Age | Commit message (Collapse) | Author | Files | Lines |
|
Now it should be more obvious when a package needs it as a dependency,
as it will fail loudly if it isn't declared as a tool.
While here, some duplicate dependencies on itstool were removed from the
MATE packages
|
|
-Fix issue #16
Not on changelog but, #19 is merged and removes one patch for us.
|
|
most of these simply extend matching from "aarch64" to "aarch64eb"
in various forms of code. most remaining uses in pkgsrc of
"MACHINE_ARCH == aarch64" are because of missing aarch64eb support,
such as most of the binary-bootstrap requiring languages like rust,
go, and java.
no pkg-bump because this shouldn't change packages on systems that
could already build all of these.
|
|
|
|
|
|
|
|
Changes:
0.9.10
------
* Fix a possible memory leak when used with ncurses and when resizing the
window. The terminal was reinitalized using setupterm() which would leak memory
and it is also very inefficient to do so, because it rereads the terminfo
entries.
* Fix a possible memory leak when used with ncurses and when exiting from an
interactive program and restoring the terminal for sfeed_curses.
* Fix an issue when the sidebar would be > 256 columns long the text on the row
would not be displayed. Especially noticable in the monocle and horizontal
layout in fullscreen on bigger screens.
* Fix a silent truncation which would make searching on titles with > 1024
bytes not work on these truncated bytes.
* Fix right-click plumb not respecting the environment variable
$SFEED_PLUMB_INTERACTIVE flag (but the keybinds did).
* Fix drawing/updating of the initial items scrollbar when switching to the
monocle layout.
* Fix to restore the terminal state more gracefully when it could not be
correctly initialized (for example with an invalid $TERM set).
* Fix for correctly decoding mouse buttons 6 and 7 (tilt scroll) using xterm
X10 decoding. These buttons were not used, but still :)
* Fix: when the pane items or scrollbar are outside the window to not draw them.
* Do not clearscreen() when the window is dirty.
This makes sure there is less flickering in xterm and urxvt, like on a window
resize or when adjusting the sidebar.
* Add the keybinds J (shift-j) and K (shift-k) to go one item up or down and
also open it for faster navigating and checking new items.
Feedback idea from Hadrien Lacour, thanks!
* Also allow scrolling when the position is exactly on the pane its scrollbar
and fix mouse button actions that don't depend on a position in a pane.
* Implement extended SGR mouse-decoding and also improve X10 mouse decoding:
* When the terminal emulator supports SGR mouse-encoding then it can
encode x, y values beyond 224 columns and rows. This is useful for big screens.
* Map the side-buttons (back, forward) to switch panes. Useful in the monocle
layout.
* Implement keymasks (shift, meta/alt, ctrl) and pass them to the function
mousereport().
* Improve resizing the sidebar:
* Avoid unnecessary redraws in some (rare) cases.
* Clamp the minimal and maximum value when adjusting the sidebar size
depending on the layout. This also fixes an issue in horizontal layout mode
when the sidebar is 1 column high. It would require 2 keypresses (instead of 1)
to increase the sidebar. This was an injustice to the fingers.
* After toggling showing new feeds or when reloading feeds try to select the
same feed again.
* Code-style improvements, typofixes and minor documentation improvements.
|
|
|
|
Changes:
0.9.22
------
* Fix a time parsing regression with the non-standard date format
"2021-02-03 05:13:03" (without T). This non-standard format is now
supported again.
* Fix/improve/refactor URI parsing and formatting. This now supports
non-authoritive URI like "magnet:" links. local file URIs like file:/// are
now also allowed (of course the user should be careful with file:/// URLs).
This now follows the RFC standard more closely (but not completely):
https://tools.ietf.org/html/rfc3986
* sfeed_update: just pass the feedurl directly as the base URL. This simplifies
the code and allows for more correct transformation of relative URLs. Note
though that feeds should not contain relative URLs though.
* sfeed_update: add an overridable parse() function, using sfeed(1) by default
This can be useful to more cleanly make connector scripts.
