# Copyright (C) 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.

#

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

# $Id$

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

"""Support page construction.

The page construction is an XHTML document consisting of several

independent components that, when put together, provide organization

to content. Each of these components is set as a method of Layout

class that can be instantiated later from application specific modules.

When you create a new application package, you need to create a page

module for it and instantiate the Layout class provided here inside

it.  Later, the following functions must be created: page_content(),

page_navibar() and main(). These functions are used to define the

content and navigation bar of your application. Both application

content and application navigation are logically organized using

variables passed through the URL.

Application

===========

URL variable: app

This variable contains the application id. It is a unique numerical

value that starts at 0 and increments one for each new application

that might be added. The application identified by number 0 is the one

used as default when no other application is provided.  The

application identified by number 0 is added to database the first time

it is created as part of the initial configuration process.

Application is the highest level of organization inside

`webenv.cgi' script. Inside applications, there is content in form

of pages and entries. Content can be grouped by categories.

Pages

=====

URL variable: page

This variable contains the page id. It is a unique numerical value

that starts at 0 and increments in one for each new page added to the

application. In contrast to applications, the page identified by

number 0 is not used as default page when no other page is provided.

This configuration is specific to each application and can be

customized inside each application individually, using string values

instead of numerical values when passing values to page variable.

Generally, when a page variable isn't passed through the URL, the

application module uses the `content_list()' method from Layout class

to display a list of all available content entries while links to

content pages are displayed in the application navigation bar so users

can access them.  The unique numerical value of content pages is

specific to each application, so there is one page 0 for each

application available. No page is added to database the first time the

database is created as part of the initial configuration process.

Pages contain similar information to that described by contents with

few exceptions. Pages, in contrast to entries, can differentiate the

page title from the page name. The page title goes in the page content

itself and describes what the page is about with a phrase. On the

other hand, the page name is generaly one word describing the page

content and is used as link on the application navigation bar.  When

no page name is explicitly provided, the first word of page title is

used instead.

Pages are always accessible inside the same application while contents

aren't.  Pages are permanently visible and linkend from each

application specific navigation bar.  This kind of pages can be

managed by editors or administrators and can be marked as `draft' to

put it on a special state where it is possible for administrator,

editors and authors to work on it, but impossible for others to read

it until the page be marked as `published' by either the page author

or any members of editor's or administrator's groups.

Pages can be converted to entires and the oposite. When convertion

occurs, unused information looses its meaning and is kept for

informative purpose, specially in situations when it might be needed

to realize a convertion back into the former state. Notice that in

order to realize such a back and forth convertion it is required that

both pages and entires share the same definition structure.  In fact,

that they be the same thing, but able to differentiate themselves

either as page or entry (e.g., through a `type' field.).

Pages content is under version control. When a page (or entry) is

changed, a verification is performed to determine whether the

information entered in edition matches the last record in the page

history table. When both the information coming from edition and the

last record in the page history table are the same (e.g., no change

happened) the edition action is cancelled and a message is printed out

to notify the action.  Otherwise, when the information entered in

edition differs from the last record in the page history table, the

information comming from edition passes to be the last record in the

page history table.  In case, a page be reverted to a revision

different to that one being currently the active page, the reverted

revision becomes the active page (e.g., by changing a `status' field

from `false' to `true' in the history table).

Categories

==========

Categories exists to organize contents. When an entry is created it is

automatically linked to a category. Categories are managed by

administrators and editors only. Categories can be nested one another

and provide another way of finding information inside the web

environment.  Categories are specific to each web application, just as

contents and pages are. The `Unknown' category is created when the

categories table is created for first time, as part of the initial

configuration process so if no explicit category assignation is set by

the user, a default value (the `Unknown' category in this case) is

used to satisfy the connection between contents and categories.

Referential integrity

=====================

Referential integrity is not handle in the logic layer provided by

this module, but set inside the database system used to store the

information handled by this module. The most we do about it here, is

to display a confirmation message before committing such actions, so

you can be aware of them.

