Posts about programming (old posts, page 33)

2010-09-09 13:50

Making deployment of desktop Python apps trivial: an idea

Proprietor and printer in front of Schwartz Print Shop in Minneapolis

Here's what I'm thinking: how hard could it be to make PyQt app deployment absolutely easy? Well, I am guessing: not very hard.

Here's the trick: see what works in the real world, and adopt it.

Question: what has deployed billions of apps and has its users happy? Answer: phones app stores.

Question: how do they work? Answer: well, that's not that short, so let's start explaining.

As I see it, a reasonable app store has the following components:

A Stable Deployment Target

You can't deploy from the store if you don't know what you are deploying into. If the target platform is shaky, you just can't know how to deploy without user assistance, and we are trying to make this easy for the user, which means that's not acceptable.

So, what's a stable deployment target we can provide?

  • PyQt (so we can deploy GUIs to all major desktop platforms)
  • Python standard library
  • Selected modules

What can be (and should be) bundled with the app?

  • Pure python modules
  • Artwork and other resources

What may be bundled:

  • Python modules written in C/C++, but you then have to redo the app for each platform, and that kinda sucks.

Deployment Services

  • Apps should be able to check if there is a new version of them in the store, to ask for upgrades.
  • Apps should be added by the deployment platform nicely into the host system's menus, desktop, etc.

Monetization Services

  • Some way to charge for apps. Even for open source apps, you could ask for U$S0.99 if you install them through the store. Optional, of course, and up to the app owner.
  • Ad platform? There must be a good one for desktop apps somewhere?

The Store Itself

  • A website that downloads a "package" associated with a local deployment application.
  • A app store app. Install things not via web, but via a desktop application.

I don't expect a functional version of this would take me more than a week working fulltime to implement. Of course then there are all sorts of usability, looks, etc. things to consider.

And... I am going to do something I very rarely do. I am going to ask for money.

As an experiment, I have setup a project at http://www.indiegogo.com/Qt-Shop and set a funding goal of U$S 600.

There you can fund me. I promise that if the project is totally funded, I will deliver. If it isn't, I may deliver anyway. I would prefer to have the money though.

The platform would be released under GPLv2 or later.

2010-09-08 13:28

Why we are here.

Warning: rant ahead.

Yesterday the government of Argentina announced that they are giving way 3 million netbooks to students. They also announced that they are giving them the option of Ubuntu or Windows 7.

There was, of course, the typical reaction from the FLOSS side: why are they giving Windows to the students when Linux is better? It's unfair that the government pays for Windows!

I am here to tell you to grow up and stop being a baby. I am here to tell you to stop treating others like babies.

I think I can do this because I am immune to criticism from the FLOSS crowd: I am a member of that crowd. I have an awesome FLOSS pedigree, I have used nothing but Linux for over 15 years. And I have a thick skin and I don't care much what other people say, in principle, unless they give me good reasons to care. And I am telling you to stop complaining.

I am telling you that if the only reason to use a specific piece of software is because it's cheaper, you are accepting that piece of software sucks.

I want people not to just use Linux, I want them to want to use Linux. I want them to wait anxiously for the next release of Ubuntu or Firefox or whatever.

And the first step towards excellence is wanting to be excellent. If having to pay nothing for Windows or Ubuntu there is a certainty that Windows will win, then Ubuntu freaking sucks and needs to improve. People are not adopting it even if it's free? Then something is wrong, and figuring out what is important.

But even more important than finding the missing piece is knowing a piece is missing. Open source has grown complacent. It's grown self righteous. It's become adolescent, sure of its awesomeness and immortality.

I don't believe in many things, but I do believe in free will. I believe that people are not morons, I believe that if they prefer Windows, it's because it does something better, and I believe that whatever that is (and I don't really know what it may be), it can be found, and can be improved, and can be replaced, and other things can be added, and people will want to use the better product.

And if they don't... well, at least we fought an honest fight, and we did our best, and we (hopefully) had fun in the process, and pushed the envelope, and created nice things, and the users are better off in the end even if our babies are not the chosen ones, because we raised the level of everything.

For example, before Linux, Windows sucked much, much more than it does now, and I think many of those improvements were because of Linux, and I am happy that today Windows users have a OS that doesn't stink.

I want free and open source software to be used because it's awesome, not because it's cheap. Awesome and cheap I can live with. Just cheap? That sucks.

