[Bioperl-l] ->add_tag_value()

[Hilmar Lapp]

On Aug 26, 2005, at 1:08 PM, Andrew Stewart wrote:

> Do many of you use bioperl-live as your primary (or exclusive) BioPerl 
> distribution, or do you keep a stable version as well?
>

I used to live off of bioperl-live and update regularly but I stopped 
updating about a year ago, so technically for production I'm on 
something close to 1.4.1.

I do keep a cvs head version too, for developing / fixing.

> I'd like to use bioperl-live for instances such as this (see message 
> history), but not necessarily for when I'm developing scripts that are 
> going to be used by others in my lab who do not necessarily have 
> bioperl-live installed.
>
> What I'm thinking is that I should maybe install a copy of 
> bioperl-live somewhere in my personal space, and then 'use' it in 
> certain scripts when needed.

What I do is I don't install Bioperl in order to avoid any precedence 
order mistakes for library paths. Since there is no compiled code 
(unless you also use bioperl-ext) you can just point PERL5LIB at the 
root of the Bioperl installation you want to work with, and if there is 
none in the standard @INC you can be sure which modules will be loaded.

>  I just have a few questions (these are probably more 'perl' questions 
> than 'bio-perl' questions)...
>
> 1. Once I obtain bioperl-live via 'cvs -d :pserver etc...', do I need 
> to actually go through the install routine or can I just access the 
> modules from where they are downloaded?

No and yes, respectively. See above.

>
> and
>
> 2. Would I then place code at the header of my script such as...
>
> use lib "/path/to/bioperl-live";
> use MODULE;
>
> and the updated module will (temporarily) override the other bioperl 
> modules in my @INC?
>
> I tried this, actually, without any noticable change in my previous 
> problem
> (_print_GenBank... still prints the feature tabs as 
> /tab="Bio::Annotation::SimpleValue=HASH(0x87f0914)" instead of 
> /tab="value"), but I don't know for certain if perl was using the 
> modules from my bioperl-live installation or the older ones.
>

I'm not sure about the search order. The POD for lib says:

        It is typically used to add extra directories to perl's search 
path so
        that later "use" or "require" statements will find modules which 
are
        not located on perl's default search path.

but also:

        The parameters to "use lib" are added to the start of the perl 
search
        path. Saying

            use lib LIST;

        is almost the same as saying

            BEGIN { unshift(@INC, LIST) }

Note you can easily test which version is loaded by using either the 
debugger or add some garbage to the module.

	-hilmar


>
> -Andrew Stewart
>
>
> Hilmar Lapp wrote:
>
>>
>>> On Aug 26, 2005, at 11:37 AM, Andrew Stewart wrote:
>>>
>>>> Would it be possible to simply update the module which contains the 
>>>> error (or are there multiple files?) rather than downgrade to 1.4 
>>>> or upgrade to the HEAD branch?
>>>> -Andrew
>>>
>>
>> You could, e.g. using Jason's suggestion, but I don't know why you 
>> wouldn't just want to upgrade to the main trunk. Currently, this is 
>> as close as you can get to upgrading to 1.5.1., which is what you 
>> will want to do anyway immediately once it's out.
>>
>>     -hilmar
>>
>
>
-- 
-------------------------------------------------------------
Hilmar Lapp                            email: lapp at gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757
-------------------------------------------------------------