Curl is a powerful tool that is mainly used to transfer data. It has way more functions, but I won't be able to cover everything. This blog post is mainly a reference for later use and not a step-by-step guide. Therefore I won't cover everything in depth.
Most of it should work on other operating systems too, but I'll use Linux as reference. I'll keep this page up-to-date and add more topics in the future.
Side note: put the URL into single or double quotes if it contains special characters.
By default, curl writes the received data to stdout and does not encode or decode received data.
curl brrl.net/ipcurl -L brrl.net/ip # -L to get through the HTTP>HTTP if necessaryYou can redirect the content from stdout to another application, save it as a file or download the target file.
curl -L -o myip.txt brrl.net/ip # save your public IP to a file called myip.txt in the current directoryIf you want to download a file and keep the original name, use the -O (capital 'o') or --remote-name option.
If you want to create a new directory, you can use --create-dirs like this:
curl -L --create-dirs -o path/from/current/dir/myip.txt brrl.net/ip
The permission used is 0750.
You can use the --interface option to use one specific interface. You are free to use the interface name, the IP address, or the hostname.
You can choose a specific DNS server with the following option. Multiple DNS servers can be chosen and must be separated by a comma.
--dns-servers 9.9.9.9:53,149.112.112.112:53
If you want curl to follow redirects, simply use the -L flag.
Some tasks require many options. To keep it organized, you can import those options from a file with the -K or --config and followed by the name of the file.
curl --config curl-options.txt https://example.comYou can set up- and download limits with --limit-rate. The default are bytes/second, and you can use K,M,G,T for Kilo-,Mega-,Giga- and Terabyte, respectively.
--limit-rate 10K
--limit-rate 1000
--limit-rate 10M
To let curl transfer data parallel, you can use the -Z or --parallel and choose --parallel-immediate to start immediately.
-Z --parallel-immediate
The default is 50 parallel transfers but can be set with --parallel-max NUMBER.
--retry 999 --retry-max-time 0 -C ---retry 999 # retry it 999 times--retry-max-time 0 # prevent the default timeout between retries-C - # continue the transfer when you run the command again, and let curl figure out where to continueSide note: make sure to put the full URL into single or double quotes if you work with wildcards and sequences.
You can tell curl to transfer multiple files by putting the names into curly brac {}
curl -O 'http://{domain1,domain2,domain3}.com'curl -O 'http://domain.com/{uri1,uri2,uri3}'curl "http://{one,two}.example.com" -o "file_#1.txt"And you can use multiple sets, as shown in this example:
kuser@pleasejustwork:~/temp/curl$ curl "http://example.com/{1,2}/{3,4}" -o "file_#1_#2.txt"
[1/4]: http://example.com/1/3 --> file_1_3.txt
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1256 100 1256 0 0 6404 0 --:--:-- --:--:-- --:--:-- 6375
[2/4]: http://example.com/1/4 --> file_1_4.txt
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1256 100 1256 0 0 12753 0 --:--:-- --:--:-- --:--:-- 12816
[3/4]: http://example.com/2/3 --> file_2_3.txt
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1256 100 1256 0 0 12765 0 --:--:-- --:--:-- --:--:-- 12816
[4/4]: http://example.com/2/4 --> file_2_4.txt
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 1256 100 1256 0 0 12804 0 --:--:-- --:--:-- --:--:-- 12816l
kuser@pleasejustwork:~/temp/curl$ ls
file_1_3.txt file_1_4.txt file_2_3.txt file_2_4.txt
[] for alphanumeric sequences:curl -O 'http://example.com/picture-[1-51].img'curl -O 'http://example.com/attach-[a-z].img'It works with leading zeros too.
Nested sequences are not supported!
curl -O 'http://example.com/picture-[1-50:2].img' # every second pictureI am not too familiar with the proxy functions. I normally just use it to download things from Tor.
--socks5-hostname localhost:9150 # On Linux--socks5-hostname localhost:9050 # Windows or Tor in Browser BundleFor normal SOCKS5 sockets, you could simply use --socks5 instead of --socks5-hostname, but with --socks5-hostname, the DNS resolution runs on the proxy.
-x, --proxy [protocol://]host[:port]curl --proxy http://proxy.example https://example.comcurl --proxy socks5://proxy.example:12345 https://example.comcurl --proxy-basic --proxy-user user:password -x http://proxy.example https://example.comcurl -u name:password --basic https://example.comcurl http://username:password@example.com/api/ -H "Authorization: Bearer reallysecuretoken"curl --oauth2-bearer "mF_9.B5f-4.1JqM" https://example.comcurl --pass secret --key file https://example.comAnd there is so much more, but I'll leave it like that for now. Things I am going to add in the future: HTTP post/get requests, certificates troubleshooting, up- and downloading data through FTP, sftp, etc., mail /SMTP. I am unfamiliar with those, so I'll test a bunch before I add those topics here.
--
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。