onionshare/cli/onionshare_cli/web/send_base_mode.py

356 lines
12 KiB
Python
Raw Normal View History

# -*- coding: utf-8 -*-
"""
OnionShare | https://onionshare.org/
2022-01-16 19:15:49 -05:00
Copyright (C) 2014-2022 Micah Lee, et al. <micah@micahflee.com>
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 3 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, see <http://www.gnu.org/licenses/>.
"""
import os
import sys
import tempfile
import mimetypes
import gzip
2021-03-10 03:40:06 -05:00
from flask import Response, request
2020-12-31 12:39:31 -05:00
from unidecode import unidecode
from urllib.parse import quote
class SendBaseModeWeb:
"""
All of the web logic shared between share and website mode (modes where the user sends files)
"""
2019-10-13 00:01:25 -04:00
def __init__(self, common, web):
super(SendBaseModeWeb, self).__init__()
self.common = common
self.web = web
# Information about the file to be shared
self.is_zipped = False
self.download_filename = None
self.download_filesize = None
self.zip_writer = None
# Create a temporary dir to store gzip files in
self.gzip_tmp_dir = tempfile.TemporaryDirectory(dir=self.common.build_tmp_dir())
self.gzip_counter = 0
# If autostop_sharing, only allow one download at a time
self.download_in_progress = False
# This tracks the history id
self.cur_history_id = 0
# Whether or not we can send REQUEST_INDIVIDUAL_FILE_STARTED
# and maybe other events when requests come in to this mode
self.supports_file_requests = True
self.define_routes()
self.init()
def fix_windows_paths(self, path):
"""
If on Windows, replace backslashes with slashes
"""
if self.common.platform == "Windows":
return path.replace("\\", "/")
return path
def set_file_info(self, filenames, processed_size_callback=None):
"""
Build a data structure that describes the list of files
"""
# If there's just one folder, replace filenames with a list of files inside that folder
if len(filenames) == 1 and os.path.isdir(filenames[0]):
2019-10-13 00:01:25 -04:00
filenames = [
os.path.join(filenames[0], x) for x in os.listdir(filenames[0])
]
# Re-initialize
2019-10-13 00:01:25 -04:00
self.files = {} # Dictionary mapping file paths to filenames on disk
self.root_files = (
{}
) # This is only the root files and dirs, as opposed to all of them
self.cur_history_id = 0
2019-10-13 00:01:25 -04:00
self.file_info = {"files": [], "dirs": []}
self.gzip_individual_files = {}
self.init()
# Windows paths use backslashes, but website paths use forward slashes. We have to
# make sure we're stripping the correct type of slash
if self.common.platform == "Windows":
slash = "\\"
else:
slash = "/"
# Build the file list
for filename in filenames:
basename = os.path.basename(filename.rstrip(slash))
# If it's a filename, add it
if os.path.isfile(filename):
self.files[self.fix_windows_paths(basename)] = filename
self.root_files[self.fix_windows_paths(basename)] = filename
# If it's a directory, add it recursively
elif os.path.isdir(filename):
self.root_files[self.fix_windows_paths(basename)] = filename
for root, _, nested_filenames in os.walk(filename):
# Normalize the root path. So if the directory name is "/home/user/Documents/some_folder",
# and it has a nested folder foobar, the root is "/home/user/Documents/some_folder/foobar".
# The normalized_root should be "some_folder/foobar"
2019-10-13 00:01:25 -04:00
normalized_root = os.path.join(
basename, root[len(filename) :].lstrip(slash)
).rstrip(slash)
# Add the dir itself
self.files[self.fix_windows_paths(normalized_root)] = root
# Add the files in this dir
for nested_filename in nested_filenames:
2019-10-13 00:01:25 -04:00
self.files[
self.fix_windows_paths(
os.path.join(normalized_root, nested_filename)
)
2019-10-13 00:01:25 -04:00
] = os.path.join(root, nested_filename)
self.set_file_info_custom(filenames, processed_size_callback)
2019-10-13 00:01:25 -04:00
def directory_listing(self, filenames, path="", filesystem_path=None):
# Tell the GUI about the directory listing
history_id = self.cur_history_id
self.cur_history_id += 1
2019-10-13 00:01:25 -04:00
self.web.add_request(
self.web.REQUEST_INDIVIDUAL_FILE_STARTED,
f"/{path}",
2019-10-13 00:01:25 -04:00
{"id": history_id, "method": request.method, "status_code": 200},
)
breadcrumbs = [("", "/")]
parts = path.split("/")
if parts[-1] == "":
parts = parts[:-1]
2019-09-15 19:44:24 -04:00
for i in range(len(parts)):
breadcrumbs.append((parts[i], f"/{'/'.join(parts[0 : i + 1])}"))
2019-09-15 19:44:24 -04:00
breadcrumbs_leaf = breadcrumbs.pop()[0]
# If filesystem_path is None, this is the root directory listing
files, dirs = self.build_directory_listing(path, filenames, filesystem_path)
2021-05-13 04:14:33 -04:00
return self.directory_listing_template(
2019-10-13 00:01:25 -04:00
path, files, dirs, breadcrumbs, breadcrumbs_leaf
)
def build_directory_listing(self, path, filenames, filesystem_path):
files = []
dirs = []
for filename in filenames:
if filesystem_path:
this_filesystem_path = os.path.join(filesystem_path, filename)
else:
this_filesystem_path = self.files[filename]
is_dir = os.path.isdir(this_filesystem_path)
if is_dir:
dirs.append(
{"link": os.path.join(f"/{path}", filename), "basename": filename}
)
else:
size = os.path.getsize(this_filesystem_path)
size_human = self.common.human_readable_filesize(size)
files.append(
{
"link": os.path.join(f"/{path}", filename),
"basename": filename,
"size_human": size_human,
}
)
return files, dirs
def stream_individual_file(self, filesystem_path):
"""
Return a flask response that's streaming the download of an individual file, and gzip
compressing it if the browser supports it.
