[ANNOUNCEMENT] Updated: tack 1.09-2.20210619

2022-11-18 Thread Cygwin tack Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* tack  1.09-2.20210619
* tack-debuginfo1.09-2.20210619

The tack program is a diagnostic that is designed to create and verify
the correctness of terminfo, and create new terminal descriptions that
are not included in the standard release.

For more information see the project home page:

https://invisible-island.net/ncurses/tack.html

As there are many changes each release please see below or read
/usr/share/doc/tack/CHANGES after installation:

https://invisible-island.net/tack/CHANGES.html


* Makefile.in: update copyright
* sysdep.c: scan-build warning about unused assignment
* Makefile.in: adjust rules to work with --program-suffix
* configure.in: add AC_ARG_PROGRAM to get --program-suffix, etc.
* aclocal.m4: resync with my-autoconf
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess: fix shellcheck warning SC2154
* config.guess: remove redundant quotes in case commands
* config.guess: simplify exit status workaround on alphaev67-dec-osf5.1
* config.sub: Handle MIPS R3 and R5 ISA levels with CPU names
* fun.c: rewrite the meta-key test to show the input character in
  readable form (from unctrl) in addition to hex. Also, make a second
  pass with the meta-mode turned off.
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess, config.sub: Recognize arc64
* config.guess: add SecBSD support
* config.sub: add SecBSD support
* aclocal.m4: resync with my-autoconf
* configure.in: minor fixes, per shellcheck
* package/debian/changelog, package/tack.spec, tack.h: bump
* tack.h: updates to work with _Noreturn changes in ncurses
* aclocal.m4: resync with my-autoconf
* config.sub: Add support for SerenityOS
* config.guess: update AROS system detection
* config.guess: fix shell variable quoting bug
* config.sub: recognize four-part configuration name for VxWorks
* config.sub: accept OS of eabi* and gnueabi*
* config.guess, config.sub: Update copyright years.
* config.guess, config.sub: Recognize riscv32be and riscv64be
* config.guess, config.sub: Recognize loongarch32, loongarch64, and loongarchx32
* config.sub: recognize thumbv7*
* tack.h, pad.c: improve interaction in padding tests by ensuring a
  time-delay happens for the cases where the terminal itself has no
  delays
* tackgen.c: fix a memory-leak
* tack.c: prefer exit_terminfo to _nc_free_tinfo
* configure.in: add check for exit_terminfo
* aclocal.m4: resync with my-autoconf
* package/debian/changelog, package/tack.spec, tack.h: bump
* tack.1: cleanup using check-manpage
* aclocal.m4: resync with my-autoconf
* configure.in: suppress checks for X
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess, config.sub: .gitattributes: specify a custom git merge driver 
for the ChangeLog file
* config.guess, config.sub: Update cgit URLs over gitweb URLs of the latest 
version of config.guess and config.sub scripts
* config.guess, config.sub: Update timestamps.
* config.sub, config.guess: Replace backtick `..` substitutions with POSIX 
$(..) command substitutions throughout.
* config.guess: Handle aarch64eb.
* config.sub: Fix whitespace problem in config.sub.
* config.sub: (i*86-pc-os2-emx): Recognise correctly.
* config.sub: (uclinux-uclibc*): Fix detection.
* config.guess: Don't use 'ldd --version' to determine the presence of musl 
libc, as this fails on Alpine Linux 3.10.
* config.sub: Fix regression in QNX recognition.
* config.guess, config.sub: Replace "if [ EXPR ]" with "if test EXPR".
* config.guess: (arm64:Darwin:*:*): Recognise.
* config.sub: (case $cpu): Whitespace fix.
* config.sub: (arm64-*): Canonicalise to aarch64-*.
* config.sub, config.guess: update
* sysdep.c: gcc-warning
* sysdep.c: two changes to allow most tests to work with mintty in
  Cygwin and MSYS2 (the meta-mode test is one case that does not work):
  - change the VTIME value in tty_set() to 0, since interbyte timing
does not work at all. Without this fix, getchar() returns
immediately without error, whether or not data is available.
  - in read_key(), do not exit from the loop as long as read() returns
-1 (error). Without this fix, read_key() returns ^@'s (nulls) in the
function-key test.
  These changes do not seem to break existing tests on other platforms,
  e.g., Unix (AIX, HPUX, Solaris).
* tack.c: fix gcc warning for non-ncurses configuration
* package/debian/changelog, package/tack.spec, tack.h: bump
* init.c: zero-out the debug/log-pointers after closing them
* tack.1: add -d option, for debug.log
* output.c: use debug.log to record putp's and tput's which hint at the
  testing done
* sysdep.c: use debug.log to record when the tty is set/reset
* tack.h: split-out log_chr and log_str
* tack.c: add -d option, for debug.log
* tack.1, tack.c, tack.h: add -l option
* tack.c: eliminate a confusing comparison for the logging-menu-entry
  state by using #define'd strings for that and the hex-output

Updated: tack 1.09-2.20210619

2022-11-18 Thread Cygwin tack Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* tack  1.09-2.20210619
* tack-debuginfo1.09-2.20210619

The tack program is a diagnostic that is designed to create and verify
the correctness of terminfo, and create new terminal descriptions that
are not included in the standard release.

For more information see the project home page:

https://invisible-island.net/ncurses/tack.html

As there are many changes each release please see below or read
/usr/share/doc/tack/CHANGES after installation:

