I made a mistake. All disk operations are assuming the files are on the current filesystem. They will not work for remote systems. I'm not sure how to make it possible for remote as well.

I guess the best thing to do right now is to abort disk operation when the remote is not localhost and print a warning.

Secondly, a workaround to run disk operations on remote is to use them in callbacks on the remote server, with the aria2p listen -c /path/to/callbacks.py functionality. Run the aria2p listen command in a background process (&, maybe followed by disown, or with nohup).

Eventually, maybe we could implement a way to communicate between different aria2p instances, to be able to run disk operations on remotes.

I was thinking of something like:

# get global option
aria2p getopt max-concurrent-downloads

# get downloads options
aria2p getopt dir 390c3d203c46ae0a

# set global option
aria2p setopt max-concurrent-downloads=10

# increment/decrement global integer option
aria2p setopt max-concurrent-downloads=+4
# or +=4? maybe support both. Should be easy as this is a positional argument

# set downloads options
aria2p setopt allow-overwrite=true 390c3d203c46ae0a 2089b05ecca3d829

We should be able to combine this with the "select" and/or "filter" flags of #22.

This issue also requires #14 to be implemented.

At least in the high-level classes: API, Download, File, BitTorrent. Maybe in Options and Client as well. It can be done progressively (one class after the other).

Not sure how this one would be implemented, or if it would even be useful.

The JSON-RPC interface does not support notifications over HTTP, but the RPC server will send notifications over Web‐Socket.

Currently we can write shell or Python script that use aria2p and hook them to aria2c.

Add commands to the CLI or options to existing commands:

• move-files
• copy-files
• save-session
• shutdown

Maybe see API class for more.

To start we could simply add the solutions found in the "question"-labeled issues. Eventually I'd like to put examples of complex callbacks in there, like a callbacks-based way to implement markers (#19)

It's because we try to "remove" it instead of "purging" it. See #30.

$ aria2p top
aria2p: error: argument : invalid choice: 'top' (choose from 'show', 'call', 'add-magnet', 'add-torrent', 'add-metalink', 'pause', 'pause-all', 'resume', 'resume-all', 'remove', 'rm', 'remove-all', 'purge', 'clear', 'autopurge', 'autoclear')

version 0.71

no top comand ?

I tried with AppVeyor, but:

1. got failures, preventing aria2c process to terminate, triggering "resources quota exceeded" error (had to send a mail to support to know what it was: incoming traffic > 7GB)
2. failed to get logs artifacts since build errored
3. appveyor refused to let me re-build, trigger a new build, or even delete the project after I forced-push...
4. at this point I gave up. Unlinked my GitHub account and signed out.

If someone is willing to setup continuous integration for Windows, PRs are welcome.

We should be able to change the output directory of a download, even if it was already started and active.

Right now the queue is a simple list. Changes are not immediately reflected, one has to call again api.get_downloads() to synchronize the list. Maybe we would implement some self-synchronizing structure.

Although the README.md is comprehensive enough, I did run into a problem when first using aria2p as a library due to not understanding the nature of aria2p. So I added some further explanations in README.md.

I tried to follow CONTRIBUTING.md and file a PR, but poetry keeps acting wired in my docker container. Anyway, a simple update in README.md can be done through an issue, I think :-|

My changes in scripts/templates/README.md

...and mypy checks:

• in the Makefile (new rule + set it as a dependency to check rule)
• in CI (as a job in .gitlab-ci.yml)

We'll need to poetry add -D mypy, add the CI job (allowing failure until it passes) then progressively add type hints.

This would require a thread-safe data file somewhere, and a configuration file I guess.

It would be great if we could change options directly from within the interface.

The options would be temporary to start, only applied during the time the process is alive. Maybe later we could add an option to "save" them inside of the aria2 config file.

It could be nice to run commands directly from within the interface, like in vim/tmux/etc.

User would hit : and a prompt would appear, with auto-completion, etc.

The main use case would be to change options, but we can imagine more use-cases.