* sfeed_update: restructure the code and allow to more easily reuse parts of
the functions by including the script. Add an example script in the README
to allow more efficient parallel execution using the non-POSIX xargs -P
option.
* Documentation improvements: add examples, typo fixes, mandoc (lint) errors,
etc.
|
|
Changes:
0.9.9
-----
- mark as read: use the ID if the link is empty
- sfeed_content: fix unescaping function
- Add feature to process a sequence of keys automatically on startup via
$SFEED_AUTOCMD environment variable
- Add horizontal and monocle layout
- Misc bugfixes and improvements
|
|
|
|
Don't hard-code the build environment's $HOME as the basis for the default
user configuration path, and use consistent means to check for this path.
Don't install configuration files in $HOME during the build. This package
also expects C++11.
|
|
pkgsrc changes:
- Use https:// for MASTER_SITES too
Changes:
0.9.8
-----
- Misc bug fixes
|
|
pkgsrc changes:
- Use https:// also for MASTER_SITES
Changes:
0.9.21
------
- Do not convert UTF-16 surrogate pairs to an invalid sequence
- If baseurl is empty use the path from the feed by default in sfeed_update
- Misc bug fixes and code and documentation improvements
|
|
Changelog
=========
3.2.2 2021-01-07
Bugfix: Header field name is not found by canlock-mhp if
there is another header field for which only the prefix
differs (and that other field is processed first)
Test suite for canlock-mhp parser extended: It now tests
headers larger than 4 KiByte and for the field prefix bug
fixed in this version
Lexer of canlock-hfp has problems to detect T_CTEXT tokens on
HP-UX (relevant only if the header field contains comments).
The syntax of the ERE for T_CTEXT was rewritten to be
compatible with the SysV lex of HP-UX 11.11
Test suite for canlock-hfp parser extended to test comments
harder
Namespace clash of internal library functions with libc on
NetBSD
Changed internal API function names hmac* to RFC2104Hmac*
Changed internal API function names hkdf* to RFC5869Hkdf*
Library API and ABI are fully compatible if no undocumented
functions are used
libcanlock-3.pc: Changed protocol in URL from http to https
|
|
pkgsrc patch removed.
Changelog
=========
3.2.2 2021-01-07
Bugfix: Header field name is not found by canlock-mhp if
there is another header field for which only the prefix
differs (and that other field is processed first)
Test suite for canlock-mhp parser extended: It now tests
headers larger than 4 KiByte and for the field prefix bug
fixed in this version
Lexer of canlock-hfp has problems to detect T_CTEXT tokens on
HP-UX (relevant only if the header field contains comments).
The syntax of the ERE for T_CTEXT was rewritten to be
compatible with the SysV lex of HP-UX 11.11
Test suite for canlock-hfp parser extended to test comments
harder
Namespace clash of internal library functions with libc on
NetBSD
Changed internal API function names hmac* to RFC2104Hmac*
Changed internal API function names hkdf* to RFC5869Hkdf*
Library API and ABI are fully compatible if no undocumented
functions are used
libcanlock-3.pc: Changed protocol in URL from http to https
|
|
|
|
Thanks to @leot for bring it up
|
|
|
|
neix is a simple, work in progress terminal feed reader for all common RSS/Atom
feeds on the web out there.
So you can read your news without advertisments or other annoying stuff.
Just the informations you need. You can import your existing collection of feeds
from an OPML file or just configure them manually.
You also have the ability to render the text with w3m, elinks or which program
you want.
|
|
Changes:
0.9.7
-----
- Permit to build with custom themes via SFEED_THEME
- Misc bug fixes
|
|
Removed POSIX shell patch (merged upstream).
Removed libcanlock patches (now supported via pkg-config).
