Download-Skript von Python 2 auf Python 3 portieren

Programmierung für GNOME und GTK+, GUI-Erstellung mit Glade.
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Samstag 22. Juni 2019, 20:23

Hallo,

ich bin gerade dabei ein angestaubtes Python 2 Skript von GitHub zum Herunterladen des Astronomischen Bild des Tages unter Python 3 lauffähig zu machen.

Bei der Funktion `set_download_folder()`, die die Methode `get_user_special_dir(glib.USER_DIRECTORY_DOWNLOAD)` aus dem Modul `glib` verwendet, um den Download-Pfad des Nutzers ausfindig zu machen, komme ich momentan nicht weiter. Das Paket "python-gtk2" ist zwar installiert, aber trotzdem erscheint folgende Fehlermeldung:

Code: Alles auswählen

Traceback (most recent call last):
  File "/home/ata/Projects/nasa-apod-desktop/nasa_apod_desktop.py", line 25, in <module>
    import glib
ModuleNotFoundError: No module named 'glib'
Der folgende alternative Code für Gtk3 auf der im Kommentar verlinkten Seite auf askubuntu funktioniert leider auch nicht, weil `GLib` ebenfalls nicht gefunden wird.

Code: Alles auswählen

from gi.repository import GLib
downloads_dir = GLib.get_user_special_dir(GLib.UserDirectory.DIRECTORY_DOWNLOAD)
Warum funktioniert die Einbindung von `glib` nicht?

Aktueller Code:

Code: Alles auswählen

#!/usr/bin/env python
# 
# Copyright (c) 2012 David Drake
# 
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# 
#    http://www.apache.org/licenses/LICENSE-2.0
# 
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# 
# 
# nasa_apod_desktop.py
# https://github.com/randomdrake/nasa-apod-desktop
# 
# Written/Modified by David Drake
# http://randomdrake.com 
# http://twitter.com/randomdrake 

import glib
import subprocess
import urllib.request, urllib.parse, urllib.error
import re
import os
import random
import glob
from PIL import Image
from sys import stdout
from sys import exit
from lxml import etree
from datetime import datetime, timedelta

DOWNLOAD_PATH = '/tmp/backgrounds/'
CUSTOM_FOLDER = 'nasa-apod-backgrounds'
RESOLUTION_TYPE = 'stretch'
DEFAULT_RESOLUTION_X = 1024
DEFAULT_RESOLUTION_Y = 768
NASA_APOD_SITE = 'http://apod.nasa.gov/apod/'
IMAGE_SCROLL = True
IMAGE_DURATION = 1200
SEED_IMAGES = 10
SHOW_DEBUG = False


# Use XRandR to grab the desktop resolution. If the scaling method is set to 'largest',
# we will attempt to grab it from the largest connected device. If the scaling method
# is set to 'stretch' we will grab it from the current value. Default will simply use
# what was set for the default resolutions.
def find_display_resolution():
    if RESOLUTION_TYPE == 'default':
        if SHOW_DEBUG:
            print(f"Using default resolution of {DEFAULT_RESOLUTION_X}x{DEFAULT_RESOLUTION_Y}")
        return DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y

    resolution_x = 0
    resolution_y = 0

    if SHOW_DEBUG:
        print("Attempting to determine the current resolution.")
    if RESOLUTION_TYPE == 'largest':
        regex_search = 'connected'
    else:
        regex_search = 'current'

    p1 = subprocess.Popen(["xrandr"], stdout=subprocess.PIPE)
    p2 = subprocess.Popen(["grep", regex_search], stdin=p1.stdout, stdout=subprocess.PIPE)
    p1.stdout.close()
    output = p2.communicate()[0]

    if RESOLUTION_TYPE == 'largest':
        # We are going to go through the connected devices and get the X/Y from the largest
        matches = re.finditer(" connected ([0-9]+)x([0-9]+)+", output)
        if matches:
            largest = 0
            for match in matches:
                if int(match.group(1)) * int(match.group(2)) > largest:
                    resolution_x = match.group(1)
                    resolution_y = match.group(2)
        elif SHOW_DEBUG:
            print("Could not determine largest screen resolution.")
    else:
        reg = re.search(".* current (.*?) x (.*?),.*", output)
        if reg:
            resolution_x = reg.group(1)
            resolution_y = reg.group(2)
        elif SHOW_DEBUG:
            print("Could not determine current screen resolution.")

