$ sdk install springboot
SDKMAN can't reach the internet so going offline. Re-enable online with:

  $ sdk offline disable

sdk:local:70: ^[[01;31m^[[Kspringboot^[[m^[[K: can't create local array elements

$ sdk version
SDKMAN 3.2.4

$ sdk version
WARNING: SDKMAN is out-of-date and requires an update. Please run:

sdk update

==== BROADCAST =================================================================
* 02/07/18: Kscript 2.4.5 released on SDKMAN! #kscript
* 28/06/18: Micronaut 1.0.0.M2 released on SDKMAN! #micronautfw
* 27/06/18: Infrastructor 0.2.0 rolling out. With stronger encryption and new task progress UI.
================================================================================

SDKMAN 5.6.4+305

$ sdk update

-bash: -1: substring expression < 0

I'm using MacOS High Sierra.

Hi,

How to find what is installed with Sdkman?

There doesn't seem to be a sdk list installed commands, and just calling sdk list simply shows all potential SDKs to install.

Thanks.

Feature Request: Offer OpenJDK options for Java.

I know that you can manually install local versions, but this is a bit kludgy and doesn't take advantage of Sdkman tracking updates.

On Cygwin, tried sdk install java and got following output:

$ sdk install java

Oracle requires that you agree with the Oracle Binary Code License Agreement
prior to installation. The license agreement can be found at:

  http://www.oracle.com/technetwork/java/javase/terms/license/index.html

Do you agree to the terms of this agreement? (Y/n): Y


Downloading: java 8u111

In progress...

######################################################################## 100.0%

Repackaging java 8u111...
-bash: /home/Jonathan/.sdkman/tmp/mbgywYdNTdwToa040cg3Azu3oUSUBGsp.bin: Device or resource busy
mv: cannot stat '/cygdrive/c/temp/jdk': No such file or directory

zip error: Nothing to do! (try: zip -qr /home/Jonathan/.sdkman/tmp/mbgywYdNTdwToa040cg3Azu3oUSUBGsp.zip . -i .)

Done repackaging...
mv: cannot stat '/home/Jonathan/.sdkman/tmp/mbgywYdNTdwToa040cg3Azu3oUSUBGsp.zip': No such file or directory
unzip:  cannot find or open /home/Jonathan/.sdkman/archives/java-8u111.zip, /home/Jonathan/.sdkman/archives/java-8u111.zip.zip or /home/Jonathan/.sdkman/archives/java-8u111.zip.ZIP.
rm: cannot remove '/home/Jonathan/.sdkman/archives/java-8u111.zip': No such file or directory

Stop! The archive was corrupt and has been removed! Please try installing again.

To check device or resource busy, I ran handle command for windows:

C:\>handle mbgywYdNTdwToa040cg3Azu3oUSUBGsp.bin

Nthandle v4.1 - Handle viewer
Copyright (C) 1997-2016 Mark Russinovich
Sysinternals - www.sysinternals.com

bash.exe           pid: 8008   type: File           3D0: C:\Users\Jonathan\.sdkman\tmp\mbgywYdNTdwToa040cg3Azu3oUSUBGsp.bin

And then I ran ps within Cygwin:

$ ps
      PID    PPID    PGID     WINPID   TTY         UID    STIME COMMAND
     1948   13268    1948       8008  pty3      197609 19:57:39 /usr/bin/bash
     7532   12708    7532       8256  pty2      197609 02:13:00 /usr/bin/ps
     1044    1164    1044      15392  pty0      197609 19:28:45 /usr/bin/bash
    13268       1   13268      13268  ?         197609 19:57:39 /usr/bin/mintty
    14852       1   14852      14852  ?         197609 19:51:14 /usr/bin/mintty
    12708   14852   12708       4204  pty2      197609 19:51:14 /usr/bin/bash
     1164       1    1164       1164  ?         197609 19:28:45 /usr/bin/mintty

It seems wrong to hard-code the candidate names when they are available from the server. It also makes it hard to add new candidates. The scripts pretty much work with a really small change to gvm-use.sh, but the tests also rely on the PATH being set, so there is a change to the tests as well to make them pass.

