Re: [144726] trunk/dports/science

2016-01-16 Thread Ryan Schmidt

> On Jan 16, 2016, at 8:54 AM, c...@macports.org wrote:
> 
> Revision
> 144726
> Author
> c...@macports.org
> Date
> 2016-01-16 06:54:58 -0800 (Sat, 16 Jan 2016)
> Log Message
> 
> esorepo: Provide easy access to ESO's ports tree
> 
> Closes #46954
> 
> Added Paths
> 
>   • trunk/dports/science/esorepo/
>   • trunk/dports/science/esorepo/Portfile
> Diff
> 
> Added: trunk/dports/science/esorepo/Portfile (0 => 144726)
> 
> --- trunk/dports/science/esorepo/Portfile (rev 0)
> +++ trunk/dports/science/esorepo/Portfile 2016-01-16 14:54:58 UTC (rev 
> 144726)
> 
> @@ -0,0 +1,127 @@
> 
> +# -*- coding: utf-8; mode: tcl; tab-width: 4; indent-tabs-mode: nil; 
> c-basic-offset: 4 -*- vim:fenc=utf-8:ft=tcl:et:sw=4:ts=4:sts=4
> +# $Id$
> +
> +PortSystem  1.0
> +
> +nameesorepo
> +version 1.1
> +categories  science
> +license GPL-2+
> +platforms   darwin
> +supported_archs noarch
> +maintainers eso.org:usd-help
> +
> +description ESO repository of software for astronomical data 
> reduction
> +
> +long_description\
> +The European Southern Observatory (ESO) maintains and distributes a 
> number \
> +of open source software packages (e.g. data reduction pipelines and 
> front-end \
> +tools) for the astronomical science community. This port provides a 
> convenient \
> +configuration mechanism to add an additional repository URL to allow 
> further \
> +installation of ESO software via MacPorts.
> +
> +homepagehttp://www.eso.org/sci/software
> +master_sites
> ftp://ftp.eso.org/pub/dfs/pipelines/repositories/macports/
> +
> +fetch {}
> +checksum {}
> +extract {}

Instead of overriding the fetch, checksum and extract phases, you should 
declare that the port has no distfiles, by writing on a line by itself:

distfiles

With no distfiles, the master_sites line isn't needed either. ("port lint" will 
complain about missing master_sites, but that's a bug in "port lint".)


> +test {}

The test phase is disabled by default. There is no need to override it with an 
empty block.


> +use_configure no
> +
> +build {
> +  set key_file [open "${workpath}/eso-pubkey.pem" w]
> +  puts ${key_file} "-BEGIN PUBLIC KEY-"
> +  puts ${key_file} 
> "MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEAy2Au5qRNn/zoRRZ7rdsn"
> +  puts ${key_file} 
> "g3lbrM33Iz3xew4yZD5L6LI69qVVfmaUsELreeK+bZAf8bLl5+HYHHvM/pBZq/+6"
> +  puts ${key_file} 
> "LBf+NnESVwmKfhEppHim3MqRNyTfSOJAM+J1xt1/dvZkBX8ehZO8piRRaKUPzauW"
> +  puts ${key_file} 
> "WhahV417jD37AWKXlRCNPxUtgEk7kRrMQuP4olWWVul+m7piGrDMzQsAOtvEMX5R"
> +  puts ${key_file} 
> "J5Ygr3iZNq8WSXhAMqfm6cksxq7kMa8sfpw8lU1jXKfN4wREH3w2oJohkkTYFqUq"
> +  puts ${key_file} 
> "0FBnqBkGZlv5e1FUEFW1rnvUpbzOxgDgfdOKwzqNWw2EGl6baQT63iIqa752kCE7"
> +  puts ${key_file} "SwIDAQAB"
> +  puts ${key_file} "-END PUBLIC KEY-"
> +  close ${key_file}
> +}

The whitespace of this portfile does not conform to what the modeline says the 
whitespace is. (Indents should be at multiples of 4 spaces, not 2.)

Why construct this file in the build phase? Why not just put the real file in 
the files directory and copy it from there? Wouldn't that be much more 
straightforward?

___
macports-dev mailing list
macports-dev@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-dev


Re: [144726] trunk/dports/science

2016-01-16 Thread Clemens Lang
Hi Ryan,

On Sat, Jan 16, 2016 at 09:25:15AM -0600, Ryan Schmidt wrote:
> Instead of overriding the fetch, checksum and extract phases, you
> should declare that the port has no distfiles, by writing on a line by
> itself:
> 
> distfiles
> 
> With no distfiles, the master_sites line isn't needed either. ("port
> lint" will complain about missing master_sites, but that's a bug in
> "port lint".)
> 
> > +test {}
> 
> The test phase is disabled by default. There is no need to override it
> with an empty block.

[...]

> Why construct this file in the build phase? Why not just put the real
> file in the files directory and copy it from there? Wouldn't that be
> much more straightforward?

I've fixed these comments in https://trac.macports.org/changeset/144734.

> The whitespace of this portfile does not conform to what the modeline
> says the whitespace is. (Indents should be at multiples of 4 spaces,
> not 2.)

Formatted to conform to the modeline in
https://trac.macports.org/changeset/144735.

-- 
Clemens
___
macports-dev mailing list
macports-dev@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-dev