    # If we couldn't find anything automatically use what was set for the defaults
    if resolution_x == 0 or resolution_y == 0:
        resolution_x = DEFAULT_RESOLUTION_X
        resolution_y = DEFAULT_RESOLUTION_Y
        if SHOW_DEBUG:
            print("Could not determine resolution automatically. Using defaults.")

    if SHOW_DEBUG:
        print(f"Using detected resolution of {resolution_x}x{resolution_y}")

    return int(resolution_x), int(resolution_y)


# Uses GLib to find the localized "Downloads" folder
# See: http://askubuntu.com/questions/137896/how-to-get-the-user-downloads-folder-location-with-python
def set_download_folder():
    downloads_dir = glib.get_user_special_dir(glib.USER_DIRECTORY_DOWNLOAD)
    if downloads_dir:
        # Add any custom folder
        new_path = os.path.join(downloads_dir, CUSTOM_FOLDER)
        if SHOW_DEBUG:
            print("Using automatically detected path:", new_path)
    else:
        new_path = DOWNLOAD_PATH
        if SHOW_DEBUG:
            print("Could not determine download folder with GLib. Using default.")
    return new_path


# Download HTML of the site
def download_site(url):
    if SHOW_DEBUG:
        print("Downloading contents of the site to find the image name")
    opener = urllib.request.build_opener()
    req = urllib.request.Request(url)
    try:
        response = opener.open(req)
        reply = response.read()
    except urllib.error.HTTPError as error:
        if SHOW_DEBUG:
            print(f"Error downloading {url} - {str(error.code)}")
        reply = f"Error: {str(error.code)})"
    return reply


# Finds the image URL and saves it
def get_image(text):
    if SHOW_DEBUG:
        print("Grabbing the image URL")
    file_url, filename, file_size = get_image_info('a href', text)
    # If file_url is None, the today's picture might be a video
    if file_url is None:
        return None

    if SHOW_DEBUG:
        print(f"Found name of image: {filename}")

    save_to = os.path.join(DOWNLOAD_PATH, os.path.splitext(filename)[0] + '.png')

    if not os.path.isfile(save_to):
        # If the response body is less than 500 bytes, something went wrong
        if file_size < 500:
            print("Response less than 500 bytes, probably an error\nAttempting to just grab image source")
            file_url, filename, file_size = get_image_info('img src', text)
            # If file_url is None, the today's picture might be a video
            if file_url is None:
                return None
            print(f"Found name of image: {filename}")
            if file_size < 500:
                # Give up
                if SHOW_DEBUG:
                    print("Could not find image to download")
                exit()

        if SHOW_DEBUG:
            print("Retrieving image")
            urllib.request.urlretrieve(file_url, save_to, print_download_status)

            # Adding additional padding to ensure entire line 
            if SHOW_DEBUG:
                print(f"\rDone downloading {human_readable_size(file_size)}       ")
        else:
            urllib.request.urlretrieve(file_url, save_to)
    elif SHOW_DEBUG:
        print("File exists, moving on")

    return save_to


# Resizes the image to the provided dimensions
def resize_image(filename):
    if SHOW_DEBUG:
        print("Opening local image")

    image = Image.open(filename)
    current_x, current_y = image.size
    if (current_x, current_y) == (DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y):
        if SHOW_DEBUG:
            print("Images are currently equal in size. No need to scale.")
    else:
        if SHOW_DEBUG:
            print("Resizing the image from", image.size[0], "x", image.size[1], "to", DEFAULT_RESOLUTION_X, "x", DEFAULT_RESOLUTION_Y)
        image = image.resize((DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y), Image.ANTIALIAS)

        if SHOW_DEBUG:
            print(f"Saving the image to {filename}")
        fhandle = open(filename, 'w')
        image.save(fhandle, 'PNG')


# Sets the new image as the wallpaper
def set_gnome_wallpaper(file_path):
    if SHOW_DEBUG:
        print("Setting the wallpaper")
    command = "gsettings set org.gnome.desktop.background picture-uri file://" + file_path
    status, output = subprocess.getstatusoutput(command)
    return status


def print_download_status(block_count, block_size, total_size):
    written_size = human_readable_size(block_count * block_size)
    total_size = human_readable_size(total_size)

    # Adding space padding at the end to ensure we overwrite the whole line
    stdout.write(f"\r{written_size} bytes of {total_size}         ")
    stdout.flush()


def human_readable_size(number_bytes):
    for x in ['bytes', 'KB', 'MB']:
        if number_bytes < 1024.0:
            return "%3.2f%s" % (number_bytes, x)
        number_bytes /= 1024.0


# Creates the necessary XML so background images will scroll through
def create_desktop_background_scoll(filename):
    if not IMAGE_SCROLL:
        return filename

    if SHOW_DEBUG:
        print("Creating XML file for desktop background switching.")

    filename = DOWNLOAD_PATH + '/nasa_apod_desktop_backgrounds.xml'

