Author Topic: Script to create photo album from the proprietary software "fotoalbum.es"  (Read 539 times)

jborruel

  • Newbie
  • *
  • Posts: 1
  • Karma: 1
  • Scribus User
Hello. I am brand new in this Scribus and scripting world, but I have to say that it really thrilled me, I absolutely love the concept.

My wife usually creates photo albums from our trips and adventures, and she uses the services from "fotoalbum.es" (Spanish version of the original Germand brand "myphotobook.de" -- https://www.myphotobook.de/), which we find very conveniently balanced regarding quality and price. The problem is that once she produces the album, we can't keep a .pdf copy of the book, but we end up with a bunch of local files that meant nothing to us as consumers.

I digged a little into the file structure of the program, and found out that it just produces a JSON file with a dictionary object with the description of all the items of the album, that I could read out and parse into a Scribus file. It took me a little and ended up with some usable code that produces an album from the program files.

I guess that it won't be very useful for most of the readers of the forum, and I don't even know if this is the right place to post it, but I ended up happy with the results, so I just share it here in case it could be helpful for someone.

It addresses problems like the one posed by rotation, as fotoalbum rotates the object over the center point of the object, while Scribus rotates by default over the upper left most corner, and therefore it is necesary a latter translation (piece of code following "if rotation > 0.5 :").

A lot of work has also been carried out as to zoom and shift every picture within its frame, since it was not straight forward as to figure out all the parameters in the original software. Mainly because the default fotoalbum action is to "fill frame", center the picture within the frame with the largest scale to fill the frame even if part of the picture falls out of the frame. This is not the default behavior in Scribus. And then the zooming and shifting (applying offset to the picture) is applied afterwards.

I guess the code might not be optimal in a few aspects, but I am new to all this world and couldn't do better. Mainly with the html parser that reads out the caption for the text boxes, I used a global variable inside a function, something a little weird, I know...

Anyway, I uploaded the code to github (https://github.com/jborruel/myphotobook-2-scribus/blob/master/myphotobook-2-scribus.py also a new world for me) so anyone interested could improve the code if interested. I attach it here below.

Thanks for such a great tool!
Code: [Select]
#!/usr/bin/env python

"""
This is myphotobook-2-scribus.py. This script will create two Scribus documents
for a specific https://www.fotoalbum.es/
album created with their proprietary software (PhotoGenie X).
I guess it will work for albus created with aney of the other "country flavours"
of the franchise:
https://www.myphotobook.de/
https://www.myphotobook.at/
https://www.myphotobook.ch/
https://www.myphotobook.fr/
https://www.myphotobook.it/
https://www.myphotobook.nl
https://www.myphotobook.be
https://www.myphotobook.co.uk/
https://www.myphotobook.ie
https://www.myphotobook.se
https://www.myphotobook.dk
The directory where the album data are stored is in my case:
~users~\AppData\Roaming\PhotoGenie X\{XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX}\Persistent\Files\projects
I haven't implemented any error checking in this version, as is is made for
my home purposes.
USAGE: It is probably better not to already have any document in Scribus,
since memory usage can become intense when a large number of images is
used.
You are first presented with a dialog to choose the .json album file, and then
the directory for the image files (usually /Persistent/Reserved/)
KNOWN ISSUES / NEED TO DEVELOP:
- Pictures across two pages are not presented properly, as they are doubled
  in both pages: Need to crop anything beyond the page limits
- Need to fix/implement image rotation, as we don't use them. There's an issue
  with some images with the attribute .contentRotation, even though the image
  be stright, the width and height attribute are interchanged.
- Sooo many things with fonts style and size and colour...
- Backgrounds are not taken into account
AUTHOR: Jesus Borruel Original version: 2018.01.30, this version: 2018.01.30
LICENSE: 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.
text following the one presented in (scribalbum_letter.py)
"""

import scribus
import os
import json
import math

from HTMLParser import HTMLParser

#Global variable for keeping the caption text. Not the finest solution though...
caption_text = ""

class MyHTMLParser(HTMLParser):
def handle_data(self, data):
global caption_text
#print "Data:", data
caption_text = caption_text + data

# Read the JSON file into a PYTHON dictionary, which is easier to handle
class JSONObject:
def __init__(self, d):
self.__dict__ = d

jsonfile = scribus.fileDialog("Select Your JSON File", "*.JSON")
imagedir = scribus.fileDialog("Select Your Persistent - Reserved Directory", "Directories",isdir=True) + "/"
#jsonfile = "~users~/AppData/Roaming/PhotoGenie X/{XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX}/Persistent/Files/projects/XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX.json"
#imagedir = "~users~/AppData/Roaming/PhotoGenie X/{XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXX}/Persistent/Reserved/"

with open(jsonfile, 'r') as f:
albumdata = json.load(f, object_hook=JSONObject)