-- 2.4.5 release 20201224 "Glen Albyn" --
022) Dennis Preiser <dennis@d--p.de>
BUG. 'B'ody and '/'article search with wildcard matching (wildcard=0)
didn't work
BUG. convert_tex2iso with UTF-8 as local charset did map "O to wrong
char
BUG. missing termination of conversion table for convert_tex2iso
BUG. avoid possible endless loop in search_group
BUG. tin_ltoa in some cases returned an error where it shouldn't
BUG. 'T'ag all parts of multipart message didn't work with thread_arts
4 (multipart) or 5 (percentage)
BUG. hung on empty (lone '.') positive (220) response to ARTICLE command
ADD. make {Group,Thread}TagParts 'T' toggle like 't'ag
ADD. don't show mime parse error for missing end marker if not running
in debug mode
ADD. increase hash-table sizes
ADD. speed up multipart threading
ADD. multipart tagging ThreadTagParts 'T' to thread level
ADD. reduce memory usage by struct reordering to avoid padding
ADD. GNKSA_MISSING_REALNAME return code for filtering
REM. redundant conde
FIX. art.c, charset.c, group.c, hashstr.c, help.c, keymap.c, lang.c
misc.c, page.c, refs.c, rfc2046.c, screen.c, search.c, string.c
tags.c, thread.c, wildmat.c, extern.h, keymap.h, proto.h, tin.h
tin.[15]
021) Thomas E. Dickey <dickey@invisible-island.net>
ADD. update configure macros
ADD. fallback to the bundled pcre if required
ADD. prefer systemwide libcanlock over bundled one
FIX. aclocal.m4, configure[.in], Makefile.[in], tin.h
020) Urs Janssen <urs@tin.org>
ADD. config.guess, config.sub update
ADD. document strip_re_regex, strip_was_regex and col_message in manpage
ADD. configure check for fchmod(2) and fallback to chmod(2) if system
lacks it (e.g. Interactive UNIX)
ADD. speed up filter on Path:-header
ADD. allow abortion of subject/author search in group level
ADD. allow abortion of filtering
ADD. raise default max length for numbers in prompts from 6 to 10 digits
ADD. restrict multipart detection to messages with more than 1 part
ADD. show an error message if append_mail() fails
ADD. allow path for posted_articles_file
ADD. document gnksa-parser return codes in tin(5)
ADD. center_line() now does BiDi
ADD. don't fail on GNKSA_MISSING_REALNAME when posting
ADD. GNKSA_RANGLE_MISSING (104) error code to gnksa-filter
BUG. when $TIN_INDEX_NEWSDIR was used INDEX_NEWSDIR was
automatically appended to the string, but for the '-I' cmd.line it
wasn't, now '-I' behaves exactly like $TIN_INDEX_NEWSDIR
REM. redundant condition
FIX. art.c, attrib.c, config.c, debug.c, filter.c, getline.c, inews.c
init.c, lang.c, main.c, makecfg.c, memory.c, misc.c, newsrc.c
nrctbl.c, options_menu.c, post.c, read.c, save.c, screen.c
search.c, tags.c, thread.c, tmpfile.c, extern.h, proto.h, tin.h
tin.[15], configure[.in], autoconf.h[in], config.guess, config.sub
en_GB.po, keymap.sample
019) Matt Anton <tin@lv223.org>
ADD. updated french translation
FIX. fr.po
018) Toomas Soome <tsoome@me.com>
ADD. updated estonian translation
FIX. et.po
|
|
Changes in 2.6.3
* Fixed the selection of the elliptic curve to use with OpenSSL 1.1.0 or
later; NIST P-256 was enforced instead of using the most secure curve.
* A new inn.conf parameter has been added to fine-tune the cipher suites
to use with TLS 1.3: the *tlsciphers13* now permits configuring them.
A separate cipher suite configuration parameter is needed for TLS 1.3
because TLS 1.3 cipher suites are not compatible with TLS 1.2, and
vice-versa. In order to avoid issues where legacy TLS 1.2 cipher
suite configuration configured in the *tlsciphers* parameter would
inadvertently disable all TLS 1.3 cipher suites, the inn.conf
configuration has been separated out.
* Fixed a regression since INN 2.6.1 that prevented articles with
internationalized header fields (that is to say encoded in UTF-8) from
being posted.
* Support for Python 3 has been added to INN. Embedded Python filtering
and authentication hooks for innd and nnrpd can now use version 3.3.0
or later of the Python interpreter. In the 2.x series, version 2.3.0
or later is still supported.