1. add a parameter update_interval to init method of class interface.Interface. This value is used to change the sleep value: self.sleep = update_interval / self.frames
   • Example with 1 second interval (default): sleep = 1 / 200 = 0.005
   • Example with 2 seconds interval: sleep = 2 / 200 = 0.01
2. add an option to the top subparser of the main CLI parser (in cli.get_parser function). Something like "-i", "--update-interval", type=float, default=1.0
3. write tests in tests/test_interface.py
4. update documentation

Currently we can only add magnets, metalinks, or torrents. We should provide a command to add URIs.

Would it be feasible to make the aria2p.downloads.Download class dynamic?

I.e. being able to return the current status of it's properties, not just the status when the object was first instantiated?

Below is an example of what I mean.

Using aria2p as a module, you can begin a bittorrent download via

import aria2p

# First initializing the API
api = aria2p.API(
            aria2p.Client(
            host="http://localhost",
            port=6800,
            secret=""
            )
# Then adding the magnetlink
download = api.add_magnet('some_magnet_link')

The download object returned here refers to the initial metadata download.
The download object returns 0% progress in perpetuity, even after the meta download has finished and the main download has begun.

>>> download.progress_string()
0.00%
>>> download.is_complete
False

To get the current download status you can:

# Get the GID for the download
>>> meta_gid = download.gid
# Then reinstantiate the download object for a more up-to-date status
>>> download = api.get_download(meta_gid)
>>> download.progress_string()
50%
>>> download = api.get_download(meta_gid)
>>> download.progress_string()
100%

However this is obviously less clean than just having to call the original download object.

Global only options:

• bt-max-open-files
• download-result
• keep-unfinished-download-result
• log
• log-level
• max-concurrent-downloads
• max-download-result
• max-overall-download-limit
• max-overall-upload-limit
• optimize-concurrent-downloads
• save-cookies
• save-session
• server-stat-of

Gloab/specific options:

• all-proxy
• all-proxy-passwd
• all-proxy-user
• allow-overwrite
• allow-piece-length-change
• always-resume
• async-dns
• auto-file-renaming
• bt-enable-hook-after-hash-check
• bt-enable-lpd
• bt-exclude-tracker
• bt-external-ip
• bt-force-encryption
• bt-hash-check-seed
• bt-load-saved-metadata
• bt-max-peers
• bt-metadata-only
• bt-min-crypto-level
• bt-prioritize-piece
• bt-remove-unselected-file
• bt-request-peer-speed-limit
• bt-require-crypto
• bt-save-metadata
• bt-seed-unverified
• bt-stop-timeout
• bt-tracker
• bt-tracker-connect-timeout
• bt-tracker-interval
• bt-tracker-timeout
• check-integrity
• conditional-get
• connect-timeout
• content-disposition-default-utf8
• continue
• dir
• dry-run
• enable-http-keep-alive
• enable-http-pipelining
• enable-mmap
• enable-peer-exchange
• file-allocation
• follow-metalink
• follow-torrent
• force-save
• ftp-passwd
• ftp-pasv
• ftp-proxy
• ftp-proxy-passwd
• ftp-proxy-user
• ftp-reuse-connection
• ftp-type
• ftp-user
• gid
• hash-check-only
• header
• http-accept-gzip
• http-auth-challenge
• http-no-cache
• http-passwd
• http-proxy
• http-proxy-passwd
• http-proxy-user
• http-user
• https-proxy
• https-proxy-passwd
• https-proxy-user
• lowest-speed-limit
• max-connection-per-server
• max-download-limit
• max-file-not-found
• max-mmap-limit
• max-resume-failure-tries
• max-tries
• max-upload-limit
• metalink-base-uri
• metalink-enable-unique-protocol
• metalink-language
• metalink-location
• metalink-os
• metalink-preferred-protocol
• metalink-version
• min-split-size
• no-file-allocation-limit
• no-netrc
• no-proxy
• parameterized-uri
• pause-metadata
• piece-length
• proxy-method
• realtime-chunk-checksum
• referer
• remote-time
• remove-control-file
• retry-wait
• reuse-uri
• rpc-save-upload-metadata
• seed-ratio
• seed-time
• split
• ssh-host-key-md
• stream-piece-selector
• timeout
• uri-selector
• use-head
• user-agent

Specific only options:

• checksum
• index-out
• out
• pause
• select-file

It would be great to provide a HTOP-like interface for showing the current downloads and the progress. Global graph for dl/up speed, collapsable graphs for each dl, sortable columns, customization with config file (columns and their order, colors), support for mouse and keys, etc.