As a Groovy developer, I often install development versions of Groovy locally. It's usually the master branch of upcoming Groovy releases, etc.

It'd be nice if I could tell GVM to use my local installation as well, in addition to all the known release versions. For example:

gvm install local groovy 2.1.0-SNAPSHOT /dev/groovy/master/target/install

Or something like that. I'm sure you'll find a better command-line to use.

The previous work-around for using GVM with the Fish shell was to add all the 'current' links to the PATH. This appears to be broken with newer versions of GVM. The 'current' link is only set to the default version of the given tool and doesn't get updated when 'use'-ing a different version.

I have a new Windows 7 laptop. First thing I did was install Cygwin. Then while setting up my dev environment I decided to try Gvm.

I added the following lines at the end of my .bashrc: export JAVA_HOME=/cygdrive/c/dev/sdks/jdk1.7.0_25/ export PATH=$PATH:$JAVA_HOME/bin

This being a default Cygwin install, with nothing customized yet, my .bashrc file is sourced by my .bash_profile file when I launch Cygwin.

I followed the simple gvm install process, by running the curl command. When I open Cygwin, post-install, I get the following problem: -bash: export: HTTP/1.1': not a valid identifier -bash: export:200': not a valid identifier ': not a valid identifier -bash: export: CONTENT-TYPE:': not a valid identifier ': not a valid identifierN -bash: export:CONTENT-LENGTH:': not a valid identifier ': not a valid identifier -bash: export: `CONNECTION:': not a valid identifier ': not a valid identifierE ': not a valid identifier A new version of GVM is available...

The current version is HTTP/1.1 200 OK Content-Length: 5 Connection: keep-alive

1.3.3, but you have 1.3.3.

Would you like to upgrade now? (Y/n)y bash: line 1: HTTP/1.1: No such file or directory bash: line 2: Content-Type:: command not found bash: line 3: Content-Length:: command not found bash: line 4: Connection:: command not found bash: line 5: $'\r': command not found

Updating gvm... warning [C:\cygwin64\home\eweibust.gvm\tmp\res-1.3.3.zip]: 97 extra bytes at beginning or within zipfile (attempting to process anyway)

Successfully upgraded GVM.

Please open a new terminal, or run the following in the existing one:

source "/home/eweibust/.gvm/bin/gvm-init.sh"

eweibust@RSNPLALT538 ~ $ echo $PATH | tr ':' '\n' /home/eweibust/.gvm/vertx/current/bin /home/eweibust/.gvm/lazybones/current/bin /home/eweibust/.gvm/groovyserv/current/bin /home/eweibust/.gvm/groovy/current/bin /home/eweibust/.gvm/griffon/current/bin /home/eweibust/.gvm/grails/current/bin Content-Typet/.gvm/HTTP/1.1 200 OK Content-Length Connection gradle/current/bin /usr/local/bin /usr/bin /cygdrive/c/Windows/system32 /cygdrive/c/Windows /cygdrive/c/Windows/System32/Wbem /cygdrive/c/Windows/System32/WindowsPowerShell/v1.0 /cygdrive/c/dev/sdks/jdk1.7.0_25//bin

eweibust@RSNPLALT538 ~ $

It appears there is a problem with the bash script gvm is running, which is then corrupting my PATH (shown in the echo $PATH above).

Using sdkman SDKMAN 5.11.2+698 on Ubuntu Linux 20.04 and bash 5.0.17(1) with the necessary code in .bashrc to put it in the path.

