Re^2: install fails for Inline::C (hide errors)

If you learn how to do the rather simple steps to install a module without using all of the automation, then it becomes trivial to run t/01syntax.t directly so you can see the error message.

When you say "without all of the automation", do you mean install modules manually without using CPAN at all? If I do that, will I ever be able to update manually-installed modules with CPAN in the future, or will I be forced to always manually update them?

Comment on Re^2: install fails for Inline::C (hide errors)

Replies are listed 'Best First'.
Re^3: install fails for Inline::C (hide errors) by tye (Sage) on Mar 15, 2013 at 14:26 UTC
CPAN is a network of copies of a computer archive. I usually use CPAN when installing Perl modules. I rarely use CPAN.pm nor its bin/cpan wrapper. "Automation", of course, refers to modules like CPAN.pm (bad name), CPANPLUS (worse name), App::cpanminus, and probably others and their command-line components. I'm sure none of these automated wrappers around CPAN are dependent on their own, separate tracking of what they have installed. You can install one module "by hand" (usually via 'make') and another module via cpanminus and then CPANPLUS will see both of them there just fine and be able to tell what version they are and be able to upgrade them or determine that they don't need to be installed or upgraded for some other module that you are installing that depends on them. It, of course, gets rather tedious to install a module "by hand" if that module requires the installation of a huge number of other modules. But my experience says that it is usually wise to avoid those types of modules anyway. So it is only very rarely that I have much motivation to try automation. But I'm not trying to tell anybody to avoid automation as much as I do. I just encourage you to install some modules by hand. - tye	[reply]

Replies are listed 'Best First'.

Re^3: install fails for Inline::C (hide errors)
by tye (Sage) on Mar 15, 2013 at 14:26 UTC

CPAN is a network of copies of a computer archive. I usually use CPAN when installing Perl modules. I rarely use CPAN.pm nor its bin/cpan wrapper. "Automation", of course, refers to modules like CPAN.pm (bad name), CPANPLUS (worse name), App::cpanminus, and probably others and their command-line components.

I'm sure none of these automated wrappers around CPAN are dependent on their own, separate tracking of what they have installed. You can install one module "by hand" (usually via 'make') and another module via cpanminus and then CPANPLUS will see both of them there just fine and be able to tell what version they are and be able to upgrade them or determine that they don't need to be installed or upgraded for some other module that you are installing that depends on them.

It, of course, gets rather tedious to install a module "by hand" if that module requires the installation of a huge number of other modules. But my experience says that it is usually wise to avoid those types of modules anyway. So it is only very rarely that I have much motivation to try automation.

But I'm not trying to tell anybody to avoid automation as much as I do. I just encourage you to install some modules by hand.

- tye

[reply]