And the constant "they use windows because they don't know better"? That's patronizing and condescending, and very, very annoying. And if it annoys me, who is not the target of the lame condescension, trust me, it annoys the crap out of Windows users.

Grow a spine, get your asses into gear, start making awesome stuff, kick ass with quality. That's why we are here. Not to be the cheapest date in town.

2010-09-01 16:36

Goodreads+webcam+python+zbar == hackfun!

I am a big fan of GoodReads a social network for people who read books.

I read a lot, and I like that I can see what other people think before starting a book, and I can put my short reviews, and I can see what I have been reading, and lots more.

In fact, goodreads is going to be a big part of a project I am starting with some PyAr guys.

One thing I have been lazy about is adding my book list to goodreads, because it's a bit of a chore.

Well, chore no more!

Here's how to do it, the hacker way...

  1. Get zbar
  2. Get a cheap webcam
  3. Get a book
  4. Get a 7-line python program (included below)

Now watch the video...

Cute, isn't it?

Here's the code:

import os

p=os.popen('/usr/bin/zbarcam','r')
while True:
    code = p.readline()
    print 'Got barcode:', code
    isbn = code.split(':')[1]
    os.system('chromium http://www.goodreads.com/search/search?q=%s'%isbn)

2010-08-29 16:58

PET: English Translation Issue 1 has a date

Because it worked once, let's do it again. I have just set a completely arbitrary, and probably too early date for the release of the first english Issue of "PET: Python Entre Todos" magazine.

The english version is called PET, which means "Python Entre Todos: English Translation".

It will have the same contents as the first spanish Issue and... it will be the last Issue done like this.

From now on, both versions will be published at the same time, if we can.

So, there will be a very short gap between the english first Issue and the second one (less than a month, we hope).

So, stay tuned

2010-08-13 18:46

Things I learned publishing a magazine

Today at 00:00:00 GMT-3 PET: Python entre todos was indeed launched, in time (arbitrary but forced) and in budget ($0).

So, what did I learn? I learned a lot!

  • The only thing you need to publish an e-mag is time and content.
  • Time can be converted into content, but if you write everything yourself it's a blog, not a magazine. Luckily, PET found great contributors.
  • If you want utilitarian design, rst2pdf can do the job
  • In fact, it can do it better than other tools in some ways
    • I can push a fixed version of the PDFs in 5 minutes for all layouts. How much would it take me using Scribus or other DTP? In a magazine where correctness matters, that's a big deal.
    • TOCs are better than in most amateur PDF magazines I've seen. The in-content-TOC is clickable, and the PDF TOC is perfect.
    • Page numbers in the PDF TOC make sense (no, the cover is not page 1)
    • I am producing 6 PDF versions: A4(bw, colour), A5(bw,colour), Booklet(bw, colour) and I could add any other I want in a few minutes.
  • I learned about PDF imposition!

Let's explain the last one:

Suppose you want to print a small booklet, and you have 32 pages of content. How do you do that?

The easiest way is to print it 2-up double-sided in A4 paper so that you can stack the pages, fold them down the middle, staple them, and get a nice A5 booklet.

The problem is that the page ordering is hard to get right. For example, for a 4-page booklet, you need to print one A4 page with pages 4-1 on one side and 2-3 on the other. For an 8 page booklet it's 8-1,2-7,3-6,4-5.

Lucklily there's a way to get this done automatically:

1. Install podofo 3. Get booklet-A4.plan (see below) 2. Run this:

podofoimpose my-A5-pages.pdf my-booklet.pdf booklet-A4.plan lua

booklet-A4.plan is this:

---Generic Booklet (A4)
---
---It is said generic as it will try to determine
---automatically how to fit the booklet onto A4
---paper sheets, scaling pages if necessary.
---it is well suited for office documents for
---which you do not care too much about resulting
---imposition artefacts since it manages to save
---paper!
---
-- print("Booklet")
-- We output an A4 booklet
PageWidth = 595.27559
PageHeight = 841.88976

print("PageCount",PageCount)