When I start a new terminal, i.e. invoke a new bash, output like the following sometimes appears on the console:

ommand not found
36m====: command not found

Command '0music' not found, did you mean:

  command 'music' from deb music-bin (1.1.16-1.1build2)

Try: sudo apt install <deb name>


Command '0music' not found, did you mean:

  command 'music' from deb music-bin (1.1.16-1.1build2)

Try: sudo apt install <deb name>


Command '0music' not found, did you mean:

  command 'music' from deb music-bin (1.1.16-1.1build2)

Try: sudo apt install <deb name>

================================================================================: command not found

I did not manage to find out why/when this happens, it appears to be almost random, though it usually happens when opening the first terminal after a restart or after a longer time.

I have observed this behaviour for several weeks now and I am certain this only happens when the SDKMAN-related code is active in .bashrc, never happens when I comment that out.

The "command not found" complaints I get appear to come from strings inside the bash prompt definition and from the names of random directories in my home directory. I have no idea at all what could be going on here, but it became so annoying that I usually deactivate sdkman completely.

So sadly, I cannot provide steps to reproduce because this problem occurs seemingly randomly (but often enough to be very annoying).

Please tick one:

• [ ] this is a new Feature Request, a conversation was started in the Gitter CLI Development Room.
• [x] this is a new Bug Report or Issue, a conversation was started on Twitter.

Please explain the Issue / Feature Request here:

SDKMAN warned me that there was an update, while there actually wasn't:

ATTENTION: A new version of SDKMAN is available...

The current version is 5.7.3+337, but you have 5.7.3+337

Would you like to upgrade now? (Y/n): Y
No update available at this time.

I don't know how to reproduce the issue, using the sdk update command immediately tells me there are no updates.

See also discussion on twitter: https://twitter.com/mhogerheijde/status/1072868313846312964

Details about my system:

$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 18.10
Release:        18.10
Codename:       cosmic
$ zsh --version
zsh 5.5.1 (x86_64-ubuntu-linux-gnu)
$ xfce4-terminal --version
xfce4-terminal 0.8.7.4 (Xfce 4.12)
$ tmux -V
tmux 2.7

Quoting #1069:

the current approach is very dotfiles unfriendly

Hi. My problem is that I am unable to install sdkman easily on a new system with get.sdkman.io, just because recently I've added a single .sdkman/etc/config to my dotfiles.

And now on a fresh new system I first deploy my dotfiles and then start installing software and tools which I need. The problem I am facing with sdkman is that now it refuses to be installed with curl -s "https://get.sdkman.io" | bash and complains: "You already have SDKMAN installed" just because the directory with .sdkman/etc/config already exists (the installation script has a simple dir existence check: [ -d "$SDKMAN_DIR" ])

As a unix-based system's user I would expect user's configuration to be stored at $XDG_HOME_CONFIG, so I totally support #941 feature request

Is there an option to disable the output to the shell when the used jvm is changed? like "Using java version xxx in this shell."

I would like that a lot, especially when the shell just got started.

This PR makes the install script "archive type"-aware, performing checksums for both zips and tarballs, and unpacks either archive type with the appropriate tool

• [x] I have checked the documentation
• [x] I have done a quick search in past issues or Stack Overflow for similar problems
• [ ] I have brought up a conversation in the Slack User Issues Channel

Question

We have .sdkmanrcs in our project to ensure everyone's using the same java-version. However, we aren't particular to the exact-version. We just need a way to enforce (e.g.) Java 11.

The problem with setting the exact version is, that it might be outdated as soon as another developer joins the team. The existing team-members already have the java-version installed via sdkman, so sdk env works as expected. The new team-member can't install the now missing Java-version and we are forced to update the .sdkmanrc across all repositories, the new team-member is expected to use.

This is a chore I would like to avoid in the future.

Is there a way to use sdkman to achieve what I'm asking for? Is this maybe a non-idiomatic request that should be handled differently?