https://invisible-island.net/tack/CHANGES.html


* Makefile.in: update copyright
* sysdep.c: scan-build warning about unused assignment
* Makefile.in: adjust rules to work with --program-suffix
* configure.in: add AC_ARG_PROGRAM to get --program-suffix, etc.
* aclocal.m4: resync with my-autoconf
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess: fix shellcheck warning SC2154
* config.guess: remove redundant quotes in case commands
* config.guess: simplify exit status workaround on alphaev67-dec-osf5.1
* config.sub: Handle MIPS R3 and R5 ISA levels with CPU names
* fun.c: rewrite the meta-key test to show the input character in
  readable form (from unctrl) in addition to hex. Also, make a second
  pass with the meta-mode turned off.
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess, config.sub: Recognize arc64
* config.guess: add SecBSD support
* config.sub: add SecBSD support
* aclocal.m4: resync with my-autoconf
* configure.in: minor fixes, per shellcheck
* package/debian/changelog, package/tack.spec, tack.h: bump
* tack.h: updates to work with _Noreturn changes in ncurses
* aclocal.m4: resync with my-autoconf
* config.sub: Add support for SerenityOS
* config.guess: update AROS system detection
* config.guess: fix shell variable quoting bug
* config.sub: recognize four-part configuration name for VxWorks
* config.sub: accept OS of eabi* and gnueabi*
* config.guess, config.sub: Update copyright years.
* config.guess, config.sub: Recognize riscv32be and riscv64be
* config.guess, config.sub: Recognize loongarch32, loongarch64, and loongarchx32
* config.sub: recognize thumbv7*
* tack.h, pad.c: improve interaction in padding tests by ensuring a
  time-delay happens for the cases where the terminal itself has no
  delays
* tackgen.c: fix a memory-leak
* tack.c: prefer exit_terminfo to _nc_free_tinfo
* configure.in: add check for exit_terminfo
* aclocal.m4: resync with my-autoconf
* package/debian/changelog, package/tack.spec, tack.h: bump
* tack.1: cleanup using check-manpage
* aclocal.m4: resync with my-autoconf
* configure.in: suppress checks for X
* package/debian/changelog, package/tack.spec, tack.h: bump
* config.guess, config.sub: .gitattributes: specify a custom git merge driver 
for the ChangeLog file
* config.guess, config.sub: Update cgit URLs over gitweb URLs of the latest 
version of config.guess and config.sub scripts
* config.guess, config.sub: Update timestamps.
* config.sub, config.guess: Replace backtick `..` substitutions with POSIX 
$(..) command substitutions throughout.
* config.guess: Handle aarch64eb.
* config.sub: Fix whitespace problem in config.sub.
* config.sub: (i*86-pc-os2-emx): Recognise correctly.
* config.sub: (uclinux-uclibc*): Fix detection.
* config.guess: Don't use 'ldd --version' to determine the presence of musl 
libc, as this fails on Alpine Linux 3.10.
* config.sub: Fix regression in QNX recognition.
* config.guess, config.sub: Replace "if [ EXPR ]" with "if test EXPR".
* config.guess: (arm64:Darwin:*:*): Recognise.
* config.sub: (case $cpu): Whitespace fix.
* config.sub: (arm64-*): Canonicalise to aarch64-*.
* config.sub, config.guess: update
* sysdep.c: gcc-warning
* sysdep.c: two changes to allow most tests to work with mintty in
  Cygwin and MSYS2 (the meta-mode test is one case that does not work):
  - change the VTIME value in tty_set() to 0, since interbyte timing
does not work at all. Without this fix, getchar() returns
immediately without error, whether or not data is available.
  - in read_key(), do not exit from the loop as long as read() returns
-1 (error). Without this fix, read_key() returns ^@'s (nulls) in the
function-key test.
  These changes do not seem to break existing tests on other platforms,
  e.g., Unix (AIX, HPUX, Solaris).
* tack.c: fix gcc warning for non-ncurses configuration
* package/debian/changelog, package/tack.spec, tack.h: bump
* init.c: zero-out the debug/log-pointers after closing them
* tack.1: add -d option, for debug.log
* output.c: use debug.log to record putp's and tput's which hint at the
  testing done
* sysdep.c: use debug.log to record when the tty is set/reset
* tack.h: split-out log_chr and log_str
* tack.c: add -d option, for debug.log
* tack.1, tack.c, tack.h: add -l option
* tack.c: eliminate a confusing comparison for the logging-menu-entry
  state by using #define'd strings for that and the hex-output

[ANNOUNCEMENT] Updated: dialog libdialog15/-devel 1.3-4.20220117

2022-11-18 Thread Cygwin dialog Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* dialog1.3-4.20220117
* dialog-debuginfo  1.3-4.20220117
* libdialog-devel   1.3-4.20220117
* libdialog15   1.3-4.20220117

A script interpreter which provides a set of curses widgets, objects
whose appearance and behavior can be customized.

For more information see the project home page:

https://invisible-island.net/dialog

As there are multiple components and many changes each release please
see below or read /usr/share/doc/dialog/CHANGES after installation:

https://invisible-island.net/dialog/CHANGES.html


2022/01/17
* update test-package for Debian to use debhelper.
* adjust character-offset in editbox.c in case it is past the end of the
  string after trimming
* updated ro.po from http://translationproject.org/latest/dialog/

2021/12/14
* fix some inconsistent #if vs #ifdef.
* correct off-by-one limit for month when validating parameters of the
  calendar widget
* modify dlg_yes_labels to not switch to ok/cancel labels if the extra
  button is used
* move leak-checking definitions from dialog.h to dlg_internals.h

2021/11/07
* use mktemp utility rather than tempfile for temporary filename in the
  sample scripts
* updated id.po, lt.po from http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/06/21
* directly include stddef.h for systems which need it for offsetof().
* modify gauge's handle_input function to use the file pointer in the
  object in preference to the global dialog_state.
* correct height-computation for fselect and timebox, broken by fixes
  for auto-layout
* improve configure --with-no-leaks option

2021/05/30
* amend adjustment to inputbox height so the autosized dimensions match
  older versions
* improve configure checks, to reduce warnings during those checks
* updated fr.po from http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/05/09
* improve configure checks, e.g., for libtool version
* updated ms.po, sr.po, zh_TW.po from
  http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/03/24
* add configure option --enable-stdnoreturn to build when C11 _Noreturn
  feature is enabled in recent ncurses development.

2021/03/19
* updated de.po, ja.po, nl.po from http://translationproject.org/latest/dialog/

2021/03/06
* add several usage-messages to dialog.pot
* improve configure-check for ld searchpath on HP-UX and Solaris.
* fix a few cppcheck warnings
* updated eo.po, fi.po, pt.po, sq.po, sv.po, ru.po, uk.po, zh_CN.po from
  http://translationproject.org/latest/dialog/

2021/01/17
* updated dialog.pot
* add option --no-hot-list, to allow suppressing the hotkey feature from lists.
* increase minimum height of inputbox to avoid input field overlapping
  with the button-box
* modify options-parsing to permit a "no" on any boolean option and
  deprecating the "no" options without a "-" , e.g., generalizing on
  "--nook" equivalence to "--no-ok".

2021/01/12
* minor spelling fixes, per codespell
* update configure macros, for quoting/escaping fixes per shellcheck
* update config.guess, config.sub

2020/11/26
* modify configure script's compiler check to work around Xcode's
  useless aliases for standard C.
* enable keep-tite feature for NetBSD 8 and up; earlier releases would dump core
* add option "--cursor-off-label"
* add option "--erase-on-exit"
* fix errata in manpage.
* fix logic: dlg_button_to_char() would return wrong value if no
  uppercase was found

2020/11/25
* add DIALOG_TIMEOUT to sample scripts, and use report-button more
  consistently to handle unexpected exit-status values.

