Forum Archive

I played around with the beta 1.6/soon-to-be AppStore 2.0 appex module, which provides a way to make application extensions in Pythonista for use on the share sheet in other apps. Pretty amazing stuff! See module documentation and included examples for details, also on how to make available on the share sheet.
Then I realised this could be a way to collect files on the web (share in Safari), but also in cloud apps (share in Dropbox), in external editors (share in Textastic should you prefer that to the Pythonista IDE), etc. Actually, Appex GetFromURL.py works pretty simple: get url by appex.get_url(), get data by urllib.urlopen().read(), store data by open().write(). The script's length is mainly caused by path manipulation, error handling, and reporting progress.
This probably has been done before, but you may find it helpful, so here it is. Comments are welcome.

Moved to my gist

Wrote more flexible version of your script: Open in Pythonista

My extension allows save into Pythonista from different sources:
- from app extensions:
- single files or entry directories, passed by file path
- single files or web-pages, passed by URL
- text data, passed as raw text (e.g. from Notes app)
- from Pythonista's Editor:
- text data of opened tab

To use this extension you also need to download and install (copy into the same folder) libs package (to be precise: misc and log modules) from the same Gitlab project.

I'll be glad to feedbacks =)

#!/usr/bin/env python2
# coding: utf-8
# Olaf, Dec 2015, Pythonista 1.6 beta

'''
Appex GetFromURL
Pythonista app extension for use on share sheet of other apps to import file from URL into Pythonista
The file is saved at HOME/DESTINATION without user interaction (no 'save as' dialog) unless duplicate
'''

from __future__ import print_function
try:
    import appex, console, contextlib, itertools, os, os.path, sys, time, urllib, urlparse
except ImportError:
    assert False, 'This script needs the appex module in Pythonista version > 1.5'

HOME, DESTINATION = 'Documents', '' # you can change DESTINATION to any name of your liking

@contextlib.contextmanager
def callfunctionafterwardswithsec(function):
    '''Context-manager that calls function with duration of with block (sec) after termination
    >>> def pr_sec(sec): print('Duration {:3.2} sec'.format(sec))
    >>> with callfunctionafterwardswithsec(pr_sec): pass
    Duration 0.0 sec'''

    start = time.clock()
    yield
    end = time.clock()
    function(end - start)

def left_subpath_upto(path, sentinel):
    '''Left part (subpath) of path upto and including sentinel
    >>> print(left_subpath_upto('a/b/c', 'b'))
    a/b'''

    while path:
        head, tail = os.path.split(path)
        if tail == sentinel:
            break
        path = head
    return path

def iter_pad(length, arg0, *args):
    '''Iterator to pad arguments (at least 1) to specified length by repetition of final argument
    >>> print(''.join(iter_pad(3, 'a', 'b')))
    abb'''

    args = (arg0,) + args
    return itertools.islice(itertools.chain(args, itertools.repeat(args[-1])), length)

def parse_into_paths(input_url, HOME=HOME, DESTINATION=DESTINATION):
    '''Parse input URL into paths tuple for further processing
    >>> parse_into_paths('http://test.org/x.py', DESTINATION='TEST') # doctest: +ELLIPSIS
    ('x.py', 'http://test.org', 'Documents/TEST', '/private/var/.../TEST', '/priv.../TEST/x.py', True)'''

    url_tuple = urlparse.urlparse(input_url)
    scheme, netloc, basename = url_tuple.scheme, url_tuple.netloc, os.path.basename(url_tuple.path)
    input_short = urlparse.urlunparse(iter_pad(len(url_tuple), scheme, netloc, ''))
    output_short = os.path.join(HOME, DESTINATION)
    output_dir = os.path.join(left_subpath_upto(sys.argv[0], HOME), DESTINATION)
    output_path = os.path.join(output_dir, basename)
    is_Python = os.path.splitext(basename)[1].lower() == '.py'
    return basename, input_short, output_short, output_dir, output_path, is_Python

def copy_url(input_url):
    '''Write a copy of the file at input_url to HOME/DESTINATION
    if the destination directory doesn't exist, it is created
    if the destination file already exists, the user can cancel or overwrite
    if it is a Python file, a comment line is added to log the origin'''

    basename, input_short, output_short, output_dir, output_path, is_Python = parse_into_paths(input_url)

    if not os.path.exists(output_dir):
        os.mkdir(output_dir)
        console.hud_alert('Created destination directory {}'.format(output_short))
    if os.path.exists(output_path):
        try:
            console.alert('Duplicate file',
                                        '{} already exists in {}'.format(basename, output_short),
                                        'Overwrite') # or Cancel
        except KeyboardInterrupt:
            return

    with contextlib.closing(urllib.urlopen(input_url)) as input:
        data = input.read()
        console.hud_alert('Got {} ({} chars) from {}'.format(basename, len(data), input_short))
    with open(output_path, 'wb') as output:
        output.write(data)
        console.hud_alert('Wrote {} to {}'.format(basename, output_short))

def main():
    '''App extension logic, with unit tests if run within Pythonista'''

    if appex.is_running_extension():
        if appex.get_url():
            copy_url(appex.get_url())
            appex.finish()
        else:
            console.hud_alert('No input URL found', 'error')
    else:
        console.hud_alert('This script must be run from the sharing extension', 'error')
        import doctest
        doctest.testmod()

if __name__ == '__main__':
    main()