    # Create our base, background element
    background = etree.Element("background")

    # Grab our PNGs we have downloaded
    images = glob.glob(DOWNLOAD_PATH + "/*.png")
    num_images = len(images)

    if num_images < SEED_IMAGES:
        # Let's seed some images
        # Start with yesterday and continue going back until we have enough
        if SHOW_DEBUG:
            print("Downloading some seed images as well")
        days_back = 0
        seed_images_left = SEED_IMAGES
        while seed_images_left > 0:
            days_back += 1
            if SHOW_DEBUG:
                print(f"Downloading seed image ({str(seed_images_left)} left):")
            day_to_try = datetime.now() - timedelta(days=days_back)

            # Filenames look like /apYYMMDD.html
            seed_filename = NASA_APOD_SITE + "ap" + day_to_try.strftime("%y%m%d") + ".html"
            seed_site_contents = download_site(seed_filename)

            # Make sure we didn't encounter an error for some reason
            if seed_site_contents == "error":
                continue

            seed_filename = get_image(seed_site_contents)
            # If the content was an video or some other error occurred, skip the
            # rest.
            if seed_filename is None:
                continue

            resize_image(seed_filename)

            # Add this to our list of images
            images.append(seed_filename)
            seed_images_left -= 1
        if SHOW_DEBUG:
            print("Done downloading seed images")

    # Get our images in a random order so we get a new order every time we get a new file
    random.shuffle(images)
    # Recalculate the number of pictures
    num_images = len(images)

    for i, image in enumerate(images):
        # Create a static entry for keeping this image here for IMAGE_DURATION
        static = etree.SubElement(background, "static")

        # Length of time the background stays
        duration = etree.SubElement(static, "duration")
        duration.text = str(IMAGE_DURATION)

        # Assign the name of the file for our static entry
        static_file = etree.SubElement(static, "file")
        static_file.text = images[i]

        # Create a transition for the animation with a from and to
        transition = etree.SubElement(background, "transition")

        # Length of time for the switch animation
        transition_duration = etree.SubElement(transition, "duration")
        transition_duration.text = "5"

        # We are always transitioning from the current file
        transition_from = etree.SubElement(transition, "from")
        transition_from.text = images[i]

        # Create our tranition to element
        transition_to = etree.SubElement(transition, "to")

        # Check to see if we're at the end, if we are use the first image as the image to
        if i + 1 == num_images:
            transition_to.text = images[0]
        else:
            transition_to.text = images[i + 1]

    xml_tree = etree.ElementTree(background)
    xml_tree.write(filename, pretty_print=True)

    return filename


def get_image_info(element, text):
    # Grabs information about the image
    regex = '<' + element + '="(image.*?)"'
    reg = re.search(regex, text, re.IGNORECASE)
    if reg:
        if 'http' in reg.group(1):
            # Actual URL
            file_url = reg.group(1)
        else:
            # Relative path, handle it
            file_url = NASA_APOD_SITE + reg.group(1)
    else:
        if SHOW_DEBUG:
            print("Could not find an image. May be a video today.")
        return None, None, None

    # Create our handle for our remote file
    if SHOW_DEBUG:
        print("Opening remote URL")

    remote_file = urllib.request.urlopen(file_url)

    filename = os.path.basename(file_url)
    file_size = float(remote_file.headers.get("content-length"))

    return file_url, filename, file_size


if __name__ == '__main__':
    if SHOW_DEBUG:
        print("Starting")

    # Find desktop resolution
    DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y = find_display_resolution()

    # Set a localized download folder
    DOWNLOAD_PATH = set_download_folder()

    # Create the download path if it doesn't exist
    if not os.path.exists(os.path.expanduser(DOWNLOAD_PATH)):
        os.makedirs(os.path.expanduser(DOWNLOAD_PATH))

    # Grab the HTML contents of the file
    site_contents = download_site(NASA_APOD_SITE)
    if site_contents == "error":
        if SHOW_DEBUG:
            print("Could not contact site.")
        exit()

    # Download the image
    filename = get_image(site_contents)
    if filename is not None:
        # Resize the image
        resize_image(filename)

    # Create the desktop switching xml
    filename = create_desktop_background_scoll(filename)
    # If the script was unable todays image and IMAGE_SCROLL is set to False,
    # the script exits
    if filename is None:
        if SHOW_DEBUG:
            print("Today's image could not be downloaded.")
        exit()

    # Set the wallpaper
    status = set_gnome_wallpaper(filename)
    if SHOW_DEBUG:
        print("Finished!")
Gruß
Atalanttore
Sirius3
User
Beiträge: 11982
Registriert: Sonntag 21. Oktober 2012, 17:20

Samstag 22. Juni 2019, 20:45

Ein ganzes Paket mit all seinen Abhängigkeiten nur um einen Pfad zu ermitteln? Das ist doch wohl etwas übertrieben.
Statt SHOW_DEBUG und den ganzen if-print-Blöcken solltest Du Dir logging anschauen.
Statt grep als externes Programm aufzurufen würde man wohl auch eher Pythons re-Modul verwenden; achso, tust Du ja sowieso.
`finditer` liefert einen Iterator, der immer „wahr" ist.