-- We assume that H > W
-- Argh, we now can do better since we have "if" ;-)
-- Scale = PageHeight / (2*SourceWidth)
if(SourceWidth <= SourceHeight)
then
--  If you A5 pages are not really A5, uncomment the next line
--  Scale = PageHeight / (2*SourceWidth)
    Scale = 1
    rot = 90
        xof = SourceHeight
        yofRA = 0
        yofRB = SourceWidth
        yofVA = 0
        yofVB = SourceWidth
else
--  If you A5 pages are not really A5, uncomment the next line
--  Scale = PageHeight / (2*SourceHeight)
    Scale = 1
    rot = 0
        xof = 0;
        yofRA = 0
        yofRB = SourceHeight
        yofVA = SourceHeight
        yofVB = 0
end

do
    rest = PageCount % 4
    totp = PageCount
    if rest ~= 0
        then
        totp = totp + ( 4 - rest)
        end
    inc = 0
    count = 0
    imax = totp/4
    while count < imax
        do
--          We assume that podofoimpose will discard invalid records
--          such as those with source page greater than PageCount
--          print(totp, inc, rot, xof,yofRA, yofRA, yofVA, yofVB)
-- Recto
        PushRecord(totp - inc , inc + 1 , rot, xof , yofRA)
        PushRecord(inc + 1 , inc + 1 , rot, xof , yofRB)
-- Verso
        PushRecord(inc + 2 , inc + 2 , rot, xof , yofVA)
        PushRecord(totp-(inc + 1) , inc + 2 , rot, xof, yofVB)

        count = count + 1
        inc = inc + 2
        end
end

That code is taken from here: http://www.oep-h.com/impose/

And voilá, you get a scrambled PDF with the pages in exactly the right order (and empty pages added as needed).

2010-08-10 16:57

Come see me in Bahía Blanca next weekend!

I will be speaking at the Jornadas del Sur in Bahía Blanca this weekend (August 14/15 and 16).

For a change, I will not be giving my old tired talks, but a brand new one, called "The Amateur", and probably a lightning talk or something else.

The usual offer of free beer will not be possible this year, so: If you mention this blog in the QA session, you get... free candy!

After that, I will be speaking at FM La Tribu on saturday August 21st in the Charlas Abiertas 2010 where I will be speaking about a lot of things between virtualenv and nose and tox and other testing-related topics.

2010-08-09 15:57

It's coming: PET - Python Entre Todos

Edited

I removed the countdown because it was wrong (timezone problem). Check http://revista.python.org.ar for one that works :-)

I have been dragged into YAP (Yet Another Project) and in this case it's the first-issue-ever of PET - Python Entre Todos a python online magazine in spanish.

This is an emergent output of PyAr the online Python community of Argentina, and it may never have a second issue unless we get a lot of help, so pay attention to the first one :-)

I have been doing webmastering, editing and PDF production, Emiliano Dalla Verde Marcozzi has been helping everywhere and getting articles.

The first issue is some 45 A5 pages (or 23 A4 ones) in small type, so it's not exactly huge, but it has half a dozen articles aimed at different python levels.

It has been a lot of fun so far, I hope it's well received!

2010-08-05 18:55

Extending rst2pdf: easy and powerful

I do almost all my business writing (and my book) using restructured text. And when I want to produce print-quality output, I tend to use my own tool, rst2pdf.

It's popular, surely my most popular program, but very few know it's also extremely easy to extend (hat tip to Patrick Maupin, who wrote this part!). And not only that, but you can make it do some amazing stuff with a little effort.

To show that, let's create the most dazzling section headings known to man ( Let's see you do what this baby can do in LaTeX ;-).

First: define the problem.

The titles rst2pdf can produce are boring. If you pull every lever and push every button, you may end with the title text, in a Comic Sans, right-aligned, in pink lettering, with avocado-green background and a red border.

And that's as far as the customization capabilities go using stylesheets. That's usually enough, because rst2pdf is not meant for brochures or something like that (but I have done it).

The real problem is that when you get all graphic-designer on rst2pdf, you lose document structure, because you are not being semantic.

Second: define the goal.

So, imagine you want to make a heading that looks like this:

fancytitles1

The image is taken from the library of congress with some light (and bad) gimping by me to leave that empty space at the left, and the title was added using Inkscape.

Can you do that with rst2pdf? Hell no you can't. Not without coding. So let's code an extension that lets you create any heading you like within the limits of Inkscape!

First, we create a SVG template for the headings (it's a bit big because it has the bitmap embedded).