2020/11/24
* add dlg_getenv_num() and dlg_getenv_str().
* add DLG_EXIT_TIMEOUT to allow scripts to exit on an expired timeout
  with an exit-status different from DLG_EXIT_ERROR or DLG_EXIT_ESC

2020/11/23
* modify dlg_ok_label() and dlg_ok_labels() to ignore --no-ok if all
  buttons would be omitted.
* add/use dlg_der_window() to account for derived windows just like
  subwindows, to fix regression in dlg_getc() in fselect/dselect widget.
* modify several widgets to make their handling of --no-ok consistent
  with the majority of the widgets: editbox.c, menubox.c, msgbox.c,
  rangebox.c, textbox.c, timebox.c
* add ^D as binding for new virtual key DLGK_LEAVE, which will close the
  current dialog by activating the currently-selected button.
* improve manual-page description of --no-ok option
* modify configure script to make Solaris -R rpath feature work

2020/11/21
* fix regression in dlg_getc() caused by not taking into account
  subwindows used for input

2020/11/20
* add validity-checks for window pointer
* improve configure macros: CF_CLANG_COMPILER: apply compiler-flags
  needed for test.  CF_WITH_SCREEN_PDCURSES: use this to reduce
  X11-checks.

2020/11/17
* amend initialization for dlg_getc to fix regression in pause widget
  from 2020/03/27

Updated: dialog libdialog15/-devel 1.3-4.20220117

2022-11-18 Thread Cygwin dialog Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* dialog1.3-4.20220117
* dialog-debuginfo  1.3-4.20220117
* libdialog-devel   1.3-4.20220117
* libdialog15   1.3-4.20220117

A script interpreter which provides a set of curses widgets, objects
whose appearance and behavior can be customized.

For more information see the project home page:

https://invisible-island.net/dialog

As there are multiple components and many changes each release please
see below or read /usr/share/doc/dialog/CHANGES after installation:

https://invisible-island.net/dialog/CHANGES.html


2022/01/17
* update test-package for Debian to use debhelper.
* adjust character-offset in editbox.c in case it is past the end of the
  string after trimming
* updated ro.po from http://translationproject.org/latest/dialog/

2021/12/14
* fix some inconsistent #if vs #ifdef.
* correct off-by-one limit for month when validating parameters of the
  calendar widget
* modify dlg_yes_labels to not switch to ok/cancel labels if the extra
  button is used
* move leak-checking definitions from dialog.h to dlg_internals.h

2021/11/07
* use mktemp utility rather than tempfile for temporary filename in the
  sample scripts
* updated id.po, lt.po from http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/06/21
* directly include stddef.h for systems which need it for offsetof().
* modify gauge's handle_input function to use the file pointer in the
  object in preference to the global dialog_state.
* correct height-computation for fselect and timebox, broken by fixes
  for auto-layout
* improve configure --with-no-leaks option

2021/05/30
* amend adjustment to inputbox height so the autosized dimensions match
  older versions
* improve configure checks, to reduce warnings during those checks
* updated fr.po from http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/05/09
* improve configure checks, e.g., for libtool version
* updated ms.po, sr.po, zh_TW.po from
  http://translationproject.org/latest/dialog/
* update config.guess, config.sub

2021/03/24
* add configure option --enable-stdnoreturn to build when C11 _Noreturn
  feature is enabled in recent ncurses development.

2021/03/19
* updated de.po, ja.po, nl.po from http://translationproject.org/latest/dialog/

2021/03/06
* add several usage-messages to dialog.pot
* improve configure-check for ld searchpath on HP-UX and Solaris.
* fix a few cppcheck warnings
* updated eo.po, fi.po, pt.po, sq.po, sv.po, ru.po, uk.po, zh_CN.po from
  http://translationproject.org/latest/dialog/

2021/01/17
* updated dialog.pot
* add option --no-hot-list, to allow suppressing the hotkey feature from lists.
* increase minimum height of inputbox to avoid input field overlapping
  with the button-box
* modify options-parsing to permit a "no" on any boolean option and
  deprecating the "no" options without a "-" , e.g., generalizing on
  "--nook" equivalence to "--no-ok".

2021/01/12
* minor spelling fixes, per codespell
* update configure macros, for quoting/escaping fixes per shellcheck
* update config.guess, config.sub

2020/11/26
* modify configure script's compiler check to work around Xcode's
  useless aliases for standard C.
* enable keep-tite feature for NetBSD 8 and up; earlier releases would dump core
* add option "--cursor-off-label"
* add option "--erase-on-exit"
* fix errata in manpage.
* fix logic: dlg_button_to_char() would return wrong value if no
  uppercase was found

2020/11/25
* add DIALOG_TIMEOUT to sample scripts, and use report-button more
  consistently to handle unexpected exit-status values.

2020/11/24
* add dlg_getenv_num() and dlg_getenv_str().
* add DLG_EXIT_TIMEOUT to allow scripts to exit on an expired timeout
  with an exit-status different from DLG_EXIT_ERROR or DLG_EXIT_ESC

2020/11/23
* modify dlg_ok_label() and dlg_ok_labels() to ignore --no-ok if all
  buttons would be omitted.
* add/use dlg_der_window() to account for derived windows just like
  subwindows, to fix regression in dlg_getc() in fselect/dselect widget.
* modify several widgets to make their handling of --no-ok consistent
  with the majority of the widgets: editbox.c, menubox.c, msgbox.c,
  rangebox.c, textbox.c, timebox.c
* add ^D as binding for new virtual key DLGK_LEAVE, which will close the
  current dialog by activating the currently-selected button.
* improve manual-page description of --no-ok option
* modify configure script to make Solaris -R rpath feature work

2020/11/21
* fix regression in dlg_getc() caused by not taking into account
  subwindows used for input

2020/11/20
* add validity-checks for window pointer
* improve configure macros: CF_CLANG_COMPILER: apply compiler-flags
  needed for test.  CF_WITH_SCREEN_PDCURSES: use this to reduce
  X11-checks.

2020/11/17
* amend initialization for dlg_getc to fix regression in pause widget
  from 2020/03/27

[ANNOUNCEMENT] Updated: ncurses/-demo libncurses/-devel/++/w10 terminfo/-extra 6.3-1.20220416

2022-11-18 Thread Cygwin ncurses Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* ncurses   6.3-1.20220416
* ncurses-demo  6.3-1.20220416
* ncurses-debuginfo 6.3-1.20220416
* libncurses-devel  6.3-1.20220416
* libncurses++w10   6.3-1.20220416
* libncursesw10 6.3-1.20220416
* terminfo  6.3-1.20220416
* terminfo-extra6.3-1.20220416

The ncurses (new curses) library is an emulation of Sys V R 4
curses, and more. It uses terminfo format, supports pads, color,
multiple highlights, forms characters, function key mapping,
and has all the other SVR4 curses enhancements over BSD curses.