# instantiate the parser and fed it some HTML
parser = MyHTMLParser()

pagenum = 0
layercount = 0
elementcount = 0

print "ALBUM datafile found:" + albumdata.descriptor.name

#In order to process only with a few pages, instead of the whole bunch of pages. Starting Page 1 is compulsory
#pages = [1,6,8]
#for pagenum in pages :

#In order to process only with the first few pages, instead of the whole bunch of pages. Starting Page 1 is compulsory
#for pagenum in range(1,16):

#To process only with the whole bunch of pages
for pagenum in range(len(albumdata.pages)):
if pagenum == 0:
# Cover page. Define font size for the Cover page captions, and new document style
fontsize = 12
size_x = albumdata.pages[pagenum].size.width
size_y = albumdata.pages[pagenum].size.height
# New doc with the cover pages
scribus.newDoc((size_x,size_y), (3,3,3,3), scribus.PORTRAIT, 1, scribus.UNIT_MM, scribus.NOFACINGPAGES, scribus.FIRSTPAGERIGHT)
elif pagenum == 1:
# Album inner pages. Define font size for the album page's captions, and new document style
fontsize = 8
size_x = albumdata.pages[pagenum].size.width
size_y = albumdata.pages[pagenum].size.height
# New doc with the inner pages style
scribus.newDoc((size_x,size_y), (3,3,3,3), scribus.PORTRAIT, 1, scribus.UNIT_MM, scribus.FACINGPAGES, scribus.FIRSTPAGERIGHT)  #libro para el resto de páginas
else:
# Insert new page within the album document
scribus.newPage(-1)
for layercount in range(len(albumdata.pages[pagenum].layers)):
#print ("LAYER: " + str(layercount))
if albumdata.pages[pagenum].layers[layercount].elements: #Check wether there are any elements within the layer, as it could be empty
for elementcount in range(len(albumdata.pages[pagenum].layers[layercount].elements)):
#print ("ELEMENT: " + str(elementcount))
# I don't implement image rotation as we don't use it... Sorry
if albumdata.pages[pagenum].layers[layercount].elements[elementcount].type == 'PICTURE':
size_x = albumdata.pages[pagenum].layers[layercount].elements[elementcount].size.width
size_y = albumdata.pages[pagenum].layers[layercount].elements[elementcount].size.height
x_pos = albumdata.pages[pagenum].layers[layercount].elements[elementcount].position.x - (size_x / 2)
y_pos = albumdata.pages[pagenum].layers[layercount].elements[elementcount].position.y - (size_y / 2)
f = scribus.createImage(x_pos, y_pos, size_x, size_y)
scribus.loadImage(imagedir + albumdata.pages[pagenum].layers[layercount].elements[elementcount].picture.id + ".JPG", f)
scribus.setScaleImageToFrame(scaletoframe=1, proportional=0, name=f)
# https://wiki.scribus.net/canvas/Fitting_an_Image_to_its_Frame
# fitimage2frame_v2.py --> An advantage of this method is that will work regardless of the proportions of the image.
# If you want to use this in Scribus versions >=1.5.0, you will need to change the scaleImage() command to setImageScale(), otherwise with the same parameters inside
xscale, yscale = scribus.getImageScale(f)
width_pi = int(albumdata.pages[pagenum].layers[layercount].elements[elementcount].picture.dimension.width) #original image width in pixels
height_pi = int(albumdata.pages[pagenum].layers[layercount].elements[elementcount].picture.dimension.height) #original image height in pixels
scribus.setScaleImageToFrame(scaletoframe=0, proportional=1, name=f)
if (xscale > yscale):
scale = xscale
scribus.setImageScale(scale, scale, f)
dpmm = width_pi/size_x # dots per mm of the current image
offset_x = 0
offset_y = -(height_pi/dpmm-size_y)/2/scribus.mm # scribus.mm --> scribus constant: millimetres in 1 pt
if albumdata.pages[pagenum].layers[layercount].elements[elementcount].contentRotation > 0.1 : # I don't know why sometimes it mixes dimensions width and height
offset_y = -(width_pi/dpmm-size_x)/2/scribus.mm # scribus.mm --> scribus constant: millimetres in 1 pt
else:
scale = yscale
scribus.setImageScale(scale, scale, f)
dpmm = height_pi/size_y # dots per mm of the current image
offset_x = -(width_pi/dpmm-size_x)/2/scribus.mm # scribus.mm --> scribus constant: millimetres in 1 pt
offset_y = 0
if albumdata.pages[pagenum].layers[layercount].elements[elementcount].contentRotation > 0.1 :# I don't know why sometimes it mixes dimensions width and height
offset_x = -(height_pi/dpmm-size_y)/2/scribus.mm # scribus.mm --> scribus constant: millimetres in 1 pt
scribus.setImageOffset(offset_x,offset_y, f) # offset comes in scribus units pts