"""

import cgi

import cgitb; cgitb.enable()

from Apps import xhtml

qs = cgi.parse()

def qs_args( names={}):

    """Returns query string arguments.

    The query string arguments are used to build links dynamically

    and, this way, to create a browsable and logically organized web

    environment.  Such a construction generally needs to retrive some

    of the values previously passed to the query string and add new

    ones to it.

    names: A dictionary containing the variable name and value pair

        used to build a new query string. 

    When a variable is provied without a value, then its value is

    retrived from the current query string. If a value isn't found

    there neither, then the variable is removed from the new query

    string.

    When a variable is provided with its value, then its value is used

    to build the new query string.

    """

    output = ''

    names_keys = names.keys()

    names_keys.sort()

    for key in names_keys:

        if names[key] == '':

            if key in qs:

                names[key] = qs[key][0]

            else:

                continue

        if output == '':

            output = '?'

        else:

            output += '&'

        output += key + '=' + str(names[key])

    return '/webenv/' + output

class Layout(xhtml.Strict):

    """The Page Layout.

    The page layout is made by combining XHTML tags in specific ways.

    These specific combinations make the page components which in turn

    can be also combined. Some of these components can be reused and

    others don't. The goal of this class is to define what such

    components are and describe them well in order to understand how

    to use them from application modules when building XHTML documents

    dynamically.

    The page layout is initialized with a functional layout that can

    be used as reference inside application modules, to create

    variations of it. Generally, inside application packages, this

    class is instantiated in a module named `page', variables are

    reset and functions created in order to satisfy that application

    needs. When you need to output one of the page components then you

    use this class instantiated methods. When the method you need

    doesn't exist in this class, then it is a good time for it to be

    created, here ;). 

    Notice that most methods defined in this class make direct use of

    methods defined by Strict class inside the `xhtml' module. The

    Strict class inside xhtml module is inherited inside this class so

    all the methods there are also available here. Methods which

    doesn't make a direct use of Strict methods are dependencies of

    those which do make direct use of Strict methods.

    """

    def __init__(self):

        """Initialize page data."""

        self.name = 'Home'

        self.title = 'The CentOS Project'

        self.description = 'Community Enterprise Operating System'

        self.keywords = 'centos, project, community, enterprise, operating system'

        self.copyright = '2009-2011 The CentOS Project. All rights reserved.'

        self.language = 'en'

        # Define page header. This is the information displayed

        # between the page top and the page content.

        self.header = self.logo()

        self.header += self.google_ad()

        self.header += self.navibar()

        self.header += self.releases()

        self.header += self.page_links()

        self.header += self.page_navibar()

        # Define page body. This is the information displayed between

        # the page header and page footer.

        self.body = 'None'

        # Define page footer. This is the information displayed

        # between the page bottom and the page content, the last

        # information displayed in the page.

        self.footer = self.credits()

    def logo(self):

        """Returns The CentOS Logo.

        The page logo is displayed on the top-left corner of the page.

        We use this area to show The CentOS Logo, the main visual

        representation of The CentOS Project. In order to print the

        page logo correctly, the image related must be 78 pixels of

        height.

        """

        attrs = []

        attrs.append({'id': 'logo'})

        attrs.append({'title': 'Community Enterprise Operating System', 'href': '/webenv/'})

        attrs.append({'src': '/webenv-pub/Images/centos-logo.png', 'alt': 'CentOS'})

        return self.tag_div(attrs[0], [8,1], self.tag_a(attrs[1], [12,1], self.tag_img(attrs[2], [0,0]), 0), 1)

    def google_ad_example(self):

        """Returns Google advertisement for offline testings."""