Bei dem Namen `set_download_folder` würde ich erwarten, dass etwas gesetzt wird, nicht dass etwas ermittelt wird.
In `resize_image` solltest Du Dateien im Binärmodus öffnen und auch wieder schließen.

In `set_gnome_wallpaper`: Statt subprocess.getstatusoutput solltest Du etwas wie subprocess.run benutzen, vor allem, wenn Du unkontrolliert Daten aus externen Quellen an eine Shell übergibst.

`human_readable_size` gibt bei Größen größer 1073741824 None zurück.
In `create_desktop_background_scoll`: Pfade setzt man nicht mit + zusammen.
Ein str-Aufruf in einem Format-String ist unnötig.

HTML bearbeitet man nicht mit regulären Asudrücken, sondern nutzt einen HTML-Parser.
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Samstag 22. Juni 2019, 20:52

@Sirius3: Danke für die Empfehlungen. Damit werde ich einige Zeit beschäftigt sein.

Wie ermittelt man den Download-Pfad des Nutzers ohne dafür ein ganzes Paket einzubinden?

Gruß
Atalanttore
Benutzeravatar
__blackjack__
User
Beiträge: 5999
Registriert: Samstag 2. Juni 2018, 10:21
Wohnort: 127.0.0.1
Kontaktdaten:

Samstag 22. Juni 2019, 21:45

Also ich würde da schon ein ganzes Paket einbinden, denn zumindest der C-Teil muss sowieso da sein, da `libglib` eine Abhängigkeit von `libgtk` ist. Das heisst ausser der Python-Anbindung an `libglib` sind sowieso schon alle Abhängigkeiten installiert. Das einzige was da fehlt ist ein Python-Modul das `libglib` wrapped, beziehungsweise für den Gtk3/`gi`-Ansatz die Datendatei mit der Beschreibung für die „GObject Introspection“.
long long ago; /* in a galaxy far far away */
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Sonntag 23. Juni 2019, 14:10

@__blackjack__: Wie bindet man dieses Paket ein?

Gruß
Atalanttore
Benutzeravatar
__blackjack__
User
Beiträge: 5999
Registriert: Samstag 2. Juni 2018, 10:21
Wohnort: 127.0.0.1
Kontaktdaten:

Sonntag 23. Juni 2019, 14:38

@Atalanttore: Man installiert das entsprechende Python-Paket, beziehungsweise die entsprechende Datei mit den Metadaten damit das `gi.repository`-Modul ein passendes Modul-Objekt bereitstellen kann. Für Python 2 ist bei mir unter Ubuntu das `glib`-Modul in `python-gobject-2` enthalten. Für Python 3 gibt's das nicht mehr (zumindest unter Ubuntu/Debian) da muss man also über `gi` gehen und die Introspection-Metadaten aus dem `gir1.2-glib-2.0`-Debianpaket installieren. Damit bekommt man von Python aus Zugriff auf `GLib`, `GObject`, `GModule`, und `Gio`.
long long ago; /* in a galaxy far far away */
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Sonntag 23. Juni 2019, 15:01

@__blackjack__: Gibt es noch eine andere Möglichkeit, wie man mit Python 3 den Download-Pfad des Nutzers ermitteln kann?

Gruß
Atalanttore
Sirius3
User
Beiträge: 11982
Registriert: Sonntag 21. Oktober 2012, 17:20

Sonntag 23. Juni 2019, 15:05

Für so ein Hintergrunzlade-Programm würde ich sowieso einen eigenen Pfad für die Bilder nehmen, und nicht den allgemeinen Download-Ordner.
Benutzeravatar
__blackjack__
User
Beiträge: 5999
Registriert: Samstag 2. Juni 2018, 10:21
Wohnort: 127.0.0.1
Kontaktdaten:

Sonntag 23. Juni 2019, 15:19

@Sirius3: Es wird ja ein Pfad in diesem Ordner angelegt. Man könnte auch `G_USER_DIRECTORY_PICTURES` nehmen und dort den Unterordner anlegen.
long long ago; /* in a galaxy far far away */
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Samstag 29. Juni 2019, 21:18

Bei Qt stellt die Klasse `QStandardPaths` entsprechende Methoden zur Abfrage von bestimmten Nutzerverzeichnissen bereitet. Allerdings frage ich mich, ob die Einbindung von Qt bei einem Hintergrund-Download-Skript nicht etwas übertrieben ist oder nicht?