Not sure if this is wanted anymore. Maybe for completed metadata downloads only? Closing for now. Comment if you're interested in this.

-> already installed aria2c,aria2& aria2p

import aria2p
aria2 = aria2p.API(
aria2p.Client(
host="http://localhost",
port=6800,
secret=""
)
)
m=aria2.add_magnet("magnet:?xt=dghgbh.......")

output
[ERROR] HTTPConnectionPool(host='localhost', port=6800): Max retries exceeded with url: /jsonrpc (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x10a2fc2b0>: Failed to establish a new connection: [Errno 61] Connection refused',))

For Bash and Zsh, to start.

this code:
aria2.add_uris(uris=[src, ], options={'dir': download_path, 'download': {'name': file_name} })

the download_path is work, but download file name is not work.

Get inspiration from HTOP.

Example: average upload/download speed from recent history.

When I download a file, how do I specify the filename instead of using the original filename

Using pyinstaller to pack it up and release, which would be much easier to spread this great project!

Currently, if I call the pause_all method, it tries to pause all the downloads, including already finished or paused downloads. Hence it returns False even if a single download has finished, failed, already paused

When adding a download using the CLI, one should be able to specify the download(s) options and position directly from the command line.

Somewhat related to #29, seen use-case in #32.

This is really wierd. What happened to the URI?

aria2p add 'https://example.com/some/path/filename.ext' .
Traceback (most recent call last):
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "C:\Users\mohan\AppData\Local\Programs\Python\Python36-32\Scripts\aria2p.exe\__main__.py", line 9, in <module>
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\site-packages\aria2p\cli.py", line 108, in main
    return subcommands.get(subcommand)(api, **kwargs)
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\site-packages\aria2p\cli.py", line 403, in subcommand_add
    if path.exists():
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\pathlib.py", line 1334, in exists
    self.stat()
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\pathlib.py", line 1156, in stat
    return self._accessor.stat(self)
  File "c:\users\mohan\appdata\local\programs\python\python36-32\lib\pathlib.py", line 387, in wrapped
    return strfunc(str(pathobj), *args)
OSError: [WinError 123] The filename, directory name, or volume label syntax is incorrect: "'https:\\example.com\\some\\path\\filename.ext'"

The SVG in the README made my fan spin really hard and made the page overall irresponsive. Consider replacing it with an asciinema

The view could contain charts as well.

The other options is to make rows collapsible/expandable in the main view, but it will complicate the logic even more...

A metadata download object returned from the API module will not return the followed_by list, even after completion, even with download.live and download.update() methods, for example:

import time

magnetlink = 'magnet:?...'
download = api.add_magnet(magnetlink, options={'dir': '/tmp'})
while not download.live.is_complete:
    time.sleep(5)
print(download.followed_by)

Will always return []
However, the following will return the correct followed_by download objects:

import time

magnetlink = 'magnet:?...'
download = api.add_magnet(magnetlink, options={'dir': '/tmp'})
while not download.live.is_complete:
    time.sleep(5)
download = api.get_download(download.gid)
print(download.followed_by)

Is this expected behaviour? I was under the impression that the Download.live and Download.update() methods would ideally return the updated followed_by list.