For more information see the project home page:

https://invisible-island.net/ncurses

As there are multiple components and many changes each release please
see below or read /usr/share/doc/ncurses/ANNOUNCE after installation:

https://invisible-island.net/ncurses/announce.html


20220416
* add a limit-check to guard against corrupt terminfo data
* add check/warning in configure script if option --with-xterm-kbs is
  missing or inconsistent
* add setlocale call to several test-programs.
* allow extended-color number in opts parameter of wattr_on.

20220409
* add test/test_unget_wch.c

20220402
* amend extended_captype(), returning CANCEL if a string is explicitly
  cancelled.
* make description-fields distinct

20220326
* update teken
* add teken-16color, teken-vt and teken-sc
* add a few missing details for vte-2018

20220319
* add xgterm
* correct setal in mintty/tmux entries, add to vte-2018
* add blink to vte-2018
* improve tic warning about XT versus redundant tsl, etc.

20220312
* add xterm+acs building-block
* add xterm-p370, for use in older terminals
* add dec+sl to xterm-new, per xterm patch
* add mosh and mosh-256color

20220305
* replace obsolescent "-gnatg" option with "-gnatwa" and "-gnatyg", to
  work around build problems with gnat 12.
* update external links in Ada95.html
* trim unused return-value from canonical_name().

20220226
* fix issues found with coverity:
  - rewrite canonical_name() function of infocmp to ensure buffer size
  - corrected use of original tty-modes in tput init/reset subcommands
  - modify tabs program to limit tab-stop values to max-columns
  - add limit-checks for palette rgb values in test/ncurses.c
* add a few null-pointer checks to help with static-analysis.
* enforce limit on number of soft-keys used in c++ binding.
* adjust a buffer-limit in write_entry.c to quiet a bogus warning from
  gcc 12.0.1

20220219
* expanded description in man/resizeterm.3x
* additional workaround for ImageMagick in test/picsmap.c

20220212
* improve font-formatting in other manpages, for consistency.
* correct/improve font-formatting in curs_wgetch.3x

20220205
* workaround in test/picsmap.c for use of floating point for rgb values
  by ImageMagick 6.9.11, which appears to use the wrong upper limit.
* improve use of "trap" in shell scripts, using "fixup-trap".

20220129
* minor updates for test-packages
* improve handling of --with-pkg-config-libdir option, allowing for the
  case where either $PKG_CONFIG_LIBDIR or the option value has a
  colon-separated list of directories
* update kitty

20220122
* add ABI 7 defaults to configure script.
* add warning in configure script if file specified for "--with-caps"
  does not exist.
* use fix for CF_FIX_WARNINGS from cdk-perl, ignoring error-exit on
  format-warnings.
* improve readability of long parameterized expressions with the infocmp
  "-f" option by allowing split before a "%p" marker.

20220115
* improve checks for valid mouse events when an intermediate mouse state
  is not part of the mousemask specified by the caller
* use newer version 1.36 of gnathtml for generating Ada html files.

20220101
* add section on releasing memory to curs_termcap.3x and
  curs_terminfo.3x manpages.

20211225
* improve markup, e.g., for external manpage links in the manpages

20211219
* install ncurses-examples programs in libexecdir, adding a wrapper
  script to invoke those.
* add help-screen and screen-dump to test/combine.c

20211211
* add test/combine.c, to demo/test combining characters.

20211204
* improve configure check for getttynam

20211127
* fix errata in description fields
* add x10term+sl, aixterm+sl, ncr260vp+sl, ncr260vp+vt, wyse+sl

20211120
* add dim, ecma+strikeout to st-0.6
* deallocate the tparm cache when del_curterm is called for the last
  allocated TERMINAL structure
* modify test-package to more closely conform to Debian multi-arch.
* if the --with-pkg-config-libdir option is not given, use
  ${libdir}/pkgconfig as a default

2025
* fix memory-leak in delwin for pads

2023
* minor clarification to clear.1
* add xterm+sl-alt, use that in foot+base
* improve search-path check for pkg-config, for Debian testing which
  installs pkg-config with architecture-prefixes.

20211106
* improve check in misc/Makefile.in

Updated: ncurses/-demo libncurses/-devel/++/w10 terminfo/-extra 6.3-1.20220416

2022-11-18 Thread Cygwin ncurses Maintainer 
The following packages have been upgraded in the Cygwin distribution:

* ncurses   6.3-1.20220416
* ncurses-demo  6.3-1.20220416
* ncurses-debuginfo 6.3-1.20220416
* libncurses-devel  6.3-1.20220416
* libncurses++w10   6.3-1.20220416
* libncursesw10 6.3-1.20220416
* terminfo  6.3-1.20220416
* terminfo-extra6.3-1.20220416

The ncurses (new curses) library is an emulation of Sys V R 4
curses, and more. It uses terminfo format, supports pads, color,
multiple highlights, forms characters, function key mapping,
and has all the other SVR4 curses enhancements over BSD curses.


For more information see the project home page:

https://invisible-island.net/ncurses

As there are multiple components and many changes each release please
see below or read /usr/share/doc/ncurses/ANNOUNCE after installation:

https://invisible-island.net/ncurses/announce.html


20220416
* add a limit-check to guard against corrupt terminfo data
* add check/warning in configure script if option --with-xterm-kbs is
  missing or inconsistent
* add setlocale call to several test-programs.
* allow extended-color number in opts parameter of wattr_on.

20220409
* add test/test_unget_wch.c

20220402
* amend extended_captype(), returning CANCEL if a string is explicitly
  cancelled.
* make description-fields distinct

20220326
* update teken
* add teken-16color, teken-vt and teken-sc
* add a few missing details for vte-2018

20220319
* add xgterm
* correct setal in mintty/tmux entries, add to vte-2018
* add blink to vte-2018
* improve tic warning about XT versus redundant tsl, etc.

20220312
* add xterm+acs building-block
* add xterm-p370, for use in older terminals
* add dec+sl to xterm-new, per xterm patch
* add mosh and mosh-256color

20220305
* replace obsolescent "-gnatg" option with "-gnatwa" and "-gnatyg", to
  work around build problems with gnat 12.
* update external links in Ada95.html
* trim unused return-value from canonical_name().

20220226
* fix issues found with coverity:
  - rewrite canonical_name() function of infocmp to ensure buffer size
  - corrected use of original tty-modes in tput init/reset subcommands
  - modify tabs program to limit tab-stop values to max-columns
  - add limit-checks for palette rgb values in test/ncurses.c
* add a few null-pointer checks to help with static-analysis.
* enforce limit on number of soft-keys used in c++ binding.
* adjust a buffer-limit in write_entry.c to quiet a bogus warning from
  gcc 12.0.1

20220219
* expanded description in man/resizeterm.3x
* additional workaround for ImageMagick in test/picsmap.c

20220212
* improve font-formatting in other manpages, for consistency.
* correct/improve font-formatting in curs_wgetch.3x

20220205
* workaround in test/picsmap.c for use of floating point for rgb values
  by ImageMagick 6.9.11, which appears to use the wrong upper limit.
* improve use of "trap" in shell scripts, using "fixup-trap".

