Excerpt of changes (+portability and bug fixes):
* Add support for `--la-file' argument to neon-config, which prints the
full path of the installed libneon.la file.
* Improved address resolver (ne_addr_*) replacing ne_name_lookup():
- use getaddrinfo() if found; include support for IPv6 (based on work
by Noriaki Takamiya)
* For a hostname with multiple addresses, each address is tried in turn
until a connection is made.
* RFC2818 compliance for certificate identity checks in SSL:
- use `dNSname' values in subjectAltName extension if present
- hostname comparison fixed to not be case-sensitive
* Added NE_DBG_SSL debug channel.
* ne_strerror changed to return the passed-in buffer.
* Added ne_strnzcpy macro to ne_string.h.
* Added reference documentation:
- ne_sock_init, ne_addr_*.
* Remove the const qualifier from the reason_phrase field in ne_status.
- ne_parse_statusline() now strdup's the reason_phrase
* Remove the status_line argument from ne_207_end_propstat and _end_response
* Change ne_session_create, ne_session_proxy, ne_sock_connect, and the
'port' field of the ne_uri structure to use an unsigned int for port numbers
* ne_uri_defaultport returns unsigned and '0' on an unknown port (not -1).
* Changes to hooks interface:
- pass an ne_request pointer to per-request hooks
- replace "accessor" hooks with ne_{get,set}_{request,session}_private
* Authentication changes:
- the hooks changes fix a segfault if auth is enabled for an SSL session
through a proxy server
- fix ne_forget_auth segfault if either proxy or server auth are not used
* Improvements to persistent connection retry logic and error handling
in request code; fixing some cases where some errors where incorrectly
treated as a persistent connection timeout
- a TCP RST at the appropriate time is now treated as a persistent
connection timeout.
- handle persistent connection timeouts on SSL connections
* Changes to SSL support:
- improved error handling
- fix for proxy CONNECT tunnelling with some proxies (e.g. Traffic-Server)
- fix potential segfault if client cert. provider callback is used
- fix to use supplied password callback for PEM-encoded client certificates
(Daniel Berlin)
* Remove ne_read_file().
* ne_version_match replaces ne_version_minimum (semantics changed slightly).
* XML request bodies use a content-type of "application/xml" now;
applications can use NE_XML_MEDIA_TYPE from ne_xml.h
update. Excerpt of changes:
Changes with APR 0.9.2
*) Renames done (deprecated functions wrapped):
apr_filename_of_pathname -> apr_filepath_name_get
apr_get_groupid -> apr_gid_get
apr_get_groupname -> apr_gid_name_get
apr_compare_groups -> apr_gid_compare
apr_parse_addr_port -> apr_port_addr_parse
apr_shutdown -> apr_socket_shutdown
apr_bind -> apr_socket_bind
apr_listen -> apr_socket_listen
apr_accept -> apr_socket_accept
apr_connect -> apr_socket_connect
apr_send -> apr_socket_send
apr_sendv -> apr_socket_sendv
apr_sendto -> apr_socket_sendto
apr_implode_gmt -> apr_time_exp_gmt_get
apr_get_home_directory -> apr_uid_homepath_get
apr_get_userid -> apr_uid_get
apr_current_userid -> apr_uid_current
apr_compare_users -> apr_uid_compare
apr_get_username -> apr_uid_name_get
apr_recvfrom -> apr_socket_recvfrom
apr_sendfile -> apr_socket_sendfile
apr_recv -> apr_socket_recv
[Thom May]
*) Add APR_IPV6_V6ONLY socket option. [Jeff Trawick]
*) Update timeout algorithm in free_proc_chain. If a subprocess
did not exit immediately, the thread would sleep for 3 seconds
before checking the subprocess exit status again. In a very
common case when the subprocess was an HTTP server CGI script,
the CGI script actually exited a fraction of a second into the 3
second sleep, which effectively limited the server to serving one
CGI request every 3 seconds across a persistent connection.
[Bill Stoddard, Kai.Risku@arrak.fi]
*) Add recognition of and socket API support for the SCTP protocol.
[Randall Stewart <randall@stewart.chicago.il.us>]
*) Add apr_atomic_casptr() to support atomic compare-and-swap
of pointers [Brian Pane]
*) Add apr_socket_create_ex() to allow protocol to be specified for the
socket. With APR 1.0, this function will be removed and apr_socket_create()
will have the additional parameter.
