See <https://red.libssh.org/issues/194>:
"In libssh 0.7.0, ssh_forward_listen() is present in the headers as
a deprecated function, but it is not actually present in the code.
This appears to have been introduced in commit 5229253f."
New features:
- Add support for auto-completing view names
- Add support for building RPM and DEB packages
- Add subsequence matching for completion
- Completion for built-in tables and temporary tables are suggested after
entering a prefix of pg_
- Add place holder doc strings for special commands that are planned for
implementation
- Updated version of prompt_toolkit, now matching braces are highlighted
- Added support of \\e command. Queries can be edited in an external editor
- Add special command \dT to show datatypes
- Add auto-completion support for datatypes in CREATE, SELECT etc
- Improve the auto-completion in WHERE clause with logical operators
Bug Fixes:
- Fix the table formatting while printing multi-byte characters
- Fix a crash when pg_catalog was present in search path
- Fixed a bug that broke \e when prompt_tookit was updated
- Fix the display of triggers as shown in the \d output
- Fix broken auto-completion for INNER JOIN, LEFT JOIN etc
- Fix incorrect super() calls in pgbuffer, pgtoolbar and pgcompleter
- Add missing schema completion for CREATE and DROP statements
- Minor fixes around cursor cleanup
- Typos: meta.maintainer -> meta.maintainers, licences -> license, ...
- Remove unnecessary stdenv.lib (already in scope for entire file)
- [ maintainers.foo maintainers.bar ] -> with maintainers; [ foo bar ]
Doing so even for single-maintainer packages makes it slightly
more inviting to add others, which can only be a good thing.
We need networking and potentially ssl/tls compiled into ffmpeg but the
minimal ffmpeg_x wrappers no longer offer that. We could of course use
ffmpeg-full and slim it down to just what we need but we'd end up
building ffmpeg twice, might as well stick with full
I presume there's something about how github creates the .zip files such
that they can change SHA.
Please note: this is a very outdated version of pasystray---I don't know
if that's intentional, but perhaps a better strategy would be to update
it wholesale.