20220129
* minor updates for test-packages
* improve handling of --with-pkg-config-libdir option, allowing for the
  case where either $PKG_CONFIG_LIBDIR or the option value has a
  colon-separated list of directories
* update kitty

20220122
* add ABI 7 defaults to configure script.
* add warning in configure script if file specified for "--with-caps"
  does not exist.
* use fix for CF_FIX_WARNINGS from cdk-perl, ignoring error-exit on
  format-warnings.
* improve readability of long parameterized expressions with the infocmp
  "-f" option by allowing split before a "%p" marker.

20220115
* improve checks for valid mouse events when an intermediate mouse state
  is not part of the mousemask specified by the caller
* use newer version 1.36 of gnathtml for generating Ada html files.

20220101
* add section on releasing memory to curs_termcap.3x and
  curs_terminfo.3x manpages.

20211225
* improve markup, e.g., for external manpage links in the manpages

20211219
* install ncurses-examples programs in libexecdir, adding a wrapper
  script to invoke those.
* add help-screen and screen-dump to test/combine.c

20211211
* add test/combine.c, to demo/test combining characters.

20211204
* improve configure check for getttynam

20211127
* fix errata in description fields
* add x10term+sl, aixterm+sl, ncr260vp+sl, ncr260vp+vt, wyse+sl

20211120
* add dim, ecma+strikeout to st-0.6
* deallocate the tparm cache when del_curterm is called for the last
  allocated TERMINAL structure
* modify test-package to more closely conform to Debian multi-arch.
* if the --with-pkg-config-libdir option is not given, use
  ${libdir}/pkgconfig as a default

2025
* fix memory-leak in delwin for pads

2023
* minor clarification to clear.1
* add xterm+sl-alt, use that in foot+base
* improve search-path check for pkg-config, for Debian testing which
  installs pkg-config with architecture-prefixes.

20211106
* improve check in misc/Makefile.in

Adding an embedded signature on setup-x86_64.exe

2022-11-18 Thread Dale McCoy 
I use Cygwin in the course of work, and while I can use the external gpg
signature to verify the validity of setup-x86_64.exe, my IT department
can't see that step. They get somewhat concerned when they see that Windows
thinks setup-x86_64.exe is unsigned, and I certainly don't blame them.
Can I convince you to also embed a signature in the installer, so Windows
recognizes the file is signed?

I couldn't find a previous request on the mailing list for this, but I may
have missed it in my attempts to grep the monthly digests.

Thanks,
Dale McCoy

-- 
Problem reports:  https://cygwin.com/problems.html
FAQ:  https://cygwin.com/faq/
Documentation:https://cygwin.com/docs.html
Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple

Re: RE: Cygwin x86 end-of-life

2022-11-18 Thread Brian Inglis 

On Fri, 18 Nov 2022 15:51:34 +, Jon Turney wrote:

On 14/11/2022 21:29, Jason Pyeron wrote:

On Monday, November 14, 2022 3:17 PM, Brian Inglis wrote:

On Mon, 14 Nov 2022 16:25:18 +, Jon Turney wrote:

On 11/11/2022 16:16, Jon Turney wrote:

On 11/11/2022 15:50, Jon Turney wrote:

As has previously been announced, Cygwin is dropping support for x86
Windows. Cygwin 3.3.6 is the final version supporting x86 (32-bit)
Windows, and the forthcoming Cygwin 3.4 will be released for x86_64 only.
Concurrent with that, updates to x86 packages will be stopped, and the
Cygwin x86 package repository will be archived.



I plan to pause package uploads this coming Monday (2022-11-14), before
starting the re-organization of the package repository to make this
archive.



Since there have been some complaints about short notice, and to give
time to work out the issues with gettext/libunistring, I'm going to
defer this by one week, until Monday 2022-11-21.



Thank you very much appreciated, hopefully we can deal with the remaining issues
quickly.



I do not have an articulate retort to ending support, but all I can say is
that I feel there must be a middle ground.
I feel that there could and should be some form of "we don’t support it"
but we are not going out of our way to prevent it.



I'm unclear if this means:
(a) "releasing Cygwin 3.4.x DLL for x86 OS"
(b) "don't require x86 uploads, but don't prevent them, either"
The problem I have with (b) is that we probably end up in a state where 
x86 is missing package updates people want or need; or broken, but we 
don't know it, because nobody uses it.



Can I throw resources at a solution? If so what?



Sure, if that's what you want to do.
According surveys, 32-bit Windows has a fraction of 1% market share, and 
declining. Our own (limited) metrics are in accord with that, so I 
basically see any time I spend on this as wasted.

So, the first resource you'll need provide is manpower.


The decision makes sense with those numbers.
Do we have numbers to say what the situation is with Windows mingw64-i686 
crosses?
Should we also be dropping those at the same time, if there is only 1% use of 
that platform?

In which case, we should announce that, and add that to the EoL notices.

--
Take care. Thanks, Brian Inglis Calgary, Alberta, Canada

La perfection est atteinte  Perfection is achieved
non pas lorsqu'il n'y a plus rien à ajouter not when there is no more to add
mais lorsqu'il n'y a plus rien à retirerbut when there is no more to cut
-- Antoine de Saint-Exupéry

[ANNOUNCEMENT] Updated: gettext{,-devel,doc} emacs-gettext lib{gettextpo,asprintf}{0,-devel} libintl{8,-devel} 0.21.1-2

2022-11-18 Thread Cygwin Gettext Maintainer 
The following packages have been uploaded to the Cygwin distribution:

* gettext   0.21.1-2
* gettext-devel 0.21.1-2
* gettext-doc   0.21.1-2
* emacs-gettext 0.21.1-2
* libgettextpo0 0.21.1-2
* libgettextpo-devel0.21.1-2
* libasprintf0  0.21.1-2
* libasprintf-devel 0.21.1-2
* libintl8  0.21.1-2
* libintl-devel 0.21.1-2

A well integrated set of tools and documentation that provides a
framework for producing multi-lingual messages in programs.
Includes a set of conventions for writing programs to support message
catalogs, a directory and file naming organization for message catalogs,
a library for retrieving translated messages, and programs for handling
translatable and translated strings.

For changes since the previous Cygwin version, see below,
/usr/share/doc/gettext/NEWS after installation, or the link:

https://git.savannah.gnu.org/gitweb/?p=gettext.git;f=NEWS;a=blob

Cygwin release 2 fixes library dependency issues with unistring.