Ein paar der Verbesserungsvorschläge von Sirius3 habe ich schon umgesetzt. Momentan hänge ich beim Speichern des heruntergeladenen Bildes fest. Es erscheint eine Fehlermeldung, die einen falschen Datentyp (byte statt string) bemängelt:

Code: Alles auswählen

2019-06-29 22:02:34,140 __main__: Starting
2019-06-29 22:02:34,140 __main__: Attempting to determine the current resolution.
2019-06-29 22:02:34,239 __main__: Using detected resolution of 3840x1080
2019-06-29 22:02:34,240 __main__: Could not determine download folder with GLib. Using default.
2019-06-29 22:02:34,240 __main__: Downloading contents of the site to find the image name
2019-06-29 22:02:34,897 __main__: Grabbing the image URL
2019-06-29 22:02:34,898 __main__: Opening remote URL
2019-06-29 22:02:35,537 __main__: Found name of image: gendlerM83-New-HST-ESO-LL.jpg
2019-06-29 22:02:39,368 __main__: Opening local image
--- Logging error ---
Traceback (most recent call last):
  File "/usr/lib/python3.6/logging/__init__.py", line 994, in emit
    msg = self.format(record)
  File "/usr/lib/python3.6/logging/__init__.py", line 840, in format
    return fmt.format(record)
  File "/usr/lib/python3.6/logging/__init__.py", line 577, in format
    record.message = record.getMessage()
  File "/usr/lib/python3.6/logging/__init__.py", line 338, in getMessage
    msg = msg % self.args
TypeError: not all arguments converted during string formatting
Call stack:
  File "/home/ata/PycharmProjects/nasa-apod-desktop/nasa_apod_desktop.py", line 373, in <module>
    resize_image(filename)
  File "/home/ata/PycharmProjects/nasa-apod-desktop/nasa_apod_desktop.py", line 225, in resize_image
    logger.info("Resizing the image from", image.size[0], "x", image.size[1], "to", DEFAULT_RESOLUTION_X, "x", DEFAULT_RESOLUTION_Y)
Message: 'Resizing the image from'
Arguments: (3500, 'x', 2846, 'to', 3840, 'x', 1080)
2019-06-29 22:02:39,785 __main__: Saving the image as /tmp/backgrounds/gendlerM83-New-HST-ESO-LL.png
Traceback (most recent call last):
  File "/home/ata/PycharmProjects/nasa-apod-desktop/nasa_apod_desktop.py", line 373, in <module>
    resize_image(filename)
  File "/home/ata/PycharmProjects/nasa-apod-desktop/nasa_apod_desktop.py", line 231, in resize_image
    image.save(file_handler, 'PNG')
  File "/usr/lib/python3/dist-packages/PIL/Image.py", line 1935, in save
    save_handler(self, fp, filename)
  File "/usr/lib/python3/dist-packages/PIL/PngImagePlugin.py", line 732, in _save
    fp.write(_MAGIC)
TypeError: write() argument must be str, not bytes

Aktueller Code:

Code: Alles auswählen

#!/usr/bin/env python
# 
# Copyright (c) 2012 David Drake
# 
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# 
#    http://www.apache.org/licenses/LICENSE-2.0
# 
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# 
# 
# nasa_apod_desktop.py
# https://github.com/randomdrake/nasa-apod-desktop
# 
# Written/Modified by David Drake
# http://randomdrake.com 
# http://twitter.com/randomdrake

import logging

#import glib
import subprocess
import urllib.request, urllib.parse, urllib.error
import re
import os
import random
import glob
from PIL import Image
from sys import stdout
from sys import exit
from lxml import etree
from datetime import datetime, timedelta

NASA_APOD_SITE = 'http://apod.nasa.gov/apod/'
TEMPORARY_DOWNLOAD_PATH = '/tmp/backgrounds/'
CUSTOM_FOLDER = 'nasa-apod-backgrounds'

RESOLUTION_TYPE = 'stretch'
DEFAULT_RESOLUTION_X = 1024
DEFAULT_RESOLUTION_Y = 768

IMAGE_SCROLL = True
IMAGE_DURATION = 1200
SEED_IMAGES = 10
SHOW_DEBUG = False

LOG_LEVEL = logging.DEBUG
LOG_FORMAT = '%(asctime)s %(name)s: %(message)s'

logger = logging.getLogger(__name__)
logger.setLevel(LOG_LEVEL)

formatter = logging.Formatter(LOG_FORMAT)

stream_handler = logging.StreamHandler()
stream_handler.setFormatter(formatter)

logger.addHandler(stream_handler)