[Randall Stewart]
*) Add APR_IPV4_ADDR_OK flag to apr_sockaddr_info_get() to allow
apps to avoid lookup of IPv6 address if IPv4 address is sufficient.
(New APR_IPV6_ADDR_OK flag is similar.) [Jeff Trawick]
Changes with APR 0.9.1
*) Add apr_array_pop(). [Justin Erenkrantz]
Changes with APR 0.9.0
*) Includes moved to INCLUDEDIR/apr-{major} (e.g. /usr/include/apr-0)
[Greg Stein]
*) libtool versioning is used to give the library sonames a real
value. The libraries will be libapr-{major}.so.0.{minor}.{patch}
[Greg Stein]
*) Add APR_PARSE_ARGUMENTS and APR_LAYOUT macros for better layout
support. [Thom May]
*) Add parallel-apr layout which utilizes the major version number in
directories and library names. [Justin Erenkrantz]
*) Add a version number to the library name (e.g. libapr-1.so) so
that apps can do things like: -lapr-1 or -lapr-2, depending on
which version they want to use and link against. [Greg Stein]
*) Add --version to apr-config so that apps can retrieve the version
information of the (installed) APR. [Greg Stein]
*) Remove the APRVARS system; apps should use apr-config. [Greg Stein]
*) Renamed apr_strtoll()/apr_atoll() to follow int64 convention,
so these new helpers are apr_strtoi64/apr_atoi64(), since
'll' (long long) is a nonportable and aspecific construct.
Used ac/m4 tests to choose the appropriate fn behind strtoi64.
[William Rowe]
As well as many portabitility and other fixes.
directly into libfltk.{la,so} so that we don't need to guess at which
additional libraries will be needed to link into a program when linking
against libfltk. Suggested in private email by Dave Sainty <dave at
dtsp dot co dot nz>.
Tnm changes since 2.1.10:
15/6/01 (bug fix) Fixed a buffer overrun in ntping. This is actually
the reason to make this bugfix release public.
2/11/99 (bug fix) Make sure that remaining job times do not increase if
the system clock moves backwards.
17/5/99 (bug fix) Make sure that commands bound to a recv event are
always evaluated.
6/4/99 (bug fix) Fixed a Y2K bug in tnmHttp.c (thanks to bkozuma@aol.com
for finding this problem).
23/3/99 (bug fix) Accept derived types (e.g. DisplayString) in a varbind
list.
9/3/99 (bug fix) Fixed bugs in straps.c which could cause security
problems due to buffer overruns or signed/unsigned conversion.
8/3/99 (new feature) Added the global tnm(cache) variable which points
to a directory where Tnm saves frozen MIB files. The default value for
tnm(cache) is compatible with previous 2.1.X releases.
Tkined changes since 1.4.10:
21/4/99 (bug fix) The size of the canvas is now independent from the
resolution of the display. Further, PostScript dumps should now fit
on the selected page size.
(christianbiere at gmx dot de) in PR pkg/19152.
Changes since 0.91.0:
* Will now verify the SHA1 of downloaded files in the background.
* New configuration parameter for the directory where "bad files" not
matching their advertised SHA1 should be put (don't forget to
configure it, as it defaults to /tmp).
* Files are now copied in the background, so it is possible to setup the
various "download" directories on different filesystems.
* Node connection time and remote node's uptime(if available) are now
shown.
* It is now possible to paste a "magnet:" URI in the search to launch a
SHA1 search.
* The "connection indicator" at the bottom left is now a toggle.
Depressing it will terminate any Gnet connection but will continue to
serve uploads and process downloads.
* The active download status now shows how many sources are available and
how many are currently active (i.e. downloading). It will also show the
estimated remaining time for the whole file, as well as the global
reception rate for the file, accounting for all currently active
sources.
* Fixed bug that prevented proper downloading from recent
BearShare(4.1.x or better) and Shareaza (1.6.0.x or better) nodes
when the SHA1 of the file is known.
* Will now protect against "PUSH floods" by banning the relevant IPs.
* At startup, scan the download "tmp" dir for orphan files and reintegrate
them into the fileinfo database, moving completed files to the relevant
directory(depending on whether its SHA1 checked OK).