        title = 'Google Advertisement'

        url = '/webenv-pub/Images/ads-sample-468x60.png'

        image = self.tag_img({'src': url, 'alt': title}, [0,0])

        link = self.tag_a({'href': url, 'title': title}, [12,1], image)

        output = self.tag_div({'class':'google-ad'}, [8,1], link, 1)

        output += self.separator({'class':'page-line'}, [8,1])

        return output

    def google_ad(self):

        """Returns Google advertisement for online using."""

        properties = {}

        properties['google_ad_client']    = 'pub-6973128787810819'

        properties['google_ad_width']     = '468'

        properties['google_ad_height']    = '60'

        properties['google_ad_format']    = '468x60_as'

        properties['google_ad_type']      = 'text_image'

        properties['google_ad_channel']   = ''

        properties['google_color_border'] = '204c8d'

        properties['google_color_bg']     = '345c97'

        properties['google_color_link']   = '0000FF'

        properties['google_color_text']   = 'FFFFFF'

        properties['google_color_url']    = '008000'

        attrs = {}

        attrs['type'] = "text/javascript"

        output = '

        for key, value in properties.iteritems():

            output += ' '*16 + key + '="' + value + '";\n'

        output += ' '*16 + '//-->\n'

        properties = self.tag_script(attrs, [12,1], output, 1)

        attrs['src'] = "http://pagead2.googlesyndication.com/pagead/show_ads.js"

        source = self.tag_script(attrs, [12,1], ' ', 0)

        output = self.tag_div({'class':'google-ad'}, [8,1], properties + source, 1)

        output += self.separator({'class':'page-line'}, [8,1])

        return output

    def navibar(self):

        """Returns webenv navigation bar. 

        The webenv navigation bar organizes links to main web

        applications The CentOS Project makes use of. Links to these

        web applications stay always visible, no matter what web

        application the user be visiting (e.g., Wiki, Lists, Forums,

        Projects, Bugs, Docs, Downloads and Sponsors.).  Notice that

        some of these web applications are out of `webenv.cgi'

        scope and they need to code their own webenv navigation bars

        in a way that coincide the one set by `webenv.cgi'.

        """

        names = ['Home', 'Wiki', 'Lists', 'Forums', 'Projects', 'Bugs', 'Docs', 'Downloads', 'Sponsors']

        attrs = []

        focus = self.name

        for i in range(len(names)):

            if names[i].lower() == 'home':

                attrs.append({'href': '/webenv/'})

            else:

                attrs.append({'href': '/webenv/?app=' + names[i].lower()})

        tabs = self.navibar_tabs(names, attrs, focus)

        tabs += self.separator()

        return tabs

    def navibar_tabs(self, names, attrs, focus=''):

        """Returns navigation tabs.

        The navigation tabs are the smaller components a navigation

        bar like "top-level navigation bar" and "application

        navigation bar" are made of.

        names: List containing link names of tabs.

        attrs: List containing a dictionary for each tab link name

            inside the `names' list. Dictionaries inside attrs

            argument contain the link attributes (e.g., accesskey,

            title, and href) used by link names so they can be

            linkable once rendered.

        focus: Name of the link marked as current.

        """

        navibar_tabs = ''

        for i in range(len(names)):

            output = self.tag_span('', [0,0], str(names[i]))

            output = self.tag_a(attrs[i], [16,1], output)

            if str(names[i]).lower() == focus.lower():

                output = self.tag_span({'class': 'current'}, [12,1], output, 1)

            else:

                output = self.tag_span('', [12,1], output, 1)

            navibar_tabs += output

        return self.tag_div({'class': 'tabs'}, [8,1], navibar_tabs, 1)

    def releases(self):

        """Returns The CentOS Distribution last releases.

        This method introduces the `releases' method by providing

        links to it.

        names: List containing release numbers in the form M.N, where M

            means major release and N minor release.

        attrs: List containing a dictionary for each release number

            provided in `names' argument. These dictionaries provide

            the link attributes required by release numbers in order

            for them to be transformed into valid links once the page

            be rendered.