# Use XRandR to grab the desktop resolution. If the scaling method is set to 'largest',
# we will attempt to grab it from the largest connected device. If the scaling method
# is set to 'stretch' we will grab it from the current value. Default will simply use
# what was set for the default resolutions.
def find_display_resolution():
    if RESOLUTION_TYPE == 'default':
        logger.info(f"Using default resolution of {DEFAULT_RESOLUTION_X}x{DEFAULT_RESOLUTION_Y}")
        return DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y

    resolution_x = 0
    resolution_y = 0

    logger.info("Attempting to determine the current resolution.")
    if RESOLUTION_TYPE == 'largest':
        regex_search = 'connected'
    else:
        regex_search = 'current'

    p1 = subprocess.Popen(["xrandr"], stdout=subprocess.PIPE)
    p2 = subprocess.Popen(["grep", regex_search], stdin=p1.stdout, stdout=subprocess.PIPE)  # TODO: Pythons re-Modul verwenden
    p1.stdout.close()
    output = str(p2.communicate()[0])

    if RESOLUTION_TYPE == 'largest':
        # We are going to go through the connected devices and get the X/Y from the largest
        matches = re.finditer(" connected ([0-9]+)x([0-9]+)+", output)  # TODO: liefert einen Iterator, der immer „wahr" ist.
        if matches:
            largest = 0
            for match in matches:
                if int(match.group(1)) * int(match.group(2)) > largest:
                    resolution_x = match.group(1)
                    resolution_y = match.group(2)
        else:
            logger.warning("Could not determine largest screen resolution.")

    else:
        reg = re.search(".* current (.*?) x (.*?),.*", output)
        if reg:
            resolution_x = reg.group(1)
            resolution_y = reg.group(2)
        else:
            logger.warning("Could not determine current screen resolution.")

    # If we couldn't find anything automatically use what was set for the defaults
    if resolution_x == 0 or resolution_y == 0:
        resolution_x = DEFAULT_RESOLUTION_X
        resolution_y = DEFAULT_RESOLUTION_Y
        logger.warning("Could not determine resolution automatically. Using defaults.")

    logger.info(f"Using detected resolution of {resolution_x}x{resolution_y}")

    return int(resolution_x), int(resolution_y)


# Uses GLib to find the localized "Downloads" folder
# See: http://askubuntu.com/questions/137896/how-to-get-the-user-downloads-folder-location-with-python
def get_user_download_directory():
    # downloads_dir = glib.get_user_special_dir(glib.USER_DIRECTORY_DOWNLOAD)  # TODO
    downloads_dir = False  # Workaround

    if downloads_dir:
        # Add any custom folder
        new_path = os.path.join(downloads_dir, CUSTOM_FOLDER)
        logger.info("Using automatically detected path:", new_path)
    else:
        new_path = TEMPORARY_DOWNLOAD_PATH
        logger.warning("Could not determine download folder with GLib. Using default.")
    return new_path


# Download HTML of the site
def download_site(url):
    logger.info("Downloading contents of the site to find the image name")
    opener = urllib.request.build_opener()
    req = urllib.request.Request(url)
    try:
        response = opener.open(req)
        reply = response.read().decode()
    except urllib.error.HTTPError as error:
        logger.error(f"Error downloading {url} - {error.code}")
        reply = f"Error: {error.code})"
    return reply


# Finds the image URL and saves it
def get_image(text):
    logger.info("Grabbing the image URL")
    file_url, filename, file_size = get_image_info('a href', text)
    # If file_url is None, the today's picture might be a video
    if file_url is None:
        return None

    logger.info(f"Found name of image: {filename}")

    save_to = os.path.join(TEMPORARY_DOWNLOAD_PATH, os.path.splitext(filename)[0] + '.png')

    if not os.path.isfile(save_to):
        # If the response body is less than 500 bytes, something went wrong
        if file_size < 500:
            print("Response less than 500 bytes, probably an error\nAttempting to just grab image source")
            file_url, filename, file_size = get_image_info('img src', text)
            # If file_url is None, the today's picture might be a video
            if file_url is None:
                return None
            print(f"Found name of image: {filename}")
            if file_size < 500:
                # Give up
                logger.error("Could not find image to download")
                exit()

            logger.info("Retrieving image")
            urllib.request.urlretrieve(file_url, save_to, print_download_status)

            # Adding additional padding to ensure entire line 
            logger.info(f"\rDone downloading {human_readable_size(file_size)}       ")
        else:
            urllib.request.urlretrieve(file_url, save_to)
    else:
        logger.info("File exists, moving on")

    return save_to


def get_image_info(element, text):  # TODO: HTML-Parser anstatt regulären Ausdrücken verwenden
    # Grabs information about the image
    regex = '<' + element + '="(image.*?)"'
    reg = re.search(regex, text, re.IGNORECASE)
    if reg:
        if 'http' in reg.group(1):
            # Actual URL
            file_url = reg.group(1)
        else:
            # Relative path, handle it
            file_url = NASA_APOD_SITE + reg.group(1)
    else:
        logger.warning("Could not find an image. May be a video today.")
        return None, None, None