Three: the image-heading flowable

Suppose you have an image of the heading just like the one above. How would you draw that in a PDF? In reportlab, you do that using flowables which are elements that compose the story that is your document. These flowables are arranged in pages, and that's your PDF.

If you are doing a heading, there's a bit more, in that you need to add a bookmark, so it appears on the PDF table of contents.

So, here's a flowable that does just that. It's cobbled from pieces inside rst2pdf, and is basically an unholy mix of Heading and MyImage:

class FancyHeading(MyImage):
  '''This is a cross between the Heading flowable, that adds outline
  entries so you have a PDF TOC, and MyImage, that draws images'''

  def __init__(self, *args, **kwargs):
      # The inicialization is taken from rst2pdf.flowables.Heading
      self.stext = kwargs.pop('text')
      # Cleanup title text
      self.stext = re.sub(r'<[^>]*?>', '', unescape(self.stext))
      self.stext = self.stext.strip()

      # Stuff needed for the outline entry
      self.snum = kwargs.pop('snum')
      self.level = kwargs.pop('level')
      self.parent_id= kwargs.pop('parent_id')


      MyImage.__init__(self, *args, **kwargs)

  def drawOn(self,canv,x,y,_sW):

      ## These two lines are magic.
      #if isinstance(self.parent_id, tuple):
          #self.parent_id=self.parent_id[0]

      # Add outline entry. This is copied from rst2pdf.flowables.heading
      canv.bookmarkHorizontal(self.parent_id,0,0+self.image.height)

      if canv.firstSect:
          canv.sectName = self.stext
          canv.firstSect=False
          if self.snum is not None:
              canv.sectNum = self.snum
          else:
              canv.sectNum = ""

      canv.addOutlineEntry(self.stext.encode('utf-8','replace'),
                                self.parent_id.encode('utf-8','replace'),
                                int(self.level), False)

      # And let MyImage do all the drawing
      MyImage.drawOn(self,canv,x,y,_sW)

And how do we tell rst2df to use that instead of a regular Heading? by overriding the TitleHandler class. Here's where the extension magic kicks in.

If you define, in an extension, a class like this:

class FancyTitleHandler(genelements.HandleParagraph, docutils.nodes.title):

Then that class will handle all docutils nodes of class docutils.nodes.title. Here, I just took rst2pdf.genelements.HandleTitle and changed how it works for level-1 headings, making it generate a FancyHeading instead of a Heading... and that's all there is to it.

class FancyTitleHandler(genelements.HandleParagraph, docutils.nodes.title):
  '''
  This class will handle title nodes.

  It takes a "titletemplate.svg", replaces TITLEGOESHERE with
  the actual title text, and draws that using the FancyHeading flowable
  (see below).

  Since this class is defined in an extension, it
  effectively replaces rst2pdf.genelements.HandleTitle.
  '''

  def gather_elements(self, client, node, style):
      # This method is copied from the HandleTitle class
      # in rst2pdf.genelements.

      # Special cases: (Not sure this is right ;-)
      if isinstance(node.parent, docutils.nodes.document):
          #node.elements = [Paragraph(client.gen_pdftext(node),
                                      #client.styles['title'])]
          # The visible output is now done by the cover template
          node.elements = []
          client.doc_title = node.rawsource
          client.doc_title_clean = node.astext().strip()
      elif isinstance(node.parent, docutils.nodes.topic):
          node.elements = [Paragraph(client.gen_pdftext(node),
                                      client.styles['topic-title'])]
      elif isinstance(node.parent, docutils.nodes.Admonition):
          node.elements = [Paragraph(client.gen_pdftext(node),
                                      client.styles['admonition-title'])]
      elif isinstance(node.parent, docutils.nodes.table):
          node.elements = [Paragraph(client.gen_pdftext(node),
                                      client.styles['table-title'])]
      elif isinstance(node.parent, docutils.nodes.sidebar):
          node.elements = [Paragraph(client.gen_pdftext(node),
                                      client.styles['sidebar-title'])]
      else:
          # Section/Subsection/etc.
          text = client.gen_pdftext(node)
          fch = node.children[0]
          if isinstance(fch, docutils.nodes.generated) and \
              fch['classes'] == ['sectnum']:
              snum = fch.astext()
          else:
              snum = None
          key = node.get('refid')
          maxdepth=4
          if reportlab.Version > '2.1':
              maxdepth=6