Version 0.21.1 - October 2022

* The base Unicode standard is now updated to 14.0.0.

* Portability:
  - Building on macOS 11/arm64 is now supported.
  - Building on Linux/powerpc64le with glibc ≥ 2.35 is now supported.

* Runtime behaviour:
  - On AIX, locale names with a script or with an uppercase language are now
supported.
For example, sr_Cyrl_RS.UTF-8 is treated like sr_RS.UTF-8@cyrillic, and
EN_US.UTF-8 is treated like en_US.UTF-8.


-- 
Problem reports:  https://cygwin.com/problems.html
FAQ:  https://cygwin.com/faq/
Documentation:https://cygwin.com/docs.html
Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple

Updated: gettext{,-devel,doc} emacs-gettext lib{gettextpo,asprintf}{0,-devel} libintl{8,-devel} 0.21.1-2

2022-11-18 Thread Cygwin Gettext Maintainer 
The following packages have been uploaded to the Cygwin distribution:

* gettext   0.21.1-2
* gettext-devel 0.21.1-2
* gettext-doc   0.21.1-2
* emacs-gettext 0.21.1-2
* libgettextpo0 0.21.1-2
* libgettextpo-devel0.21.1-2
* libasprintf0  0.21.1-2
* libasprintf-devel 0.21.1-2
* libintl8  0.21.1-2
* libintl-devel 0.21.1-2

A well integrated set of tools and documentation that provides a
framework for producing multi-lingual messages in programs.
Includes a set of conventions for writing programs to support message
catalogs, a directory and file naming organization for message catalogs,
a library for retrieving translated messages, and programs for handling
translatable and translated strings.

For changes since the previous Cygwin version, see below,
/usr/share/doc/gettext/NEWS after installation, or the link:

https://git.savannah.gnu.org/gitweb/?p=gettext.git;f=NEWS;a=blob

Cygwin release 2 fixes library dependency issues with unistring.


Version 0.21.1 - October 2022

* The base Unicode standard is now updated to 14.0.0.

* Portability:
  - Building on macOS 11/arm64 is now supported.
  - Building on Linux/powerpc64le with glibc ≥ 2.35 is now supported.

* Runtime behaviour:
  - On AIX, locale names with a script or with an uppercase language are now
supported.
For example, sr_Cyrl_RS.UTF-8 is treated like sr_RS.UTF-8@cyrillic, and
EN_US.UTF-8 is treated like en_US.UTF-8.

[ANNOUNCEMENT] Updated: gsasl, libgsasl{18, -common, -devel, -doc} 2.2-2

2022-11-18 Thread Cygwin gsasl Maintainer 
The following packages have been uploaded for testing in Cygwin:

* gsasl 2.2-2
* libgsasl182.2-2
* libgsasl-common   2.2-2
* libgsasl-devel2.2-2
* libgsasl-doc  2.2-2

GNU SASL is an implementation of the Simple Authentication and Security
Layer (SASL) framework, consisting of a library with several plugins,
and command-line application.

For more information see the project home page:

https://www.gnu.org/software/gsasl/

See below or read /usr/share/doc/gsasl/NEWS after installation for
a summary of changes since the last Cygwin release, or read
/usr/share/doc/gsasl/ChangeLog after installation for details.

Cygwin release 2 fixes dependency problems affecting curl and git.


Noteworthy changes in release 2.2.0 2022-09-03

* Fix build error with too old GnuTLS.

* Tests: New tests/gsasl-mailutils-tls.sh.
  It performs integration checks between GNU SASL and GNU MailUtils
  imapd with TLS enabled, thereby testing TLS support in the 'gsasl'
  command line tool.

* Various minor bug fixes and improvements.
  Mainly to pacify improved CI/CD checking.


Noteworthy changes in release 2.1.1 2022-08-16 [beta]

* Tests: New tests/gsasl-mailutils-gs2krb5-gssapi.sh.
  It perform integration checks between GNU SASL and GNU MailUtils imapd
  (GSSAPI and GS2-KRB5). They can be used externally from the GNU SASL
  build environment to perform system integration tests, see
  .gitlab-ci.yml for inspiration.

* Various minor bug fixes and improvements.
  Fix two crashes in 'gsasl' introduced in 2.1.0.


Noteworthy changes in release 2.1.0 2022-08-05 [beta]

* Support new "tls-exporter" channel binding.
  The "tls-exporter" channel binding is specified in RFC 9266
  . Now we can support
  SCRAM-*-PLUS over TLS 1.3 channels, and address some of the security
  problems with "tls-unique".
  The library add new callback property GSASL_CB_TLS_EXPORTER and error
  code GSASL_NO_CB_TLS_EXPORTER. These are documented in the manual.
  The 'gsasl' command-line tool set it if system GnuTLS has
  GNUTLS_CB_TLS_EXPORTER, which was introduced with GnuTLS 3.7.2
  released on 2021-05-29.

* SCRAM: Support for "tls-exporter".
  The SCRAM client will now query the application for
  GSASL_CB_TLS_EXPORTER before it query for GSASL_CB_TLS_UNIQUE. Supply
  it to support TLS 1.3. The SCRAM server will query the application for
  the channel binding type requested by the client (tls-unique or
  tls-exporter), and it is up to the application to decide what to do.

* SCRAM: Fix memory leaks on incremental application usage.
  See tests/scram-incremental.c for application behaviour that trigger
  the leaks. We run valgrind --leak-check=full to catch future
  regressions.

* Tests: New tests/gsasl-dovecot-gssapi.sh & tests/gsasl-mailutils-cram.sh.
  These perform integration checks between GNU SASL and Dovecot
  (GSS-API) and GNU MailUtils imapd (CRAM-MD5, DIGEST-MD5, SCRAM-SHA-*).
  They can be used externally from the GNU SASL build environment to
  perform system integration tests, see .gitlab-ci.yml for inspiration.

* API and ABI modifications.
  GSASL_CB_TLS_EXPORTER: Added.
  GSASL_NO_CB_TLS_EXPORTER: Added.


Noteworthy changes in release 2.0.1 2022-07-15

* Support for the libgssglue GSS-API library were added.
  We encourage you to build with libgssglue, as that allows system
  administrators and end-users to chose between MIT Kerberos, Heimdal
  and GNU GSS during run-time. Read about the background here:
  https://blog.josefsson.org/2022/07/14/towards-pluggable-gss-api-modules/

* GSSAPI client: don't use AUTHID as fallback for AUTHZID.
  The code historically used the AUTHID as authorization identity, but
  in 2012 we changed it to first query for AUTHZID, and only if that is
  not available, fall back to using AUTHID as the authorization
  identity. The change was not released until version 1.8.1 on
  2019-08-02, when it was properly documented to be removed 'after the
  year 2012'. While documented behaviour, this seems like just
  surprising behaviour and we now finally make the change.