When configuring INN with the --with-python flag, the "PYTHON"
environment variable, when set, is used to select the interpreter to
embed. Otherwise, it is searched in standard paths.
In case you change the Python interpreter to embed, make sure that the
Python scripts you use are written in the expected syntax for that
version of the Python interpreter. Notably, buffer objects have been
replaced with memoryview objects in Python 3, and UTF-8 encoding now
really matters for string literals (Python 3 uses bytes and Unicode
objects).
INN documentation and samples of Python hooks have been updated to
provide more examples.
* When a Python or Perl filter hook rejects an article, innd now
mentions the reason in response to CHECK and TAKETHIS commands.
Previously, the reason was given only for the IHAVE command.
* nnrpd now properly logs the hostname of clients whose connection
failed owing to an issue during the negotiation of a TLS session or
high load average.
Changes in 2.6.2
* A new *syntaxchecks* parameter has been added in inn.conf. It permits
controlling the level of checks performed by innd and nnrpd. Up to
now, only one check can be enabled/disabled: when *laxmid* is
mentioned in the values of this new parameter, INN accepts Message-IDs
that contain ".." in the left part, as well as Message-IDs with two
"@" (such Message-IDs would otherwise be considered as syntactically
invalid). See the inn.conf(5) man page for more details.
The check is disabled by default (*no-laxmid*), which corresponds to
the legacy behaviour of INN 2.6.1 and earlier.
* Use of the ovdb_server helper server is now the default when using the
ovdb overview method, that is to say the default value for the
*readserver* parameter in ovdb.conf is now set to true. It improves
stability and avoids deadlocks, timing issues and corrupted ovdb
databases.
* mailpost now removes empty header fields before attempting to post
articles, and keeps trace of them in the X-Mailpost-Empty-Hdrs: newly
generated header field body. Also, mailpost now sanitizes header
fields with regards to empty continuation header lines. Thanks to
Kamil Jonca for these bug reports.
* A new -z parameter has been added to mailpost to mention a list of
header fields to remove from the gated message. Thanks to Dieter
Stussy for the patch.
* Fixed a bug in inews that was rejecting articles containing header
fields whose length exceeded 998 bytes. This limitation is for the
length of a single line of a header field (and not for the length of
the whole header field, as it was wrongly the case).
* Added support for GnuPG's gpg binary (in addition to gpgv) in
pgpverify. Indeed, gpg still validates signatures made with weak
digest algorithms like MD5 whereas gpgv no longer do. Thanks to
Thomas Hochstein for the patch, which permits validating control
articles for hierarchies that are still using old PGP keys.
* Added similar support for GnuPG's gpg binary in perl-nocem to validate
NoCeM notices from issuers who are still using old PGP keys.
* A few commands listed in the "Control commands to INND" section in
daily Usenet reports were appearing as a mere letter; all of them are
now properly converted to meaningful words.
* The *tlsprotocols* parameter in inn.conf now recognizes the "TLSv1.3"
value (for OpenSSL versions implementing TLS 1.3, that is to say
starting from OpenSSL 1.1.1).
* The buffindexed overview method will now hopefully work properly on
systems with a native page size larger than 16KB.
* Other minor bug fixes and documentation improvements.
|
|
|
|
Changes
0.9.6
-----
- Add keybinds < (decrease), > (increase) and = (reset to automatic
size) to resize feed sidebar pane for small terminals
- Make lazyload a run-time configuration option via SFEED_LAZYLOAD
environment variable
|
|
For the Python 3.8 default switch.
|
|
|
|
No changes. Use same tarball as canlock-hp.
|
|
Changelog
=========
3.2.1 2020-11-28
Bugfix: off-by-one heap buffer overflow fixed in canlock-mhp
(If header is larger than 4 KiByte. Found by Valgrind)
Bugfix: canlock-mhp end of header detection fixed
(Found by test suite on openSUSE. Reported by Martin Hauke)
|
|
Changes:
- sfeed_xmlenc: be more paranoid in printing encoding names
- sfeed_web: improve parsing <link> if it has no type attribute
|
|
Prevent stack overruns if /message.id contains filenames >= 1000.