    # Create our handle for our remote file
    logger.info("Opening remote URL")

    remote_file = urllib.request.urlopen(file_url)

    filename = os.path.basename(file_url)
    file_size = float(remote_file.headers.get("content-length"))

    return file_url, filename, file_size


# Resizes the image to the provided dimensions
def resize_image(filename):
    logger.info("Opening local image")

    image = Image.open(filename)
    current_x, current_y = image.size
    if (current_x, current_y) == (DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y):
        logger.info("Images are currently equal in size. No need to scale.")
    else:
        logger.info("Resizing the image from", image.size[0], "x", image.size[1], "to", DEFAULT_RESOLUTION_X, "x", DEFAULT_RESOLUTION_Y)
        image = image.resize((DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y), Image.ANTIALIAS)

        logger.info(f"Saving the image as {filename}")

        file_handler = open(filename, 'w')
        image.save(file_handler, 'PNG')
        file_handler.close()


# Sets the new image as the wallpaper
def set_gnome_wallpaper(file_path):
    logger.info("Setting the wallpaper")
    command = "gsettings set org.gnome.desktop.background picture-uri file://" + file_path
    status, output = subprocess.getstatusoutput(command)  # TODO: Statt subprocess.getstatusoutput etwas wie subprocess.run benutzen
    return status


def print_download_status(block_count, block_size, total_size):
    written_size = human_readable_size(block_count * block_size)
    total_size = human_readable_size(total_size)

    # Adding space padding at the end to ensure we overwrite the whole line
    stdout.write(f"\r{written_size} bytes of {total_size}         ")
    stdout.flush()


def human_readable_size(number_bytes):  # TODO: gibt bei Größen größer 1073741824 None zurück.
    for x in ['bytes', 'KB', 'MB']:
        if number_bytes < 1024.0:
            return "%3.2f%s" % (number_bytes, x)
        number_bytes /= 1024.0


# Creates the necessary XML so background images will scroll through
def create_desktop_background_scroll(filename):
    if not IMAGE_SCROLL:
        return filename

    logger.info("Creating XML file for desktop background switching.")

    filename = os.path.join(TEMPORARY_DOWNLOAD_PATH, '/nasa_apod_desktop_backgrounds.xml')

    # Create our base, background element
    background = etree.Element("background")

    # Grab our PNGs we have downloaded
    images = glob.glob(TEMPORARY_DOWNLOAD_PATH + "/*.png")
    num_images = len(images)

    if num_images < SEED_IMAGES:
        # Let's seed some images
        # Start with yesterday and continue going back until we have enough
        logger.info("Downloading some seed images as well")
        days_back = 0
        seed_images_left = SEED_IMAGES
        while seed_images_left > 0:
            days_back += 1
            logger.info(f"Downloading seed image ({seed_images_left} left):")
            day_to_try = datetime.now() - timedelta(days=days_back)

            # Filenames look like /apYYMMDD.html
            seed_filename = os.path.join(NASA_APOD_SITE, "ap", day_to_try.strftime("%y%m%d") + ".html")
            seed_site_contents = download_site(seed_filename)

            # Make sure we didn't encounter an error for some reason
            if seed_site_contents == "error":
                continue

            seed_filename = get_image(seed_site_contents)
            # If the content was an video or some other error occurred, skip the
            # rest.
            if seed_filename is None:
                continue

            resize_image(seed_filename)

            # Add this to our list of images
            images.append(seed_filename)
            seed_images_left -= 1
        logger.info("Done downloading seed images")

    # Get our images in a random order so we get a new order every time we get a new file
    random.shuffle(images)
    # Recalculate the number of pictures
    num_images = len(images)

    for i, image in enumerate(images):
        # Create a static entry for keeping this image here for IMAGE_DURATION
        static = etree.SubElement(background, "static")

        # Length of time the background stays
        duration = etree.SubElement(static, "duration")
        duration.text = str(IMAGE_DURATION)

        # Assign the name of the file for our static entry
        static_file = etree.SubElement(static, "file")
        static_file.text = images[i]

        # Create a transition for the animation with a from and to
        transition = etree.SubElement(background, "transition")

        # Length of time for the switch animation
        transition_duration = etree.SubElement(transition, "duration")
        transition_duration.text = "5"

        # We are always transitioning from the current file
        transition_from = etree.SubElement(transition, "from")
        transition_from.text = images[i]