#If there is cropping, the image is "shifted" within the image frame, by a certain scale in points.
if albumdata.pages[pagenum].layers[layercount].elements[elementcount].cropping:
#Apply zoom and recenter
zoom_x = float(albumdata.pages[pagenum].layers[layercount].elements[elementcount].cropping.width)
zoom_y = float(albumdata.pages[pagenum].layers[layercount].elements[elementcount].cropping.height)
if (zoom_x > zoom_y):
scale = scale / zoom_x
dpmm = dpmm * zoom_x
elif (zoom_y > zoom_x):
scale = scale / zoom_y
dpmm = dpmm * zoom_y
scribus.setImageScale(scale, scale, f)
#Apply image shift within frame
#Cropping is defined in myphotobook as the percentage of displacement from the centered picture, where 0.5 is centered, and extreme values (0,1)
# is the maximum displacement and therefore to take out completely the image out of the frame
x_shift = float(albumdata.pages[pagenum].layers[layercount].elements[elementcount].cropping.x) # % / 100 respect the center 0.5
y_shift = float(albumdata.pages[pagenum].layers[layercount].elements[elementcount].cropping.y) # % / 100 respect the center 0.5
offset_x = -(width_pi/dpmm-size_x)/2/scribus.mm+(0.5-x_shift)*width_pi/dpmm/scribus.mm # scribus.mm --> scribus constant: number of pts per mm
offset_y = -(height_pi/dpmm-size_y)/2/scribus.mm+(0.5-y_shift)*height_pi/dpmm/scribus.mm
scribus.setImageOffset(offset_x,offset_y, f)
elif (albumdata.pages[pagenum].layers[layercount].elements[elementcount].type == 'TEXT') and (albumdata.pages[pagenum].layers[layercount].elements[elementcount].text != ""):
size_x = albumdata.pages[pagenum].layers[layercount].elements[elementcount].size.width
size_y = albumdata.pages[pagenum].layers[layercount].elements[elementcount].size.height
x_pos = albumdata.pages[pagenum].layers[layercount].elements[elementcount].position.x - (size_x / 2)
y_pos = albumdata.pages[pagenum].layers[layercount].elements[elementcount].position.y - (size_y / 2)
#This is the kind of text defined within the "text" label
#<p style=\"text-align: left;\"><span style=\"color:#000000;font-family: Calibri; font-size: 2.82mm;\">Cataratas de Iguazú (Argentina) 27/11/2016</span></p>\n
caption_text = ""
parser.feed(albumdata.pages[pagenum].layers[layercount].elements[elementcount].text)
#print "caption_text : " + caption_text
captionBox = scribus.createText(x_pos,y_pos,size_x,size_y)
scribus.setText(caption_text, captionBox)
#scribus.setFont("Copper Std", captionBox)
scribus.setFontSize(fontsize, captionBox)
#The following are defined within the element attributes, but they don't seem to be used. The fonts are rather taken from the html info within the "text" attribute
#scribus.setFont(albumdata.pages[pagenum].layers[layercount].elements[elementcount].font.family, captionBox)
#scribus.setFontSize(albumdata.pages[pagenum].layers[layercount].elements[elementcount].font.size, captionBox)
rotation = - albumdata.pages[pagenum].layers[layercount].elements[elementcount].rotation % (2 * math.pi)
if rotation > 0.5 :
# https://stackoverflow.com/questions/34372480/rotate-point-about-another-point-in-degrees-python
# Scribus by default rotates the text box over the upper leftmost corner, while myphotobook uses the center point.
# Therefore, I use the function in the above site to translate the frame after rotation 
# It works for most of our rotated text (upside down, as in the album spine) but sometimes not in the straight horizontal text
# ox, oy = origin
ox, oy = 0,0
# px, py = point
px, py = size_x/2, size_y/2
qx = ox + math.cos(rotation) * (px - ox) - math.sin(rotation) * (py - oy)
qy = oy + math.sin(rotation) * (px - ox) + math.cos(rotation) * (py - oy)
scribus.rotateObjectAbs(math.degrees(rotation), captionBox)
scribus.moveObject(size_x/2, size_y/2, captionBox)
scribus.moveObject(qx, qy, captionBox)
#print ("texto: " + str(caption_text))
#print ("rotation: " + str(rotation))
#print ("rotation degrees: " + str(math.degrees(rotation) % 360))
#print ("----------------------------------------"+"\n")
#else:
#print ("no Elements within LAYER containing PICTURE or TEXT")
#else:
#print ("No elements within the LAYER")

scribus.setRedraw(1)
scribus.redrawAll()

a.l.e

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 1394
  • Karma: 19
    • the graphics lab
Re: Script to create photo album from the proprietary software "fotoalbum.es"
« Reply #1 on: February 22, 2018, 11:55:03 am »
well done!

interesting idea...

 

palmate