Fetch v2rayN subscription and test node availability on Linux

Some wrapper scripts for v2ray-core and xray-core on Linux:

1. Fetch v2rayN subscription link and generate json config files.
2. Test node availability by accessing www.google.com via these nodes.
3. Conveniently switch to different nodes by specifying an index.

Overview

Some important files:

• v2ray-wrapper.cfg: configuration for v2ray-wrapper.sh.

• template.json: template for v2ray config json. This file, together with the generated config file (which only has outbounds), will be given to v2ray.

• v2ray-wrapper.sh: the main script file, which can fetch subscription, test node availability and select different nodes.

Usage

1. Download v2ray or xray binaries.

Extract into this directory, e.g. v2ray-4.44.0. It should contain the v2ray binary (or xray binary).

2. Config v2ray-wrapper.sh

Copy the sample-v2ray-wrapper.cfg to v2ray-wrapper.cfg and edit it.

# Will be sourced by bash
# Relative paths are relative to this cfg file

# v2rayN subscription link
URL='https://example.com/...'

# If not empty, override the default user agent
USERAGENT=''

# Template json file
TEMPLATE=template.json

# Destination of the generated config files
CFGDIR=example_dir

# The directory that contains the v2ray (or xray) binary
BINDIR=v2ray-4.44.0

# Number of concurrent instances when testing nodes
TESTJOBS=16

# Verbose mode. Comment out to disable it.
#VERBOSE=y

# Use xray instead of v2ray. You have to use xray for XTLS.
# Comment out to disable it.
#XRAY=y

# Generate a copy of the original config, where the server address is
# overridden by the TLS serverName. This may fix some connection issues.
# The copy will have '-tlsServ.json' suffix.
# After testing the config files (-t), if we find that the original config
# is working properly, then the '-tlsServ.json' copy will be deleted.
# Comment out to disable it.
#TLSNAMEASSERVER=y

3. Config template.json

Copy the sample-template.json to template.json. Modify it if necessary ( e.g. change local port, log level...).

4. Run v2ray-wrapper.sh

The main script file is v2ray-wrapper.sh. To use -t (test availability), you have to install GNU Parallel.

Usage:
    v2ray-wrapper.sh [-u] [-t] [-c] [-n] [<cfg>]

Update the v2ray config files, test the current config files, select
the config file to use with its index, and run v2ray.

    <cfg>   Config file for this script. Default is v2ray-wrapper.cfg.
            This file will be sourced, so you can even put your own
            scripts here.

Options:
    -u      Update the config files using the subscription link.
    -t      Test the current config files and remove the unusable ones.
            This is done by trying to access www.google.com via these
            nodes.
    -c      Let the user choose which config file to use (by index).
            If not specified, choose the config file used last time
            (a symlink named last.json).
    -n      Don't run v2ray in the end.

Note that for simplicity the order of the options is fixed (i.e. '-u -c' is
OK but '-c -u' is invalid). And combination (like '-uc') is not supported.

To have more control over fetching v2rayN subscription link, you can manually invoke v2ray-subscr.py (it's used internally in v2ray-wrapper.sh).

usage: v2ray-subscr.py [-h] [-v] [-n] [-o OUTPUT] [-a AGENT] [--tlsNameAsServer] url

Parse a v2rayN subscription link and generate json config files for v2ray into a
directory. The unused json files in that directory will be removed.

positional arguments:
  url                   v2rayN subsription link

options:
  -h, --help            show this help message and exit
  -v, --verbose         Verbose mode
  -n, --dryrun          Don't modify anything on disk
  -o OUTPUT, --output OUTPUT
                        Output directory
  -a AGENT, --agent AGENT
                        User agent for the HTTP request
  --tlsNameAsServer     Generate a copy of the original config, where the server address
                        is overridden by the TLS serverName. This may fix some
                        connection issues. The copy will have '-tlsServ.json' suffix.

Supported Configs

The content returned by v2rayN subscription link is a base64-encoded string. After decoding, we will get line-separated urls.

The following urls are supported (it will show a detailed warning and skip the url whenever something is not supported):

vmess://<base64_string>

• outbounds[0].settings.vnext[0]: address, port, users[0].id, users[0].alterId
• outbounds[0].streamSettings.network: tcp, ws
• outbounds[0].streamSettings.security: none, tls
• outbounds[0].streamSettings.tlsSettings: serverName
• outbounds[0].streamSettings.wsSettings: path

<proto>://<uuid>@<host>:<port>?...#<desc>

<proto> may be vless, trojan. This format is described here. Currently XTLS must be used for trojan protocol.

• (vless) outbounds[0].settings.vnext[0]: address, port, users[0].id, users[0].encryption, users[0].flow
• (trojan) outbounds[0].settings.servers[0]: address, port, password, flow (xray only)
• outbounds[0].streamSettings.network: tcp, ws
• outbounds[0].streamSettings.security: tls, xtls (deprecated)
• outbounds[0].streamSettings.tlsSettings: serverName, fingerprint (xray only)
• outbounds[0].streamSettings.xtlsSettings (deprecated): serverName
• outbounds[0].streamSettings.wsSettings: path