* GSSAPI server: don't set AUTHZID to empty string when absent.
  The GSS-API SASL protocol does not differentiate between an absent
  authorization identity and an authorization identity that is the empty
  string. Previously libgsasl would set it to the empty string but now
  it is set to NULL. The manual explains that this is a protocol
  limitation.

* The examples/smtp-server.c now supports GSSAPI/GS2-KRB5.
  The example is used during CI/CD testing of GNU SASL and thus it made
  sense to extend it. Some bugs related to getline error conditions were
  also fixed.

* GSSAPI server: Fix out-of-bounds read.
  A malicious client can after it has authenticated with Kerberos send a
  specially crafted message that causes Libgsasl to read out of bounds
  and cause a crash in the server.


Noteworthy changes in release 2.0.0

Updated: gsasl, libgsasl{18,-common,-devel,-doc} 2.2-2

2022-11-18 Thread Cygwin gsasl Maintainer 
The following packages have been uploaded for testing in Cygwin:

* gsasl 2.2-2
* libgsasl182.2-2
* libgsasl-common   2.2-2
* libgsasl-devel2.2-2
* libgsasl-doc  2.2-2

GNU SASL is an implementation of the Simple Authentication and Security
Layer (SASL) framework, consisting of a library with several plugins,
and command-line application.

For more information see the project home page:

https://www.gnu.org/software/gsasl/

See below or read /usr/share/doc/gsasl/NEWS after installation for
a summary of changes since the last Cygwin release, or read
/usr/share/doc/gsasl/ChangeLog after installation for details.

Cygwin release 2 fixes dependency problems affecting curl and git.


Noteworthy changes in release 2.2.0 2022-09-03

* Fix build error with too old GnuTLS.

* Tests: New tests/gsasl-mailutils-tls.sh.
  It performs integration checks between GNU SASL and GNU MailUtils
  imapd with TLS enabled, thereby testing TLS support in the 'gsasl'
  command line tool.

* Various minor bug fixes and improvements.
  Mainly to pacify improved CI/CD checking.


Noteworthy changes in release 2.1.1 2022-08-16 [beta]

* Tests: New tests/gsasl-mailutils-gs2krb5-gssapi.sh.
  It perform integration checks between GNU SASL and GNU MailUtils imapd
  (GSSAPI and GS2-KRB5). They can be used externally from the GNU SASL
  build environment to perform system integration tests, see
  .gitlab-ci.yml for inspiration.

* Various minor bug fixes and improvements.
  Fix two crashes in 'gsasl' introduced in 2.1.0.


Noteworthy changes in release 2.1.0 2022-08-05 [beta]

* Support new "tls-exporter" channel binding.
  The "tls-exporter" channel binding is specified in RFC 9266
  . Now we can support
  SCRAM-*-PLUS over TLS 1.3 channels, and address some of the security
  problems with "tls-unique".
  The library add new callback property GSASL_CB_TLS_EXPORTER and error
  code GSASL_NO_CB_TLS_EXPORTER. These are documented in the manual.
  The 'gsasl' command-line tool set it if system GnuTLS has
  GNUTLS_CB_TLS_EXPORTER, which was introduced with GnuTLS 3.7.2
  released on 2021-05-29.

* SCRAM: Support for "tls-exporter".
  The SCRAM client will now query the application for
  GSASL_CB_TLS_EXPORTER before it query for GSASL_CB_TLS_UNIQUE. Supply
  it to support TLS 1.3. The SCRAM server will query the application for
  the channel binding type requested by the client (tls-unique or
  tls-exporter), and it is up to the application to decide what to do.

* SCRAM: Fix memory leaks on incremental application usage.
  See tests/scram-incremental.c for application behaviour that trigger
  the leaks. We run valgrind --leak-check=full to catch future
  regressions.

* Tests: New tests/gsasl-dovecot-gssapi.sh & tests/gsasl-mailutils-cram.sh.
  These perform integration checks between GNU SASL and Dovecot
  (GSS-API) and GNU MailUtils imapd (CRAM-MD5, DIGEST-MD5, SCRAM-SHA-*).
  They can be used externally from the GNU SASL build environment to
  perform system integration tests, see .gitlab-ci.yml for inspiration.

* API and ABI modifications.
  GSASL_CB_TLS_EXPORTER: Added.
  GSASL_NO_CB_TLS_EXPORTER: Added.


Noteworthy changes in release 2.0.1 2022-07-15

* Support for the libgssglue GSS-API library were added.
  We encourage you to build with libgssglue, as that allows system
  administrators and end-users to chose between MIT Kerberos, Heimdal
  and GNU GSS during run-time. Read about the background here:
  https://blog.josefsson.org/2022/07/14/towards-pluggable-gss-api-modules/

* GSSAPI client: don't use AUTHID as fallback for AUTHZID.
  The code historically used the AUTHID as authorization identity, but
  in 2012 we changed it to first query for AUTHZID, and only if that is
  not available, fall back to using AUTHID as the authorization
  identity. The change was not released until version 1.8.1 on
  2019-08-02, when it was properly documented to be removed 'after the
  year 2012'. While documented behaviour, this seems like just
  surprising behaviour and we now finally make the change.

* GSSAPI server: don't set AUTHZID to empty string when absent.
  The GSS-API SASL protocol does not differentiate between an absent
  authorization identity and an authorization identity that is the empty
  string. Previously libgsasl would set it to the empty string but now
  it is set to NULL. The manual explains that this is a protocol
  limitation.

* The examples/smtp-server.c now supports GSSAPI/GS2-KRB5.
  The example is used during CI/CD testing of GNU SASL and thus it made
  sense to extend it. Some bugs related to getline error conditions were
  also fixed.

* GSSAPI server: Fix out-of-bounds read.
  A malicious client can after it has authenticated with Kerberos send a
  specially crafted message that causes Libgsasl to read out of bounds
  and cause a crash in the server.


Noteworthy changes in release 2.0.0

Re: RE: Cygwin x86 end-of-life

2022-11-18 Thread Jon Turney 

On 14/11/2022 21:29, Jason Pyeron wrote:

-Original Message-
From: Brian Inglis
Sent: Monday, November 14, 2022 3:17 PM

On Mon, 14 Nov 2022 16:25:18 +, Jon Turney wrote:> On 11/11/2022 16:16, Jon
Turney wrote:

On 11/11/2022 15:50, Jon Turney wrote:

As has previously been announced, Cygwin is dropping support for x86
Windows. Cygwin 3.3.6 is the final version supporting x86 (32-bit)
Windows, and the forthcoming Cygwin 3.4 will be released for x86_64 only.

