This is what was lost in Base:System due to the revert of bash in openSUSE:Factory
See https://bugzilla.opensuse.org/show_bug.cgi?id=1129036
It is up to you to decide what to do with it - but at least in its current form,
it is not working (install cycle with u-a/bash, detailed in the bug)
At least the non-/bin/sh related changes, like the patches, are certainly still wanted though.
- Update temporary patch from upstream to make sed testsuite
work again
- Requires(postun) -> Requires(preun)
- Do not link /bin/sh as slave alternative to /usr/bin/sh
- Add temporary patch from upstream to fix boo#1127700
- Refine and harden update-alternatives work flow
- Move header and Makefile from bash-loadables to bash-devel
- Make update-alternatives work flawless
- Put "sh" under control of update-alternatives
OBS-URL: https://build.opensuse.org/request/show/684575
OBS-URL: https://build.opensuse.org/package/show/Base:System/bash?expand=0&rev=286
- Update temporary patch from upstream to make sed testsuite
work again
- Requires(postun) -> Requires(preun)
- Do not link /bin/sh as slave alternative to /usr/bin/sh
- Add temporary patch from upstream to fix boo#1127700
- Refine and harden update-alternatives work flow
- Move header and Makefile from bash-loadables to bash-devel
- Make update-alternatives work flawless
- Put "sh" under control of update-alternatives
OBS-URL: https://build.opensuse.org/request/show/683701
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/bash?expand=0&rev=151
- Update to bash 5.0 (final)
a. Tilde expansion isn't performed on indexed array subscripts, even for
backwards compatibility.
b. The shell doesn't exit in posix mode if the eval builtin gets a parse
error when run by the command builtin.
c. Fixed a bug that caused a shell comment in an alias to not find the end
of the alias properly.
d. Reverted a change from April, 2018 that caused strings containing
backslashes to be flagged as glob patterns.
- Modify patches
* bash-2.03-manual.patch
* bash-4.3-loadables.dif
* bash-5.0.dif
- Update to bash 5.0 rc1
a. Fix to initial word completion detection code.
b. Fixed a bug that caused issues with assignment statements containing ^A in
the value assigned when IFS contains ^A.
c. Added a fallback to fnmatch() when strcoll can't correctly deal with
bracket expression character equivalence classes.
d. Fixed a bug that caused $BASH_COMMAND to contain the trap handler command
when running a trap handler containing [[ or (( commands.
e. Fixed a bug that caused nameref assignments in the temporary environment
to potentially create variables with invalid names.
f. Fixed a bug that caused `local -' to turn off alias expansion in scripts.
g. Fixed a parser issue with a command string containing EOF after an invalid
command as an argument to a special builtin not causing a posix-mode shell
to exit.
h. Made a slight change to the FNV-1 string hash algorithm used for associative
arrays (corrected the initial seed).
OBS-URL: https://build.opensuse.org/request/show/664075
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/bash?expand=0&rev=149
- Add patch 20, 21, 22 and 23 to bash-4.4-patches.tar.bz2
* 20: In circumstances involving long-running scripts that create
and reap many processes, it is possible for the hash table bash
uses to store exit statuses from asynchronous processes to
develop loops. This patch fixes the loop causes and adds code
to detect any future loops.
* 21: A SIGINT received inside a SIGINT trap handler can possibly
cause the shell to loop.
* 22: There are cases where a failing readline command (e.g.,
delete-char at the end of a line) can cause a multi-character
key sequence to `back up' and attempt to re-read some of the
characters in the sequence.
* 23: When sourcing a file from an interactive shell, setting the
SIGINT handler to the default and typing ^C will cause the
shell to exit.
- remove bash-4.4-wait-sigint-handler.patch (upstreamed)
OBS-URL: https://build.opensuse.org/request/show/613679
OBS-URL: https://build.opensuse.org/package/show/Base:System/bash?expand=0&rev=264
- Make sure that correct readline-devel version is used (current 7.0)
- Correct documentation path
- Due package split removed patches (for the bots)
* readline-6.2-xmalloc.dif
* readline-6.2-endpw.dif
* readline-6.3-destdir.patch
* readline-6.2-metamode.patch
* readline-7.0.dif
* readline-6.3-input.dif
* readline-5.2-conf.patch
* readline-6.3-rltrace.patch
- Split readline off into its own package
OBS-URL: https://build.opensuse.org/request/show/587961
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/bash?expand=0&rev=145
- Add patches 13-18 to bash-4.4-patches.tar.bz2
* 13: If a here-document contains a command substitution, the
command substitution can get access to the file descriptor used
to write the here-document.
* 14: Under some circumstances, functions that return via the
`return' builtin do not clean up memory they allocated to keep
track of FIFOs.
* 15: Process substitution can leak internal quoting to the
parser in the invoked subshell.
* 16: Bash can perform trap processing while reading command
substitution output instead of waiting until the command
completes.
* 17: There is a memory leak when `read -e' is used to read a
line using readline.
* 18: Under certain circumstances (e.g., reading from /dev/zero),
read(2) will not return -1 even when interrupted by a signal.
The read builtin needs to check for signals in this case.
- partial cleanup with spec-cleaner
OBS-URL: https://build.opensuse.org/request/show/570849
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/bash?expand=0&rev=142
- Add patches 13-18 to bash-4.4-patches.tar.bz2
* 13: If a here-document contains a command substitution, the
command substitution can get access to the file descriptor used
to write the here-document.
* 14: Under some circumstances, functions that return via the
`return' builtin do not clean up memory they allocated to keep
track of FIFOs.
* 15: Process substitution can leak internal quoting to the
parser in the invoked subshell.
* 16: Bash can perform trap processing while reading command
substitution output instead of waiting until the command
completes.
* 17: There is a memory leak when `read -e' is used to read a
line using readline.
* 18: Under certain circumstances (e.g., reading from /dev/zero),
read(2) will not return -1 even when interrupted by a signal.
The read builtin needs to check for signals in this case.
- partial cleanup with spec-cleaner
OBS-URL: https://build.opensuse.org/request/show/570827
OBS-URL: https://build.opensuse.org/package/show/Base:System/bash?expand=0&rev=248
