix are what Perl was originally intended for, and sits quite at home in this platform. If moving to linux is out of the question, you might find it easier (and more secure, etc, etc...) to get the win32 version of Apache, to replace crusty IIS.

If MYSQL is bogging you down (I'm so impressed with MYSQL, but it is quite bloated and generally best suited for applications where numerous simulatenous queries are performed), you might consider reimplementing the built in functionality of Berkely DBASE access in Perl (not sure how this works on a win32 box, if at all). If you have the Llama, Berkely DBASE is covered in one of the last chapters, and in only a few pages can get you on your feet using this aspect. I use Berkely almost exclusively, because my databases are generally large, but not accessed simultaneously. Also, I don't really need all the inherent permission capabilities, etc, that MYSQL has.

I think everyone has difficulties now and again installing modules from CPAN; with such a disparate number of extension Perl can take on, it's easy to find yourself in dependency hell. I would recommend <plug> installing Slackware Linux </plug> and building Perl from source. Maybe even create to parallel installs, one for mucking about with new/untested modules, and a stable/quasi-vanilla one that you can use for important/main development. That's what I do at least.

That's long and my thoughts are rambling, but I hope it gets you on track.

------------------------------------
"There is more in heaven and earth
than is dreamt of in your philosophy"

Thanks for the encouragement. One of the advantages of experience is having a few buddies in the biz. One of them offered the use of a linux system and so I have moved everything over there. I learned about "apt" which I wish I had known about a few days ago. He has helped me make some great progress.

This is a great resource, too, and I will definetly check in from time to time. It's really quite amazing, I think, that open source works so well.