<feed xmlns='http://www.w3.org/2005/Atom'>
<title>gpgme/src/w32-io.c, branch gpgme-1.7.1</title>
<subtitle>GnuPG Made Easy.</subtitle>
<id>http://git.bktus.com/gpgfrontend/gpgme/atom?h=gpgme-1.7.1</id>
<link rel='self' href='http://git.bktus.com/gpgfrontend/gpgme/atom?h=gpgme-1.7.1'/>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/'/>
<updated>2016-08-18T15:14:36Z</updated>
<entry>
<title>core: Remove (now) useless diagnostic</title>
<updated>2016-08-18T15:14:36Z</updated>
<author>
<name>Andre Heinecke</name>
<email>aheinecke@intevation.de</email>
</author>
<published>2016-08-18T15:14:36Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=30f156280f18500ee522db58aecd40711c8af685'/>
<id>urn:sha1:30f156280f18500ee522db58aecd40711c8af685</id>
<content type='text'>
* src/w32-io.c(_gpgme_io_spawn): Remove spawnhelper not found
diagnostic.

--
When spawnhelper is not found the error is now catched before
the CreateProcess call so the added diagnostic in bb2d11c
(which was not helpful because the value for spawnhelper
would be NULL in that case) is now no longer needed.
</content>
</entry>
<entry>
<title>core: Fail loudly in case w32 spawner not found</title>
<updated>2016-08-18T15:09:04Z</updated>
<author>
<name>Andre Heinecke</name>
<email>aheinecke@intevation.de</email>
</author>
<published>2016-08-18T14:56:58Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=9cf983b0199950c8f8cccee2cb8e45aafcba9fd1'/>
<id>urn:sha1:9cf983b0199950c8f8cccee2cb8e45aafcba9fd1</id>
<content type='text'>
* src/w32-io.c (_gpgme_io_spawn): Show a message box in
case gpgme-w32spawn.exe not found.

--
Otherwise every engine call will just fail with unsupported
protocol. Even in the debug output the problem was not made
clear because CreateProcess will fail with error code 87
(Invalid Parameter) because spawnhelper is NULL. The helpful
error message for ERROR_INVALID_PARAMETER would have been:
"is 'NULL' correctly installed"

As GpgME basically becomes useless on Windows without the
spawnhelper we want to fail very loud in that case.
</content>
</entry>
<entry>
<title>Fix typos found by codespell.</title>
<updated>2015-12-02T10:28:07Z</updated>
<author>
<name>Justus Winter</name>
<email>justus@g10code.com</email>
</author>
<published>2015-12-02T10:28:07Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=8c61cbfb8ff2e1056840ecf8511810ed2482eb1f'/>
<id>urn:sha1:8c61cbfb8ff2e1056840ecf8511810ed2482eb1f</id>
<content type='text'>
--
Signed-off-by: Justus Winter &lt;justus@g10code.com&gt;
</content>
</entry>
<entry>
<title>w32: Add extra diagnostic about possible missing gpgme-w32spawn.exe.</title>
<updated>2015-10-28T15:28:48Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2015-10-28T15:27:49Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=bb2d11c1eebd4bcfb0f2cfce728026a7420dca47'/>
<id>urn:sha1:bb2d11c1eebd4bcfb0f2cfce728026a7420dca47</id>
<content type='text'>
* src/w32-io.c (_gpgme_io_spawn): Add a new diagnostic.

Signed-off-by: Werner Koch &lt;wk@gnupg.org&gt;
</content>
</entry>
<entry>
<title>w32: Fix another memleak on error.</title>
<updated>2014-04-16T08:19:54Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2014-04-15T14:40:48Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=62711e5614e44e65a7c7bb7d21493d09d3081271'/>
<id>urn:sha1:62711e5614e44e65a7c7bb7d21493d09d3081271</id>
<content type='text'>
* src/w32-io.c (create_reader): free CTX.
--