        """

        releases = ''

        names = []

        names.append('6.0')

        attrs = []

        attrs.append({'href': qs_args({'p':'releases', 'id': 6.0})})

        title = self.tag_a({'href': qs_args({'p':'releases'})}, [0,0], 'Last Releases') + ':'

        title = self.tag_span({'class': 'title'}, [16,1], title)

        for i in range(len(names)):

            link = self.tag_a(attrs[i], [20,1], names[i])

            if i == len(names) - 1:

                span = self.tag_span({'class': 'last release'}, [16,1], link, 1) 

            else:

                span = self.tag_span({'class': 'release'}, [16,1], link, 1) 

            releases += span

        releases = self.tag_div({'class': 'left'}, [12,1], title + releases, 1)

        rsslink = self.tag_span('', [0,0], 'RSS')

        rsslink = self.tag_a({'href': qs_args({'rss':'releases'}), 'title': 'RSS'}, [20,1], rsslink)

        rsslink = self.tag_span({'class': 'rss'}, [16,1], rsslink, 1)

        rsslink = self.tag_div({'class': 'right'}, [12, 1], rsslink, 1)

        return self.tag_div({'id': 'last-releases'}, [8,1], releases + rsslink, 1)

    def user_links_logs(self):

        """Return links related to user's logs.

        This function introduces the `logs' module. The `logs' module

        registers all user's activity, from login to logout. This link

        must be display/accessible only after a user has successfully

        login.

        """

        last_visit = self.tag_a({'href': qs_args({'app':'', 'p':'logs'})}, [0,0], 'Logs')

        return self.tag_div({'class': 'logs'}, [12, 1], last_visit, 1)

    def user_links_session(self):

        """Returns links related to user's session.

        This function introduces the `session' module. The `session'

        module provides state to user interactions so their action can

        be registered individually.

        """

        names = []

        attrs = []

        session = ''

        names.append('Lost your password?')

        attrs.append({'href': qs_args({'app':'', 'p':'lostpwd'})})

        names.append('Register')

        attrs.append({'href': qs_args({'app':'', 'p':'register'})})

        names.append('Login')

        attrs.append({'href': qs_args({'app':'', 'p':'login'})})

        for i in range(len(names)):

            output = self.tag_a(attrs[i], [20,1], str(names[i]), 0)

            if i == len(names) - 1:

                output = self.tag_span({'class': 'last'}, [16,1], output, 1)

            else:

                output = self.tag_span('', [16,1], output, 1)

            session += output

        return self.tag_div({'class': 'session'}, [12,1], session, 1)

    def user_links_trails(self, names=['None'], attrs=[{'href': '/webenv/'}]):

        """Returns page trails (a.k.a. breadcrumbs).

        The page breadcrumbs record the last pages the user visited

        inside the current web application. Notice that page

        breadcrumbs are user-specific information, so it isn't

        possible to implement them until a way to manage user sessions

        be implemeneted inside `webenv.cgi' script. Until then,

        keep the tag construction commented and return an empty value.

        names: List with trail link names.

        attrs: Dictionary with trail link attributes.

        """

        links = ''

        for i in range(len(names)):

            if i == len(names) - 1:

                output = self.tag_span({'class':'last'}, [16,1], self.tag_a(attrs[i], [20, 1], names[i]), 1)

            else:

                output = self.tag_span('', [16,1], self.tag_a(attrs[i], [20, 1], names[i], 0), 1)

            links += output

        return self.tag_div({'class': 'trail'}, [12,1], links, 1)

    def user_links(self):

        """Returns user related links.

        The user links are specific to each web application. They are

        shown over the application navigation bar.

