#!/bin/bash

#

# cli_getPathComponent.sh -- This function evaluates one repository

# path and output parts/components from it. Generally, the path

# information is passed to the function's first positional argument

# and the part/component we want to retrive is passed to the

# function's second positional argument. If second argument is not

# passed, then first argument is assumed to be the part/component we

# want to retrive, and action value (ACTIONVAL) variable is used

# instead as source path information.

#

# Copyright (C) 2009, 2010, 2011 The CentOS Project

#

# This program is free software; you can redistribute it and/or modify

# it under the terms of the GNU General Public License as published by

# the Free Software Foundation; either version 2 of the License, or (at

# your option) any later version.

#

# This program is distributed in the hope that it will be useful, but

# WITHOUT ANY WARRANTY; without even the implied warranty of

# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU

# General Public License for more details.

#

# You should have received a copy of the GNU General Public License

# along with this program; if not, write to the Free Software

# Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.

#

# ----------------------------------------------------------------------

# $Id$

# ----------------------------------------------------------------------

function cli_getPathComponent {

    local -a PATTERN

    local LOCATION=''

    local OPTION=''

    # Define location which we retrive information from.

    if [[ "$#" -eq 1 ]];then

        LOCATION="$ACTIONVAL"

        OPTION="$1"

    elif [[ "$#" -eq 2 ]];then

        LOCATION="$1"

        OPTION="$2"

    else

       cli_printMessage "`gettext "Invalid arguments."`" --as-error-line

    fi

    # Define release pattern.

    PATTERN[0]="(([[:digit:]]+)(\.([[:digit:]]+)){,1})"

    # Define architecture pattern. Make it match the architectures the

    # CentOS distribution is able to be installed on.

    PATTERN[1]="(i386|x86_64)"