Found by Hans-Christoph Steiner with cppcheck.
</content>
</entry>
<entry>
<title>w32: Fix memleak in an error code paths.</title>
<updated>2014-04-15T20:18:04Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2014-04-15T14:40:48Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=85a07ca7e3dffdefc8ae74beebb59e47a6e6bd1b'/>
<id>urn:sha1:85a07ca7e3dffdefc8ae74beebb59e47a6e6bd1b</id>
<content type='text'>
* src/w32-io.c (create_writer): Free CTX in cased of bad FD.
* src/w32-util.c (_gpgme_mkstemp): Free TMPNAME in case of a failed
mkstemp.
--

Found by Hans-Christoph Steiner with cppcheck.
</content>
</entry>
<entry>
<title>Make use of internal iospawn flags more flexible.</title>
<updated>2014-04-10T09:39:14Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2014-04-10T09:39:14Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=443f6b76a964b77e2e61c03592a83ffa0a6da4b1'/>
<id>urn:sha1:443f6b76a964b77e2e61c03592a83ffa0a6da4b1</id>
<content type='text'>
* src/priv-io.h (IOSPAWN_FLAG_DETACHED): New. Renumber the others.
* src/w32-io.c (_gpgme_io_spawn): Use DETACHED_PROCESS process only if
IOSPAWN_FLAG_DETACHED is given.
* src/w32-qt-io.cpp (_gpgme_io_spawn): Ditto.
* src/w32-glib-io.c (_gpgme_io_spawn): Ditto.
* src/assuan-support.c (my_spawn): Pass IOSPAWN_FLAG_DETACHED flags.
* src/dirinfo.c (read_gpgconf_dirs): Ditto.
* src/engine-gpg.c (start): Ditto.
* src/engine-gpgconf.c (gpgconf_read, gpgconf_write): Ditto.
* src/version.c (_gpgme_get_program_version): Ditto.
</content>
</entry>
<entry>
<title>w32: Change the way the I/O threads are cleaned up.</title>
<updated>2013-05-11T13:38:08Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2013-04-25T11:00:16Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=9e7df9aa6d81f0abbabb03a2346d80eb5d375f81'/>
<id>urn:sha1:9e7df9aa6d81f0abbabb03a2346d80eb5d375f81</id>
<content type='text'>
* src/w32-io.c (reader_context_s, create_reader)
(writer_context_s, create_writer): Rename STOPPED to CLOSE_EV.
(reader, writer): Remove setting of STOPPED.  Wait for CLOSE_EV and
then release the context.
(destroy_reader, destroy_writer): Do not wait but set the CLOSE_EV.
(kill_reader, kill_writer): Remove.
(_gpgme_io_close): Add code from kill_reader and kill_writer.
--

The old code was prone to deadlocks which were actually exhibited at
Kleopatra startup.  The new code is much more straightforward and
easier to understand.  The reason for the complex old code was
probably due to our former idea to allow re-use of the I/O threads.
However we have long given up on this.
</content>
</entry>
<entry>
<title>Fix hang in socket closing.</title>
<updated>2013-05-08T18:38:50Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2013-04-25T11:00:16Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=9f330be8210d2498fe93d4166b6f6c02fca76475'/>
<id>urn:sha1:9f330be8210d2498fe93d4166b6f6c02fca76475</id>
<content type='text'>
* src/w32-io.c (destroy_reader): Call shutdown.
(reader): Do not print an error in the shutdown case.
</content>
</entry>
<entry>
<title>Improve debug output of the I/O reader and writer.</title>
<updated>2013-05-08T17:10:41Z</updated>
<author>
<name>Werner Koch</name>
<email>wk@gnupg.org</email>
</author>
<published>2013-05-06T18:23:47Z</published>
<link rel='alternate' type='text/html' href='http://git.bktus.com/gpgfrontend/gpgme/commit/?id=2118f497010a9a41c29d062a7605ff2e136f8f4e'/>
<id>urn:sha1:2118f497010a9a41c29d062a7605ff2e136f8f4e</id>
<content type='text'>
* src/w32-io.c (reader, writer): Also print file_sock.
</content>
</entry>
</feed>