        """

        userlinks = self.user_links_logs()

        userlinks += self.user_links_session()

        userlinks += self.user_links_trails()

        return self.tag_div({'class': 'userlinks'}, [8,1], userlinks, 1)

    def page_navibar(self, names=['Welcome'], attrs=[{'href':'/webenv/?p=welcome'}], focus='Welcome'):

        """Returns navigation bar for application main pages.

        names: List containing link names.

        attrs: List containing one dictionary for each link name in

            `names' argument. Dictionaries here contain the link

            attributes needed to make linkable tabs once the page is

            rendered.

        """

        navibar_app = self.navibar_tabs(names, attrs, focus)

        navibar_app += self.separator({'class': 'page-line white'}, [8,1])

        return navibar_app

    def separator(self, attrs={'class': 'page-line'}, indent=[16,1]):

        """Returns separator.

        The separator construction is mainly used to clear both sides

        inside the page, specially when floating elements are around.

        attrs: Dictionary containing hr's div attributes.

        indent: List containing hr's div indentation values.

        """

        line = self.tag_hr({'style': 'display:none;'}, [0,0])

        line = self.tag_div(attrs, indent, line)

        return line

    def license(self):

        """Retruns license link."""

        license = 'Creative Commons Attribution-Share Alike 3.0 Unported License'

        license = self.tag_a({'href': 'http://creativecommons.org/licenses/by-sa/3.0/'}, [0,0], license) + '.'

        return license

    def metadata(self):

        """Returns metadata."""

        metadata = self.tag_meta({'http-equiv': 'content-type', 'content': 'text/html; charset=UTF-8'}, [4,1])

        metadata += self.tag_meta({'http-equiv': 'content-style-type', 'content': 'text/css'}, [4,0])

        metadata += self.tag_meta({'http-equiv': 'content-language', 'content': str(self.language)}, [4,1])

        metadata += self.tag_meta({'name': 'keywords', 'content': str(self.keywords)}, [4,0])

        metadata += self.tag_meta({'name': 'description', 'content': str(self.description)}, [4,1])

        metadata += self.tag_meta({'name': 'copyright', 'content': 'Copyright © ' + str(self.copyright)}, [4,0])

        metadata += self.tag_title('', [4,1], self.title)

        metadata += self.tag_link({'href': '/webenv-pub/stylesheet.css','rel': 'stylesheet', 'type': 'text/css'}, [4,0])

        metadata += self.tag_link({'href': '/webenv-pub/centos-fav.png', 'rel': 'shortcut icon', 'type': 'image/png'}, [4,1])

        return self.tag_head('', [0,1], metadata)

    def searchform(self, size=15):

        """Returns search form.

        The search form redirects user from the current page onto the

        search page, where the keywords previously introduced in the

        input field are processed then.

        size: A number discribing how large the search box is.

        """

        input = self.tag_input({'type':'text', 'value':'', 'size':size}, [0,0])

        action = self.tag_dt({}, [20,1], 'Search')

        action += self.tag_dd({}, [20,1], input)

        action = self.tag_dl({'class':'search'}, [16,1], action, 1)

        return self.tag_form({'action': qs_args({'app':'', 'p':'search'}), 

                              'method':'post', 'title':'Search'},

                              [12,1], action, 1)

    def content_resumen(self, attrs, id, title, user_id, commit_date,

                        update_date, category_id, comments, abstract):

        """Returns content resumen.

        The content resumen is used to build the list of contents,

        output by `content_list()' method. The content resumen intends

        to be concise and informative so the user can grab a general

        idea about the related content and what it is about.

        attrs: A dictionary discribing the rows style.  This is useful

            to alternate the row background colors.

        id: A unique numerical value referring the content

            identification. This is the value used on administrative

            tasks like updating and deleting.

        title: A few words phrase describing the content, up to 255

            characters.

        author_id: A string referring the user email address, as

            specified by RFC2822. The user email address is used as id

            inside The CentOS User LDAP server, where user specific

            information (e.g., surname, lastname, office, phone, etc.)

            are stored in. This is the field that bonds the user with

            the content he/she produces.

        commit_date: A string referring the timestamp the content

            arrived to database for time.

        update_date: A string representing the timestamp the content

            was updated/revised for last time.

        category_id: A number refering the category id the content is

            attached to.

        abstract: One paragraphs describing the content.  This

            information is used to build the page metadata

            information. When this value is not provided no abstract

            information is displayed in the page, but the 

            name="description".../> is built using article's first 255

            characters.

        comments: A number representing how many comments the content

            has received since it is in the database.