System info

Not relevant (will provide if seemingly relevant though)

Bug report

Installation fails with a 404 - not found message when downloading native binary. I've added a verbose flag to curl, see the output below:

Looking for a previous installation of SDKMAN...
Looking for unzip...
Looking for zip...
Looking for curl...
Looking for sed...
Installing SDKMAN scripts...
Create distribution directories...
Getting available candidates...
Prime the config file...
Installing script cli archive...
* Downloading...
############################################################################################################################################################################################################## 100.0%
* Checking archive integrity...
* Extracting archive...
* Copying archive contents...
* Cleaning up...

* Downloading...
*   Trying 45.55.42.78:443...                                                           #                #                #               #                                                                         
* Connected to api.sdkman.io (45.55.42.78) port 443 (#0)
* ALPN, offering h2                                                                            #                #               #               #                                                                   
* ALPN, offering http/1.1
* successfully set certificate verify locations:
*  CAfile: /etc/ssl/certs/ca-certificates.crt
*  CApath: /etc/ssl/certs
} [5 bytes data]
* TLSv1.3 (OUT), TLS handshake, Client hello (1):
} [512 bytes data]
* TLSv1.3 (IN), TLS handshake, Server hello (2):                                                      #               #               #               #                                                             
{ [122 bytes data]
* TLSv1.3 (IN), TLS handshake, Encrypted Extensions (8):
{ [25 bytes data]
* TLSv1.3 (IN), TLS handshake, Certificate (11):
{ [4578 bytes data]
* TLSv1.3 (IN), TLS handshake, CERT verify (15):
{ [264 bytes data]
* TLSv1.3 (IN), TLS handshake, Finished (20):
{ [52 bytes data]
* TLSv1.3 (OUT), TLS change cipher, Change cipher spec (1):
} [1 bytes data]
* TLSv1.3 (OUT), TLS handshake, Finished (20):
} [52 bytes data]
* SSL connection using TLSv1.3 / TLS_AES_256_GCM_SHA384
* ALPN, server accepted to use http/1.1
* Server certificate:
*  subject: CN=*.sdkman.io
*  start date: Apr 10 00:00:00 2022 GMT
*  expire date: Apr 28 23:59:59 2023 GMT
*  subjectAltName: host "api.sdkman.io" matched cert's "*.sdkman.io"
*  issuer: C=GB; ST=Greater Manchester; L=Salford; O=Sectigo Limited; CN=Sectigo RSA Domain Validation Secure Server CA
*  SSL certificate verify ok.
} [5 bytes data]                                                                                            #                #               #              #                                                       
> GET /2/broker/download/native/install/0.0.15/linuxarm32hf HTTP/1.1
> Host: api.sdkman.io
> User-Agent: curl/7.74.0
> Accept: */*
> 
{ [5 bytes data]
* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):
{ [265 bytes data]
* TLSv1.3 (IN), TLS handshake, Newsession Ticket (4):
{ [265 bytes data]
* old SSL session ID is stale, removing
{ [5 bytes data]                                                                                                   #               #               #              #                                                 
* Mark bundle as not supporting multiuse
* The requested URL returned error: 404 Not Found
* Closing connection 0
} [5 bytes data]
* TLSv1.3 (OUT), TLS alert, close notify (256):
} [2 bytes data]
curl: (22) The requested URL returned error: 404 Not Found

'curl -v --fail --location --progress-bar "$download_url" > "$sdkman_zip_file"': command failed with exit code 22.

To reproduce

On a Raspberry Pi 2 Model B, running raspbian newest raspbian (see OS version below), try running install script $ curl -s https://get.sdkman.io | bash, notice the 404 error.

System info

System: GNU/Linux

uname -a
Linux raspberrypi 5.15.61-v7+ #1579 SMP Fri Aug 26 11:10:59 BST 2022 armv7l GNU/Linux

Shell Bash:

bash --version
GNU bash, version 5.1.4(1)-release (arm-unknown-linux-gnueabihf)
Copyright (C) 2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>