    # Define theme pattern for trunk, branches, and tags directory

    # structures.

    #

    # trunk:

    # ------

    # Themes are made of `Models' and `Motifs'. `Models' controls the

    # `Motifs' characteristcs and it is not rendereable. On the other

    # hand, `Motifs' controls the theme visual style and do is

    # renderable. Since we only need to know the theme name when we

    # render something, we take the `Motifs' directory structure as

    # reference to find out the theme name we are producing images

    # for.

    #

    # `Motifs' are organized by names, so when we say `a theme name'

    # we are really saying `the artistic motif name of a theme', but

    # for short `theme name' is used instead. `Models' are also

    # organized by names, but we do never use model names to name a

    # thame. We always say the `theme models' or `the models of the

    # theme` to refere the theme component that controls the

    # characteristics of artistic motifs. We never name a thame as its

    # model. We use artistic motifs name for such purpose.

    #

    # Inside artistic motif names, we organize different versions of

    # the same artistic motif by means of numerical worklines.

    # Worklines, inside trunk, have an integer numerical form and

    # begin at `1'. They increment one unit each time a new/fresh

    # visual style ---for the same artistic motif--- is conceived.

    #

    # For example, the Flame theme uses the flame filter of Gimp to

    # produce different fractal patterns randomly. We use the flame

    # filter to produce different visual styles under the same theme

    # since all patterns we produce are based on the same pattern

    # (i.e., the random fractal pattern of Gimp's flame filter). This

    # way, if you no longer want to produce visual styles from flame

    # filter, it is time to create a new artistic motif name for it.

    #

    # This schema is very convenient for designers and packagers,

    # since different designers can create their own worklines and go

    # on with them until CentOS distribution release date is close. At

    # that time, a workline is selected and tagged for packaging.

    # Translators and programmers can continue working indepently, and

    # without affection, in their own directory structures.

    #

    #   Example:

    #                                +------> theme name

    #                                |-->| 

    #   trunk/Identity/Images/Themes/Flame/1

    #   trunk/Identity/Images/Themes/Flame/2 

    #   trunk/Identity/Images/Themes/Flame/3

    #                                      |

    #                                      +----> theme version

    #

    # Sometimes we only need to retrive the theme name, but othertimes

    # both the theme name and its work-line is required as well.

    #

    # branches:

    # ---------

    # The branch development line is treated just as trunk development

    # line does, but we rarely use it since it could be confusing to

    # know whether a tag was created from trunk or branches.

    #

    # Conceive an enumeration schema in order to differentiate

    # branches from trunk is not convenient either, it would introduce

    # an intermediate point to the final production of tags we would

    # need to be aware of.  Instead of such configuration, we prefer

    # to go straightforward from trunk to tags.

    #

    # Intermediate branches for quality assurance might be good in

    # some situations, but not when we produce themes. We need a

    # simple structure, where we could design, render content (using

    # centos-art.sh), and release for testing (through tags).  If

    # something goes wrong in the released tag, it would be fixed in

    # trunk and later released in another tagged relase.

    #

    # Of course, care should be taken to avoid making of trunk

    # development line a chaotic place.  Everbody should know exactly

    # what they are doing therein. We need to design protections to

    # isolate possible damages and that way, we could know exactly

    # what and where we need to concentrate in and put our time on.

    #

    # In resume, do not use branches if you don't need to.  Use trunk

    # development line instead.

    #

    # tags:

    # -----

    # The tag frozen line is mainly used to perform theme releases.

    #

    #   Example:

    #                                   +------> theme name

    #                                   |-->| 

    #       tags/Identity/Images/Themes/Flame/1.0

    #      trunk/Identity/Images/Themes/Flame/1 |--> minor update

    #                                         |----> major udpate

    #

    # Tags have the format X.Z, where X is the first number in the

    # name (e.g., `1') and represents the trunk/branches artistic

    # motif version. The Z represents the second number in the name

    # (e.g., `0') which is the tag version itself.  

    #

    # Tag versions start at `0' and increment one unit each time a new

    # tag is created from the same artistic motif version.  When a new

    # tag is created for the same artistic motif version, the first

    # number in the tag name remains intact.  The first number in the

    # tag name only changes when we create a new tag for a new

    # artistic motif version.

    #

    #   Consider the following relations: 

    #

    #       trunk/Identity/Images/Themes/Flame/1

    #       tags/Identity/Images/Themes/Flame/1.0

    #       tags/Identity/Images/Themes/Flame/1.1

    #       tags/Identity/Images/Themes/Flame/1.2

    #

    #       trunk/Identity/Images/Themes/Flame/2

    #       tags/Identity/Images/Themes/Flame/2.0

    #       tags/Identity/Images/Themes/Flame/2.1

    #       tags/Identity/Images/Themes/Flame/2.2

    #

    #       trunk/Identity/Images/Themes/TreeFlower/1

    #       ...

    #       and so on. 

    #

    # Tag versions are created to release fixes and improvements, Tags

    # are immutable (i.e., once tags are created, they shouldn't be

    # modified.).

    PATTERN[2]="Identity/Images/Themes/(([A-Za-z0-9]+)/(${PATTERN[0]}))/"

    # Identify which part of the release we want to output.

    case "$OPTION" in

        '--release' )

            echo "$LOCATION" | egrep "${PATTERN[0]}" | sed -r "s!.*/${PATTERN[0]}/.*!\1!"

            ;;

        '--release-major' )

            echo "$LOCATION" | egrep "${PATTERN[0]}" | sed -r "s!.*/${PATTERN[0]}/.*!\2!"

            ;;

        '--release-minor' )

            echo "$LOCATION" | egrep "${PATTERN[0]}" | sed -r "s!.*/${PATTERN[0]}/.*!\4!"

            ;;

        '--release-pattern' )

            echo "${PATTERN[0]}"

            ;;

        '--architecture' )

            echo "$LOCATION" | egrep "${PATTERN[1]}" | sed -r "s!${PATTERN[1]}!\1!"

            ;;

        '--architecture-pattern' )

            echo "${PATTERN[1]}"

            ;;

        '--theme' )

            echo "$LOCATION" | egrep "${PATTERN[2]}" | sed -r "s!.*${PATTERN[2]}.*!\1!"

            ;;

        '--theme-name' )

            echo "$LOCATION" | egrep "${PATTERN[2]}" | sed -r "s!.*${PATTERN[2]}.*!\2!"

            ;;

        '--theme-release' )

            echo "$LOCATION" | egrep "${PATTERN[2]}" | sed -r "s!.*${PATTERN[2]}.*!\3!"

            ;;

        '--theme-pattern' )

            echo "${PATTERN[2]}"

            ;;

    esac

}