* Minor bug fixing: improper countdown for push timeout when falling back,
fixed corner-case processing that could lead to crashes or incorrect
behaviour, accidental localization on X-Live-Since headers, freezing of
the GUI when clearing many uploads, obscure portability fixes.
- Apply disabled official patch since the patch's content has corrected.
* Impossible to define acls with spaces in them
- Remove "@unexec ${RMDIR} %D/etc/squid ..." line from PLIST since
there is already removing directory line which use more generic
PKG_SYSCONFDIR variable.
- include bsd.pkg.mk explicitly, rather than in the JRE's Makefile.common
- be more thorough at avoiding installation of files that are also
found in the JRE
- include bsd.pkg.mk explicitly, rather than in Makefile.common
- replace dependency on linuxppc_lib with suse_base to get a
sufficiently recent glibc
- teach the remaining two usages of `uname -m` output about our
various powerpc platforms
This makes the Blackdown JRE finally work on powerpc again.
Note that suse_base requires a Linux kernel version >= 2.2.
NetBSD-current's COMPAT_LINUX reports itself as 2.4.18 by default
on i386 and powerpc. On NetBSD/powerpc 1.6, you'll need to override
the default value of the "emul.linux.kern.osrelease" sysctl. On
NetBSD < 1.6, you can try setting the sysctl, but the emulation
may be insufficient to run this software properly.
Also note that on powerpc, you must specify the "-green" option,
because the Classic VM hangs.
* correct inverted logic of restrictions table which made "-restrict=default"
provide incorrect values for several items. This was broken in 2.8.4dev.19
(reported by Jeff Long <long@ukans.edu> and RobertM <robm@bob.bofh.org>) -TD
* correct check for calling endwin() to allow for curses implementations
without newterm (report/patch by Brett Lymn).
Here are some of them, excerpted from NEWS:
- New code to handle multibyte characters.
- `select' was changed to be more ksh-compatible
- There is now a bindable edit-and-execute-command readline command,
like the vi-mode `v' command, bound to C-xC-e in emacs mode.
- The shell now performs arithmetic in the largest integer size the
machine supports (intmax_t), instead of long.
- There is a new configuration option `--enable-mem-scramble', controls
bash malloc behavior of writing garbage characters into memory at
allocation and free time.
- The `complete' and `compgen' builtins now have a new `-s/-A service'
option to complete on names from /etc/services.
- `read' has a new `-u fd' option to read from a specified file descriptor.
- The expansion of $LINENO inside a shell function is only relative to the
function start if the shell is interactive -- if the shell is running a
script, $LINENO expands to the line number in the script. This is as
POSIX-2001 requires.
- The bash debugger in examples/bashdb has been modified to work with the
new DEBUG trap semantics, the command set has been made more gdb-like,
and the changes to $LINENO make debugging functions work better. Code
from Gary Vaughan.
- New [n]<&word- and [n]>&word- redirections from ksh93 -- move fds (dup
and close).
- The `echo' builtin now accepts \0xxx (zero to three octal digits following
the `0') in addition to \xxx (one to three octal digits) for SUSv3/XPG6/
POSIX.1-2001 compliance.
- Added support for DESTDIR installation root prefix, so you can do a
`make install DESTDIR=bash-root' and do easier binary packaging.
- New `-A group/-g' option to complete and compgen; does group name
completion.
- The ksh-like `ERR' trap has been added. The `ERR' trap will be run
whenever the shell would have exited if the -e option were enabled.
It is not inherited by shell functions.
- configure has a new `--enable-largefile' option, like other GNU utilities.
- `for' loops now allow empty word lists after `in', like the latest POSIX
drafts require.
- The builtin `ulimit' now takes two new non-numeric arguments: `hard',
meaning the current hard limit, and `soft', meaning the current soft
limit, in addition to `unlimited'
Also, there is a "New unwind-protect implementation from Paul
Eggert", which I believe obviates the need for two sparc64-related
patches.
Changes since 1.11.1p1:
* The "log" and "rlog" commands now have a -S option to suppress the
header information when no revisions are selected.
* A serious error that allowed read-only users to tag files has been
corrected.
* The "annotate" command will no longer annotate binary files unless
you specify the new -F option.