"""
use_gzip = self.should_use_gzip()
# gzip compress the individual file, if it hasn't already been compressed
if use_gzip:
if filesystem_path not in self.gzip_individual_files:
gzip_filename = os.path.join(
self.gzip_tmp_dir.name, str(self.gzip_counter)
)
self.gzip_counter += 1
self._gzip_compress(filesystem_path, gzip_filename, 6, None)
self.gzip_individual_files[filesystem_path] = gzip_filename
file_to_download = self.gzip_individual_files[filesystem_path]
filesize = os.path.getsize(self.gzip_individual_files[filesystem_path])
else:
file_to_download = filesystem_path
filesize = os.path.getsize(filesystem_path)
path = request.path
# Tell GUI the individual file started
history_id = self.cur_history_id
self.cur_history_id += 1
2019-10-13 00:01:25 -04:00
self.web.add_request(
self.web.REQUEST_INDIVIDUAL_FILE_STARTED,
path,
{"id": history_id, "filesize": filesize},
)
def generate():
chunk_size = 102400 # 100kb
2019-10-13 00:01:25 -04:00
fp = open(file_to_download, "rb")
done = False
while not done:
chunk = fp.read(chunk_size)
2019-10-13 00:01:25 -04:00
if chunk == b"":
done = True
else:
try:
yield chunk
# Tell GUI the progress
downloaded_bytes = fp.tell()
percent = (1.0 * downloaded_bytes / filesize) * 100
2019-10-13 00:01:25 -04:00
if (
not self.web.is_gui
or self.common.platform == "Linux"
or self.common.platform == "BSD"
):
sys.stdout.write(
2019-10-13 00:01:25 -04:00
"\r{0:s}, {1:.2f}% ".format(
self.common.human_readable_filesize(
downloaded_bytes
),
percent,
)
)
sys.stdout.flush()
2019-10-13 00:01:25 -04:00
self.web.add_request(
self.web.REQUEST_INDIVIDUAL_FILE_PROGRESS,
path,
{
"id": history_id,
"bytes": downloaded_bytes,
"filesize": filesize,
},
)
done = False
2021-04-29 20:13:05 -04:00
except Exception:
# Looks like the download was canceled
done = True
# Tell the GUI the individual file was canceled
2019-10-13 00:01:25 -04:00
self.web.add_request(
self.web.REQUEST_INDIVIDUAL_FILE_CANCELED,
path,
{"id": history_id},
)
fp.close()
2019-10-13 00:01:25 -04:00
if self.common.platform != "Darwin":
sys.stdout.write("\n")
basename = os.path.basename(filesystem_path)
r = Response(generate())
if use_gzip:
2019-10-13 00:01:25 -04:00
r.headers.set("Content-Encoding", "gzip")
r.headers.set("Content-Length", filesize)
2021-01-01 18:56:53 -05:00
filename_dict = {
2021-02-22 16:35:14 -05:00
"filename": unidecode(basename),
"filename*": "UTF-8''%s" % quote(basename),
2021-01-01 18:56:53 -05:00
}
r.headers.set("Content-Disposition", "inline", **filename_dict)
(content_type, _) = mimetypes.guess_type(basename, strict=False)
if content_type is not None:
2019-10-13 00:01:25 -04:00
r.headers.set("Content-Type", content_type)
return r
def should_use_gzip(self):
"""
Should we use gzip for this browser?
"""
2019-10-13 00:01:25 -04:00
return (not self.is_zipped) and (
"gzip" in request.headers.get("Accept-Encoding", "").lower()
)
2019-10-13 00:01:25 -04:00
def _gzip_compress(
self, input_filename, output_filename, level, processed_size_callback=None
):
"""
Compress a file with gzip, without loading the whole thing into memory
Thanks: https://stackoverflow.com/questions/27035296/python-how-to-gzip-a-large-text-file-without-memoryerror
"""
bytes_processed = 0
2019-10-13 00:01:25 -04:00
blocksize = 1 << 16 # 64kB
with open(input_filename, "rb") as input_file:
output_file = gzip.open(output_filename, "wb", level)
while True:
if processed_size_callback is not None:
processed_size_callback(bytes_processed)
block = input_file.read(blocksize)
if len(block) == 0:
break
output_file.write(block)
bytes_processed += blocksize
output_file.close()
def init(self):
"""
Inherited class will implement this
"""
pass
def define_routes(self):
"""
Inherited class will implement this
"""
pass
def directory_listing_template(self):
"""
Inherited class will implement this. It should call render_template and return
the response.
"""
pass
def set_file_info_custom(self, filenames, processed_size_callback):
"""
Inherited class will implement this.
"""
pass
2019-10-13 00:01:25 -04:00
def render_logic(self, path=""):
"""
Inherited class will implement this.
"""
pass