          # The parent ID is the refid + an ID to make it unique for Sphinx
          parent_id=(node.parent.get('ids', [None]) or [None])[0]+u'-'+unicode(id(node))
          if client.depth > 1:
              node.elements = [ Heading(text,
                      client.styles['heading%d'%min(client.depth, maxdepth)],
                      level=client.depth-1,
                      parent_id=parent_id,
                      node=node,
                      )]
          else: # This is an important title, do our magic ;-)
              # Hack the title template SVG
              tfile = open('titletemplate.svg')
              tdata = tfile.read()
              tfile.close()
              tfile = tempfile.NamedTemporaryFile(dir='.', delete=False, suffix='.svg')
              tfname = tfile.name
              tfile.write(tdata.replace('TITLEGOESHERE', text))
              tfile.close()

              # Now tfname contains a SVG with the right title.
              # Make rst2pdf delete it later.
              client.to_unlink.append(tfname)

              e = FancyHeading(tfname, width=700, height=100,
                  client=client, snum=snum, parent_id=parent_id,
                  text=text, level=client.depth-1)

              node.elements = [e]

          if client.depth <= client.breaklevel:
              node.elements.insert(0, MyPageBreak(breakTo=client.breakside))
      return node.elements

The full extension is in SVN and you can try it this way:

[fancytitles]$ rst2pdf -e fancytitles -e inkscape demo.txt -b1

You need to enable the Inkscape extension so the SVG will look nice. And here's the output:

fancytitles2

You can override how any element is handled. That's being extensible :-)

2010-07-24 20:37

This is why Qt (and PyQt) are cool

Alejandro Dolina once wrote (and this is from memory that's probably 25 years old) of a round table discussing "What's Tango?", and how after two hours of discussing the nature, characteristics and history of tango, one of the members of the panel picked up a bandoneón, played "El apache argentino" stood up and left without saying a word.

So, why are Qt and PyQt cool?

Audio player widget:

# -*- coding: utf-8 -*-

import sys, os
from PyQt4 import QtCore, QtGui, uic
from PyQt4.phonon import Phonon
import icons_rc

class AudioPlayer(QtGui.QWidget):
    def __init__(self, url, parent = None):

        self.url = url

        QtGui.QWidget.__init__(self, parent)
        self.setSizePolicy(QtGui.QSizePolicy.Expanding,
            QtGui.QSizePolicy.Preferred)


        self.player = Phonon.createPlayer(Phonon.MusicCategory,
            Phonon.MediaSource(url))
        self.player.setTickInterval(100)
        self.player.tick.connect(self.tock)

        self.play_pause = QtGui.QPushButton(self)
        self.play_pause.setIcon(QtGui.QIcon(':/icons/player_play.svg'))
        self.play_pause.clicked.connect(self.playClicked)
        self.player.stateChanged.connect(self.stateChanged)

        self.slider = Phonon.SeekSlider(self.player , self)

        self.status = QtGui.QLabel(self)
        self.status.setAlignment(QtCore.Qt.AlignRight |
            QtCore.Qt.AlignVCenter)

        self.download = QtGui.QPushButton("Download", self)
        self.download.clicked.connect(self.fetch)

        layout = QtGui.QHBoxLayout(self)
        layout.addWidget(self.play_pause)
        layout.addWidget(self.slider)
        layout.addWidget(self.status)
        layout.addWidget(self.download)

    def playClicked(self):
        if self.player.state() == Phonon.PlayingState:
            self.player.pause()
        else:
            self.player.play()

    def stateChanged(self, new, old):
        if new == Phonon.PlayingState:
            self.play_pause.setIcon(QtGui.QIcon(':/icons/player_pause.svg'))
        else:
            self.play_pause.setIcon(QtGui.QIcon(':/icons/player_play.svg'))

    def tock(self, time):
        time = time/1000
        h = time/3600
        m = (time-3600*h) / 60
        s = (time-3600*h-m*60)
        self.status.setText('%02d:%02d:%02d'%(h,m,s))

    def fetch(self):
        print 'Should download %s'%self.url

def main():
    app = QtGui.QApplication(sys.argv)
    window=AudioPlayer(sys.argv[1])
    window.show()
    # It's exec_ because exec is a reserved word in Python
    sys.exit(app.exec_())

if __name__ == "__main__":
    main()