Bump PKGREVISION.
|
|
- MIME support: Implement decoding quoted-printable
- View command to view multiple messages at once
- Remember received articles if interrupted or timed out
- Post to multiple servers status still Open with 5 other changes
- Check for errors properly after reading a dot at the beginning of a line
from the NNTP server; also add a "G" progress indicator
|
|
- Searching for Unicode string in current article is now supported (case
insensitive variant based on Unicode case folding algorithm)
- Unicode database updated to version 13.0.0 (CaseFolding.txt added)
- The new 'force_unicode' entry in configfile set to a nonzero value sends
all outgoing articles in Unicode, if they contain non-ASCII characters
- Dark background color is now usable
- MIME conformance reached (as defined in RFC 2049 Section 2)
- If selected text is present, quote only selected text for followup
- XDG desktop entry and icon theme added (option CFG_XDG_DISABLE)
- Configuration window in GUI now allows to disable the "User-Agent" header
field for outgoing articles
- It is now possible to enable AUTHINFO USER/PASS without encrypted connection
(option CFG_NNTP_AUTH_UNENCRYPTED)
This allows to use stunnel instead of the internal TLS module
- TLS module no longer calls deprecated functions HMAC*(), RSA_size() and
SSL_get_peer_certificate() for OpenSSL API 3.0.0
|
|
Changes:
0.9.5
-----
- Minor bug fixes and improvements
|
|
|
|
Changes:
0.9.4
-----
- Add configurable interactive modes for executed programs (executed
program interactive in the same terminal)
- Add support for new categories field
- Misc bug fixes and improvements
|
|
0.9.19
------
- Parse categories and add multiple field values support (for categories)
separated by `|'
- Misc bug fixes and improvements
|
|
Changes:
0.9.3
-----
- Several bug fixes and performance improvements
|
|
|
|
|
|
Changes:
0.9.2
-----
- Fix scrolling issues on some terminals
- Fix a file descriptior leak on startup when stdin is not a tty
- Improve cursor visibility and behaviour on some terminals
- Misc fixes and improvements
0.9.1
-----
- Add pledge support for OpenBSD
- Add support to SFEED_YANKER to yank the URL/enclosure to a program (by
default `xclip -r')
- Fix a crash with small window width and garbage data if utf8pad fails
|
|
|
|
Changes:
0.9
---
- Add sfeed_content
- Misc bug fixes and improvements
0.8.2
-----
- Reread urls if using an url list for mark read and loading a feed
- Rework updatesidebar drawing and width calculation
0.8.1
-----
- Misc bug fixes (cursor problem with urxvt and reloadall and handling of the
current open feed)
|
|
|
|
MultiMail is an offline mail packet reader, supporting the Blue Wave,
QWK, OMEN, SOUP and OPX formats. It uses a simple curses-based
interface.
SOUP is used for Internet email and Usenet. The other formats are
primarily used with dialup (or telnet) BBSes, to save connect time and
to provide a better interface to the message base.
|
|
No PKGREVISION bump since it should failed on all platforms where there
is not a native curses and/or only worked by accident.
|
|
|
|
sfeed_curses is a curses UI front-end for sfeed.
It shows the TAB-separated feed items in a graphical command-line UI.
The interface has a look inspired by the mutt mail client. It has a
sidebar panel for the feeds, a panel with a listing of the items and a
small statusbar for the selected item/url. Some functions like
searching and scrolling are integrated in the interface itself.
|
|
Changes:
0.9.18
------
- format tools: don't skip items with a missing/invalid timestamp field
- sfeed_atom: the updated field is mandatory: use the current time
- sfeed_atom: fix timezone, output if timestamp is set
- sfeed_atom: make the output more conform by adding id, updated tags and add
<feed> tags (author name, id, updated, title).
Thanks lich for the feedback and testing.
- sfeed_gopher: do not use URL: prefix for gopher:// urls.
- Documentation improvements
|
|
|