* The "tag" and "rtag" commands will no longer move or delete branch
tags unless you use the new -B option. (This prevents accidental
changes to branch tags that are hard to undo.)
* We've standardized on the 1.5 Automake release for the moment. Again, this
should only really affect developers. See the section of the INSTALL file
about using the autotools if you are compiling CVS yourself.
Changes from 1.11.1 to 1.11.1p1:
* Read only access was broken - now fixed.
Changes from 1.11 to 1.11.1:
* The "cvs diff" command now accepts the -y/--side=by-side and -T/
--initial-tab options. (To use these options with a remote repository,
both the client and the server must support them.)
* The expansion of the loginfo format string has changed slightly.
Previously, the expansion was surrounded by single quotes ('); if a file
name contained a single quote character, the string would not be parsed
as a single entity by the Unix shell (and it would not be possible to
parse it unambiguously). Now the expansion is surrounded by double
quotes (") and any embedded dollar signs ($), backticks (`), backslashes
(\), and double quotes are preceded by a backslash. This is parsed as a
single entity by the shell reguardless of content. This change should
not be noticable unless you're not using a Unix shell or you have
embedded the format string inside a double quoted string.
* There was a bug in the diff code which sometimes caused conflicts to
be flagged which shouldn't have been. This has been fixed.
* New "cvs rlog" and "cvs rannotate" commands have been added to get log
messages and annotations without having to have a checked-out copy.
* Exclusive revision ranges have been added to "cvs log" using ::
(similar to "cvs admin -o").
* The VMS client now accepts wildcards if you're running VMS 7.x.
* ZLIB has been updated to version 1.1.3, the most current version. This
includes mostly some optimizations and minor bug fixes.
* The ~/.cvspass file has a slightly modified format. CVSROOTs are now
stored in a new canonical form - hostnames are now case insensitive and
port numbers are always stored in the new format. Until a new login for
a particular CVSROOT is performed with the new version of CVS, new and
old versions of CVS should interoperate invisibly. After that point, an
extra login using the old version of CVS may be necessary to continue to
allow the new and old versions of CVS to interoperate using the same
~/.cvspass file and CVSROOT. The exception to this rule occurs when the
CVSROOTs used with the different versions use case insensitively
different hostnames, for example, "empress", and "empress.2-wit.com".
* A password and a port number may now be specified in CVSROOT for
pserver connections. The new format is:
:pserver:[[user][:password]@]host[:[port]]/path
Note that passwords specified in a checkout command will be saved in the
clear in the CVS/Root file in each created directory, so this is not
recommended, except perhaps when accessing anonymous repositories or the
like.
* The distribution has been converted to use Automake. This shouldn't
affect most users except to ease some portability concerns, but if you
are building from the repository and encounter problems with the
makefiles, you might try running ./noautomake.sh after a fresh update
-AC.
This is needed so fetchmail properly detects shark and cats as arm boxen.
Note it does mean that we now give warnings about missing, but there's no
glue in mk/bsd.pkg.mk to link the gnu-config/missing into a package.
Maybe there should be.
Tested on shark, cats and i386.
Changes since 2.3.0 are too numerous to mention all here. Refer to the homepage
for details.
The most important changes:
* Project page was moved to sourceforge
* Implemented user-defined columns in TV schedule output, which allow
to define the column content depending on matches against a number
of filter shortcuts; the content can consist of static text,
pre-defined images or a standard attribute. See the manual or this
screenshot for details.
* Implemented inverted filters, i.e. you can now search for all
programmes which do not match one or more of given filter criteria
or simply invert the outcome of a complete filter setting.
* Added a new filter type: "VPS/PDC" which allows to search for
programmes with a VPS/PDC timestamp attached, and/or whose VPS/PDC
start time differs from the actual start time (i.e. shifted or newly
added programmes)
* Enhanced start time filter: added an option "ignore date" which
allows to search for programmes in a daily time window
(e.g. restrict a search for German new magazine "Tagesschau" to
20:00-20:10)
* Added database export in XML format (according to xmltv.dtd), provided by
Massis Sirapian.
* Bugfix for Power-PC architecture in pioutput.c; Thanks to Gerhard Tonn
* Internal changes: split Tcl/Tk script for GUI into modules (the
* daemon only loads the script which manages the rcfile - saves
* apx. 400 kB RAM)
Plus many bug fixes.