If we want to add several magnets from a file, we have to use a shell loop:

while read -r magnet; do
  aria2p add-magnet "$magnet"
done < magnets_file

It would be easier if we were able to do this:

aria2p add-magnet $(cat magnets_file)
# maybe also support one multi-line string:
aria2p add-magnet "$(cat magnets_file)"

Or even provide a flag to load from a file:

aria2p add-magnet -f magnets_file

When I try to download something with HTTP (not torrent), I expect the Download.bittorrent to be None however this is never the case.
Here is Download._struct for https://speed.hetzner.de/100MB.bin

{'completedLength': '0', 'connections': '1', 'dir': '/home/shivam/downloads//816/', 'downloadSpeed': '0', 'files': [{'completedLength': '0', 'index': '1', 'length': '0', 'path': '', 'selected': 'true', 'uris': [{'status': 'used', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}, {'status': 'waiting', 'uri': 'https://speed.hetzner.de/100MB.bin'}]}], 'gid': 'be0fa29b8fbc0c96', 'numPieces': '0', 'pieceLength': '1048576', 'status': 'active', 'totalLength': '0', 'uploadLength': '0', 'uploadSpeed': '0'}

While here is the Download._struct when a torrent is added:

{'bittorrent': {'announceList': [['http://nyaa.tracker.wf:7777/announce'], ['udp://open.stealth.si:80/announce'], ['udp://tracker.opentrackr.org:1337/announce'], ['udp://tracker.coppersurfer.tk:6969/announce'], ['udp://exodus.desync.com:6969/announce']]}, 'completedLength': '0', 'connections': '0', 'dir': '/home/shivam/downloads//817/', 'downloadSpeed': '0', 'files': [{'completedLength': '0', 'index': '1', 'length': '0', 'path': '[METADATA][Anime Time] Princess Connect! ReDive - 1 [720p].mp4', 'selected': 'true', 'uris': []}], 'gid': 'df71125cecd63d13', 'infoHash': '985e7f658509dca05b6b1560a4213950b11b7373', 'numPieces': '0', 'numSeeders': '0', 'pieceLength': '16384', 'seeder': 'false', 'status': 'active', 'totalLength': '0', 'uploadLength': '0', 'uploadSpeed': '0'}

As asked in #16 by @mnabila, add a shortcut that reads the clipboard to add new downloads.

aria2c start up, no token daemon ok

$ aria2p call aria2.getVersion
{"enabledFeatures": ["Async DNS", "BitTorrent", "Firefox3 Cookie", "GZip", "HTTPS", "Message Digest", "Metalink", "XML-RPC"], "version": "1.33.1"}
`

aria2c start up, token daemon bug
$aria2p -s token call aria2.getVersion
Unauthorized

Display the F keys.

I would like to be able to "select" downloads based on filters and patterns and their fields, and then apply an action (both with API and with CLI). For example:

aria2p pause --filter status=waiting
aria2p resume --pattern "dimmu borgir"
aria2p resume --re-pattern '\.iso$'

...but also proper printing. Logs should be logs, not used as stdout/stderr. For example, deprecation warnings about *-all commands should be logged (level=warning) but also printed to stderr.

The documentation states 3.6 and above, yet the .toml file states "~3.6" which indicates 3.6.* but not 3.7

Found this out when trying to install the package on my system that is running python 3.7.

Are there any known issues on 3.7 or is it safe to change the required version to "^3.6"?

The list command lists all downloads, sorts them, limits the result, etc.
It can also be used to quickly see the information about specific downloads in the terminal: it accepts gids or filter patterns.

Maybe do this:

• By default only outputs downloads names.
• Option -l, --long for additional information (--format=long).
• Option -f, --format with choices and custom string.
• Option -s, --sort with choices and custom sort.
• Option -H, --human-readable.

I think it would make aria2p easier to understand and use. Currently we "remove" downloads and "purge" download results. I'd like to get rid of the "purge" methods and only use "remove". The "remove" methods would be smart enough to know if we have to internally purge or remove, based on the download object status.

I think it would simplify things, but maybe it's a not a good idea. Any thoughts, users of aria2p?

Sometimes network calls can block forever if no timeouts are specified. For example here

It would be good if a timeout could be passed to requests.

Maybe it could be specified globally for the client like Client(..., timeout=60) and then use requests.post(..., timeout=self.timeout).

1. Follow https://poetry.eustace.io/docs/pyproject/#extras to set asciimatics as optional. Users will then have to install aria2p with pip install aria2p[tui] if they want the interactive interface.
2. Once peterbrittain/asciimatics#95 is done, use the tui extra for asciimatics (to reduce number of dependencies).