        The content itself is not displayed in the resumen, but in

        `content_details()'.

        """

        title = self.tag_a({'href': qs_args({'app':'', 'p':'entry', 'id':id})}, [0,0], title)

        title = self.tag_h3({'class': 'title'}, [20,1], title, 0)

        info = self.content_info(id, user_id, commit_date,

                                 update_date, category_id, comments,

                                 abstract)

        return self.tag_div(attrs, [16,1], title + info, 1)

    def pagination(self):

        """Return content pagination."""

        previous = self.tag_a({'href':''}, [0,0], 'Previous')

        previous = self.tag_span({'class':'previous'}, [20,1], previous)

        next = self.tag_a({'href':''}, [0,0], 'Next')

        next = self.tag_span({'class':'next'}, [20,1], next)

        separator = self.separator({'class':'page-line'}, [20,1])

        return self.tag_div({'class':'pagination'}, [16,1], previous +

                            next + separator, 1)

    def content_info(self, content_id, user_id, commit_date,

                     update_date, category_id, comments, abstract):

        """Return content information.

        The content information provides a reduced view of content so

        people can make themselves an idea of what the content talks

        about. The content information displays content's title,

        author, timestamp, related category, number of comments and an

        abstract of the whole content.

        """

        categories = []

        categories.append('Unknown')

        categories.append('Erratas')

        categories.append('Articles')

        categories.append('Events')

        if category_id <= len(categories):

            category_name = categories[category_id].capitalize()

        else:

            category_id = 0

            category_name = categories[category_id].capitalize()

        category_name = self.tag_a({'href': qs_args({'app':'', 'p':'categories', 'id':category_id})}, [0,0], category_name)

        category_name = self.tag_span({'class':'category'}, [24,1], category_name) 

        users = {}

        users['al@centos.org'] = 'Alain Reguera Delgado'

        users['ana@centos.org'] = 'Ana Tamara Reguera Gattorno'

        users['alina@centos.org'] = 'Alina Reguera Gattorno'

        if user_id in users.keys():

            user_name = self.tag_a({'href':'mailto:' + user_id}, [0,0], users[user_id])

            user_name = self.tag_span({'class':'author'}, [24,1], 'Written by ' + user_name)

        if update_date != commit_date:

            date = self.tag_span({'class':'date'}, [24,1], update_date)

        else:

            date = self.tag_span({'class':'date'}, [24,1], commit_date)

        comments_attrs = {'href': qs_args({'app':'', 'p':'entry', 'id':content_id}) + '#comments'}

        if comments == 1:

            comments = self.tag_a(comments_attrs, [0,0], str(comments) + ' comment')

        elif comments > 1:

            comments = self.tag_a(comments_attrs, [0,0], str(comments) + ' comments')

        else:

            comments = 'No comments'

        comments = self.tag_span({'class':'comment'}, [24,1], comments)

        abstract = self.tag_p({'class':'abstract'}, [24,1], abstract)

        return self.tag_div({'class': 'info'}, [20,1], user_name + date + category_name + comments + abstract, 1)

    def content_list(self):

        """Return list of content.

        The list of content is used to explore the content available

        inside specific pages of specific web applications. The

        information is displayed through paginated rows of content

        that can be filtered to reduce the search results based on

        patterns.  By default, the list of content displays 15 rows,

        but this value can be changed in user's preferences.

        """

        output = ''

        count = 0

        rows = []

        rows.append([0, 'Introduction to CentOS Web Environment',

                    'al@centos.org',