Video player widget:

import sys, os
from PyQt4 import QtCore, QtGui, uic
from PyQt4.phonon import Phonon
import icons_rc

class VideoPlayer(QtGui.QWidget):
    def __init__(self, url, parent = None):

        self.url = url

        QtGui.QWidget.__init__(self, parent)
        self.setSizePolicy(QtGui.QSizePolicy.Expanding,
            QtGui.QSizePolicy.Preferred)


        self.player = Phonon.VideoPlayer(Phonon.VideoCategory,self)
        self.player.load(Phonon.MediaSource(self.url))
        self.player.mediaObject().setTickInterval(100)
        self.player.mediaObject().tick.connect(self.tock)

        self.play_pause = QtGui.QPushButton(self)
        self.play_pause.setIcon(QtGui.QIcon(':/icons/player_play.svg'))
        self.play_pause.clicked.connect(self.playClicked)
        self.player.mediaObject().stateChanged.connect(self.stateChanged)

        self.slider = Phonon.SeekSlider(self.player.mediaObject() , self)

        self.status = QtGui.QLabel(self)
        self.status.setAlignment(QtCore.Qt.AlignRight |
            QtCore.Qt.AlignVCenter)

        self.download = QtGui.QPushButton("Download", self)
        self.download.clicked.connect(self.fetch)
        topLayout = QtGui.QVBoxLayout(self)
        topLayout.addWidget(self.player)
        layout = QtGui.QHBoxLayout(self)
        layout.addWidget(self.play_pause)
        layout.addWidget(self.slider)
        layout.addWidget(self.status)
        layout.addWidget(self.download)
        topLayout.addLayout(layout)
        self.setLayout(topLayout)

    def playClicked(self):
        if self.player.mediaObject().state() == Phonon.PlayingState:
            self.player.pause()
        else:
            self.player.play()

    def stateChanged(self, new, old):
        if new == Phonon.PlayingState:
            self.play_pause.setIcon(QtGui.QIcon(':/icons/player_pause.svg'))
        else:
            self.play_pause.setIcon(QtGui.QIcon(':/icons/player_play.svg'))

    def tock(self, time):
        time = time/1000
        h = time/3600
        m = (time-3600*h) / 60
        s = (time-3600*h-m*60)
        self.status.setText('%02d:%02d:%02d'%(h,m,s))

    def fetch(self):
        print 'Should download %s'%self.url

def main():
    app = QtGui.QApplication(sys.argv)
    window=VideoPlayer(sys.argv[1])
    window.show()
    # It's exec_ because exec is a reserved word in Python
    sys.exit(app.exec_())

if __name__ == "__main__":
    main()

...

2010-07-23 02:50

Desktop apps and clouds (with video)

I enjoy creating desktop applications. That means I may be a member of a dying breed, since web apps are going to make us all obsolete next week, but I do enjoy doing it.

The bad side of it is, of course that sometimes it's much more convenient to use a web application. For example, I have abandoned my own baby (uRSSus) because google reader is just easier and more convenient to use.

But then I thought... what bothers me of uRSSus? And there are quite a few things!

  1. It's not in all computers I may use

    That means I will not ever be able to use it exclusively.

  2. It's pretty useless without an Internet connection (but so is google reader mostly)

  3. Since I can't use it exclusively, I end with feeds on uRSSus that are not on google reader and viceversa.

  4. It's freaking slow

So, I decided to see what I could do about that without giving up the good side of uRSSus:

  1. It looks much nicer than a web app, because it looks like a desktop app
  2. It does things like opening the site instead of showing the feed item (great for partial content feeds)
  3. I wrote it (yes, that's a feature for me. I like self-made programs)

So, this attempt at rewriting the desktop RSS reader produced this:

As you can see in the above video, this reader syncs the subscription list to google reader. It will also eventually sync your read/unread posts.

It still can open full sites instead of feed items, it has/will have a heck of an offline mode (full pages captured as images, for example), and... it's very very fast.

It's much faster than google reader in Chromium, and hella faster than uRSSus. That was done via smarter coding, so it probably means I was braindead before and experienced a minor recovery.

The code is not fit for release (for example, the database schema will change) but you can try it: http://code.google.com/p/kakawana/source/checkout

Contents © 2000-2019 Roberto Alsina