Concurrent with that, updates to x86 packages will be stopped, and the
Cygwin x86 package repository will be archived.



I plan to pause package uploads this coming Monday (2022-11-14), before
starting the re-organization of the package repository to make this
archive.



Since there have been some complaints about short notice, and to give
time to work out the issues with gettext/libunistring, I'm going to
defer this by one week, until Monday 2022-11-21.


Thank you very much appreciated, hopefully we can deal with the remaining issues
quickly.


I do not have an articulate retort to ending support, but all I can say is that 
I feel there must be a middle ground.

I feel that there could and should be some form of "we don’t support it" but we 
are not going out of our way to prevent it.


I'm unclear if this means:

(a) "releasing Cygwin 3.4.x DLL for x86 OS"
(b) "don't require x86 uploads, but don't prevent them, either"

The problem I have with (b) is that we probably end up in a state where 
x86 is missing package updates people want or need; or broken, but we 
don't know it, because nobody uses it.



Can I throw resources at a solution? If so what?


Sure, if that's what you want to do.

According surveys, 32-bit Windows has a fraction of 1% market share, and 
declining. Our own (limited) metrics are in accord with that, so I 
basically see any time I spend on this as wasted.


So, the first resource you'll need provide is manpower.

Re: Sharing my packages for CYGWIN

2022-11-18 Thread Keith Christian 
Carlo,

Since you are successfully creating Cygwin packages, one thing that
would be very helpful is a detailed set of instructions for cygport to
create a Cygwin package from source.
There are instructions posted on the Cygwin website and other info
about cygport, but I have not found a comprehensive document that
makes packaging nearly "mistake-proof" for uninitiated packagers.
(Maybe I'm not looking in the right place.)
E.g., I want to create an up to date Ansible package for personal use,
as the current version in Cygwin is 2.8.4, and it appears Ansible is
up to at least 2.10.7, as that version is available in Debian
Bullseye.

Thanks.

-- 
Problem reports:  https://cygwin.com/problems.html
FAQ:  https://cygwin.com/faq/
Documentation:https://cygwin.com/docs.html
Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple

Re: [PATCH] Cygwin: pty: Fix 'Bad address' error when running 'cmd.exe /c dir'

2022-11-18 Thread Johannes Schindelin 
Hi Corinna,

On Mon, 24 Oct 2022, Corinna Vinschen wrote:

> However, two points:
>
> - I'm wondering if the patch (both of yours) doesn't actually just cover
>   a problem in child_info_spawn::worker().  Different runpath values,
>   depending on the app path being "cmd" or "cmd.exe"?  That sounds like
>   worker() is doing weird stuff.  And it does in line 400ff.
>
>   So, if the else branch of this code is apparently working fine for
>   "cmd" per Takashi's observation in
>   https://cygwin.com/pipermail/cygwin-patches/2022q4/012032.html, how
>   much sense is in the if branch handling "command.com" and "cmd.exe"
>   specially?  Wouldn't a better patch get rid of this extra if and
>   the null_app_name variable instead?

I never understood why the pcon code was allowed to be so Hydra-like as to
sprawl into corners far, far beyond `winsup/cygwin/fhandler*`.

FWIW I would be in favor of getting rid of this special handling (unless
it causes a regression). Given the recent experience, I expect Takashi to
want to work on this without any interference from my side.

Ciao,
Dscho

Re: [PATCH v2 1/2] Allow deriving the current user's home directory via the HOME variable

2022-11-18 Thread Johannes Schindelin 
Hi Corinna,

On Thu, 10 Nov 2022, Corinna Vinschen wrote:

> On Nov 10 16:16, Johannes Schindelin wrote:
>
> > On Mon, 24 Oct 2022, Corinna Vinschen wrote:
> >
> > > On Oct 23 23:04, Johannes Schindelin wrote:
> > > > On Tue, 18 Oct 2022, Corinna Vinschen wrote:
> > > > [...]
> > > > > That means, the results from the "env" method is equivalent to the
> > > > > "windows" method, just after checking $HOME.  That's a bit of a 
> > > > > downer.
> > > > >
> > > > > Assuming the "env" method would *only* check for $HOME, the user would
> > > > > have the same result by simply setting nsswitch.conf accordingly:
> > > > >
> > > > >   home: env windows
> > > >
> > > > Except when the domain controller is (temporarily) unreachable, e.g. 
> > > > when
> > > > sitting in a train with poor or no internet connection. Then that latter
> > > > approach would have the "benefit" of having to wait 10-15 seconds before
> > > > the network call says "nope".
> > > >
> > > > This particular issue has hit enough Git for Windows users that I found
> > > > myself being forced to implement these patches and run with them for the
> > > > past seven years.
> > > >
> > > > Given the scenario of an unreachable domain controller, I hope you agree
> > > > that the `env` support added in the proposed patches _has_ merit.
> > >
> > > Yes, I don't doubt an `env' method checking for $HOME even a bit.
> >
> > Cool!
> >
> > > I'm just not sure as far as HOMEDRIVE/HOMEPATH/USERPROFILE are
> > > concerned.  Those vars should be left alone, but we can't control that,
> > > so reading them from genuine sources is preferred.
> >
> > I do not recall the exact reasons because it has been a good while since I
> > worked on these patches. But I do remember that we had to have a fall-back
> > for the many scenarios in Git for Windows where `HOME` is not even set,
> > and we specifically had to add HOMEDRIVE/HOMEPATH handling because
> > USERPROFILE alone would lead to problems (IIRC there were plenty of
> > corporate setups where USERPROFILE pointed to a potentially-disconnected
> > network drive).
> >
> > > Sure, the downside in terms of the LDAP server is clear to me
> > >
> > > So I guess it's ok to allow the env method to read the values of those
> > > vars from the env.  I would just feel better if we urge the
> > > user to set $HOME and read that exclusively.
> >
> > I would feel better about that, too, if it was practical.
> >
> > But I cannot ask millions of Git for Windows users to please go ahead and
> > first configure their `HOME` variable correctly, it took much less time to
> > implement the patch we're discussing than asking all users individually
> > ;-)
> >
> > And since there is nothing specific about Git for Windows here, I expect
> > Cygwin users to benefit from this feature, too.
> >
> > With this context in mind, I would like to ask to integrate the patch
> > as-is, including the HOMEDRIVE/HOMEPATH and USERPROFILE fall-backs.
>
> Can't do that, sorry.  Two replies before I sent a necessary change,
> which needs inclusion first.

I am a bit confused. Do you need anything from me to move this along, i.e.
are those two replies you mention some emails I failed to address yet?

Thanks,
Dscho