summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorTatsuhiro Tsujikawa <tatsuhiro.t@gmail.com>2017-01-16 22:34:28 +0900
committerTatsuhiro Tsujikawa <tatsuhiro.t@gmail.com>2017-01-16 22:34:28 +0900
commit36902526b66907f1f4b0add9900d8a82521b45d4 (patch)
tree53703fdd86a2b42bf399ec7d0484de24f4157adb
parentdc84ca14e5951646e49d2eaccde9f9b8a7841a44 (diff)
Update NEWSrelease-1.31.0
-rw-r--r--NEWS120
1 files changed, 15 insertions, 105 deletions
diff --git a/NEWS b/NEWS
index 47e01edf..f4c45b00 100644
--- a/NEWS
+++ b/NEWS
@@ -1,121 +1,31 @@
-aria2 1.30.0
+aria2 1.31.0
============
Release Note
------------
-This release fixes several bugs. See Changes below for detailed bug
-fixes. This release adds an ability to expand `${HOME}` to user's
-home directory in several options. It would be useful when used in
-configuration file. -S option against torrent file now prints out DHT
-nodes.
+This release fixes assertion error in
+SimpleRandomizer::getRandomBytes. It adds new option
+content-disposition-default-utf8.
Changes
-------
-* mingw: Use sqlite 3.15.2
+* Better error message when local file status cannot be retrieved
-* Fix bug that --disable-websocket causes build error
+ GH-836
-* Fix bug that lower bound of optimize-concurrent-downloads becomes 1
+* Fix assertion failure in SimpleRandomizer::getRandomBytes
- GH-798
+ errno might not be initialized to 0, and we may get both rv == -1
+ and errno != ENOSYS. This leads to assertion failure. Since
+ getrandom_linux always returns -1 on failure, checking errno is
+ useless in this function.
-* Print DHT nodes with -S option
+ GH-823
- GH-795
+* Add option content-disposition-default-utf8
-* Take into account address family when resolving DHT node address
+ Patch from JimmyZ
- GH-794
-
-* Allow block device file to -i option
-
- GH-785
-
-* Expand `${HOME}` to user's home directory in several options
-
- The following options implement this substitution:
-
- * --ca-certificate
- * --certificate
- * --dht-file-path
- * --dht-file-path6
- * --dir
- * --input-file
- * --load-cookies
- * --log
- * --metalink-file
- * --netrc-path
- * --on-bt-download-complete
- * --on-download-complete
- * --on-download-error
- * --on-download-start
- * --on-download-stop
- * --on-download_pause
- * --out
- * --private-key
- * --rpc-certificate
- * --rpc-private-key
- * --save-cookies
- * --save-session
- * --server-stat-if
- * --server-stat-of
- * --torrent-file
-
- GH-780
-
-* Document that libaria2 is not thread safe
-
- GH-775
-
-* configure: Use pkg-config to find libs whenever possible
-
- Use pkg-config to find cppunit, libxml2 and expat. All those
- libraries provide pkg-config files, and in all of those cases the
- pkg-config macro is superior to the custom macros used currently.
-
- The advantages of pkg-config files include:
-
- - Explicit static linking support via --static. Currently, e.g.
- 'xml2-config --libs' prints all libraries needed for static
- linking when doing dynamic linking unnecessary, resulting in
- unnecessary direct deps.
-
- - Better cross-build support. You don't have to build the additional
- *-config tools for target.
-
- - Better multilib support. Per-ABI pkgconfig directories are
- commonly supported while packages usually fail to look for
- per-CHOST *-config variants.
-
- - Better override support. The current macros allow little to no
- result overrides, the pkg-config macros let you pass FOO_CFLAGS
- and FOO_LIBS manually.
-
- - Cleaner version checks. The code used in libxml.m4 is really
- creepy.
-
- Patch from Michał Górny
-
- GH-773
-
-* WinTLS: Rewrite writeData
-
- We re-wrote WinTLSSession::writeData. The major points are:
-
- * Buffer is now preallocated once handshake is finished.
- Previously, they are allocated each time when we send one TLS
- record.
-
- * Schannel uses header, body and trailer for each secBuffer. Now we
- send them off at once using WSASend which is windows counterpart
- of sendv. Previously, we do memmove if some of them are
- truncated.
-
- * We don't try to send application data in
- WinTLSSession::closeConnection, since semantically we need same
- application data used to create TLS record before. Using 0 length
- data to finish sending buffered data looks like a hack.
-
- GH-772
+ GH-813