        # Create our tranition to element
        transition_to = etree.SubElement(transition, "to")

        # Check to see if we're at the end, if we are use the first image as the image to
        if i + 1 == num_images:
            transition_to.text = images[0]
        else:
            transition_to.text = images[i + 1]

    xml_tree = etree.ElementTree(background)
    xml_tree.write(filename, pretty_print=True)

    return filename


if __name__ == '__main__':
    logger.info("Starting")

    # Find desktop resolution
    DEFAULT_RESOLUTION_X, DEFAULT_RESOLUTION_Y = find_display_resolution()

    # Set a localized download folder
    TEMPORARY_DOWNLOAD_PATH = get_user_download_directory()

    # Create the download path if it doesn't exist
    if not os.path.exists(os.path.expanduser(TEMPORARY_DOWNLOAD_PATH)):
        os.makedirs(os.path.expanduser(TEMPORARY_DOWNLOAD_PATH))

    # Grab the HTML contents of the file
    site_contents = download_site(NASA_APOD_SITE)
    if site_contents == "error":
        logger.error("Could not contact site.")
        exit()

    # Download the image
    filename = get_image(site_contents)
    if filename is not None:
        # Resize the image
        resize_image(filename)

    # Create the desktop switching xml
    filename = create_desktop_background_scroll(filename)
    # If the script was unable todays image and IMAGE_SCROLL is set to False,
    # the script exits
    if filename is None:
        logger.error("Today's image could not be downloaded.")
        exit()

    # Set the wallpaper
    status = set_gnome_wallpaper(filename)
    logger.info("Finished!")
Gruß
Atalanttore
Sirius3
User
Beiträge: 11982
Registriert: Sonntag 21. Oktober 2012, 17:20

Samstag 29. Juni 2019, 22:20

Wenn Du binär-Dateien schreiben willst, mußt Du sie auch im Binärmodus öffnen. Dateien öffnet man auch mit dem with-Statement. Hier wäre das einfachste, ›image.save‹ gleich den Dateinamen zu übergeben.
Benutzeravatar
__blackjack__
User
Beiträge: 5999
Registriert: Samstag 2. Juni 2018, 10:21
Wohnort: 127.0.0.1
Kontaktdaten:

Samstag 29. Juni 2019, 22:37

@Atalanttore: Warum nicht `glib`? Das ist ja ein Skript um die Hintergrundbilder unter GNOME zu setzen. Da ist `libglib` auf jeden Fall installiert, denn das ist ja Teil/Grundlage von Gtk und GNOME benutzt Gtk. Das `glib`-Funktion die das Skript benutzt ist das ”Gegenstück” zu der Qt-Klasse. Dafür Qt zu verwenden wäre irgendwie schräg.
long long ago; /* in a galaxy far far away */
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Samstag 6. Juli 2019, 20:19

Mit dem Code unten wird die URL zur hochauflösenden Version des Bildes ohne reguläre Ausdrücke extrahiert. Der Code setzt aber voraus, dass die URL zum Bild der 2. Link (Index 1) auf der Webseite ist.

Kann man den Code, ohne reguläre Ausdrücke zu verwenden, robuster gegen Änderungen auf der Webseite machen, wenn die URL zum hochauflösenden Bild mal nicht der 2. Link auf der Webseite ist?

Code: Alles auswählen

from bs4 import BeautifulSoup
import requests

url = "https://apod.nasa.gov/apod/ap190629.html"

response = requests.get(url)
html_source = response.text
soup = BeautifulSoup(html_source, 'lxml')
tags = soup.find_all('a')

urls = []

for tag in tags:
    urls.append(tag.get('href'))

print(urls[1])
Gruß
Atalanttore
Benutzeravatar
__blackjack__
User
Beiträge: 5999
Registriert: Samstag 2. Juni 2018, 10:21
Wohnort: 127.0.0.1
Kontaktdaten:

Samstag 6. Juli 2019, 22:54

@Atalanttore: Du suchst einen Link (<a>) zu einer Bilddatei (href="….jpg") das als Kindelement ein <img> enthält.
long long ago; /* in a galaxy far far away */
Atalanttore
User
Beiträge: 407
Registriert: Freitag 6. August 2010, 17:03

Sonntag 7. Juli 2019, 19:28

@__blackjack__: Danke. "Kindelement <img>" war der richtige Hinweis.

Der Code sieht nun so aus:

Code: Alles auswählen

from bs4 import BeautifulSoup
import requests
import os

url = "https://apod.nasa.gov/apod/ap190629.html"

NASA_APOD_SITE = 'http://apod.nasa.gov/apod/'

response = requests.get(url)
html_source = response.text
soup = BeautifulSoup(html_source, 'lxml')
tags = soup.find_all('a')

for tag in tags:
    if tag.findChild("img"):
        image_path = os.path.join(NASA_APOD_SITE, tag.get('href'))
        print(image_path)
Gruß
Atalanttore
Antworten