+++ /dev/null
-2005-12-28 Jamey Sharp <jamey@minilop.net>
-
- * configure.ac:
- Replace obsolete AC_OUTPUT(...) call with AC_CONFIG_FILES(...)
- and the no-argument AC_OUTPUT macro.
-
-2005-12-24 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_xlib.c, src/xcbxlib.h, src/Makefile.am:
- New header and two functions specifically for Xlib's use, so
- we can quit installing xcbint.h.
-
- * src/xcb_out.c, src/xcbint.h:
- Now that Xlib uses entirely public API, force_sequence_wrap is
- purely internal to xcb_out.
-
-2005-12-20 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_util.c:
- Use AF_UNIX instead of PF_UNIX, because they're interchangeable
- on Linux (and 4.x BSD?) but on MacOS X the latter is not
- available. Thanks to Ian Osgood for testing.
-
-2005-12-12 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_util.c:
- Rewrite XCBParseDisplay to handle all error cases correctly.
- Thanks to Travis Spencer for pointing out a problem with
- the previous sscanf call; turns out that sscanf won't do what
- I want anyway, so this version just uses strtoul.
-
-2005-12-12 Jamey Sharp <jamey@minilop.net>
-
- * tests/check_public.c:
- Fix segfault in fail_unless calls and provide more useful error
- messages. Also remove DISPLAY from the environment and test
- ParseDisplay on a null pointer.
-
-2005-12-07 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h, src/xcb_in.c:
- Use GCC attribute extension, if available, to mark some
- functions deprecated; and rename XCBWaitEvent to
- XCBWaitForEvent, with a deprecated backwards compatibility
- function.
-
-2005-12-07 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h, src/xcb_conn.c, src/xcb_util.c:
- Rename XCBConnect to XCBConnectToFD, and implement XCBConnect
- and XCBConnectToDisplayWithAuthInfo, as specified by the XcbApi
- documentation. Provide a trivial implementation of deprecated
- XCBConnectBasic for backwards compatibility. Fix XCBParseDisplay
- to accept a null screen pointer.
-
-2005-09-30 Trevor Woerner <twoerner.x@gmail.com>
-
- reviewer: Jamey Sharp <jamey@minilop.net>
-
- * Makefile.am, configure.ac, src/Makefile.am, tests/Makefile.am,
- src/check_all.c, src/check_public.c, src/check_suites.h:
- Created a new directory called "tests" where the unit tests
- will reside, separate from other files. Moved the unit
- test files from the "src" directory into the new "tests"
- directory.
-
- * configure.ac, tests/CheckLog.xsl, tests/.cvsignore,
- tests/Makefile.am, tests/check_all.c:
- Added an XSLT transform to post-process the XML log file which
- is generated by the unit test suite into an HTML page. Added
- the necessary hooks into the build system to have this
- translation occur when necessary as part of the build process.
-
-2005-08-15 Trevor Woerner <twoerner.x@gmail.com>
-
- reviewer: Jamey Sharp <jamey@minilop.net>
-
- * src/Makefile.am:
- Make builds work when the source and build directories
- are different.
-
-2005-08-03 Trevor Woerner <twoerner.x@gmail.com>
-
- reviewer: Jamey Sharp <jamey@minilop.net>
-
- * configure.ac:
- Allow people to not have check as long as they have at
- least automake version 1.9. Older versions of automake
- still work as long as check is installed or the NoCheck
- instructions are followed.
-
-2005-07-21 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_conn.c:
- Fix one uninitialized memory warning from Valgrind.
- More will have to wait.
-
-2005-07-21 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_ext.c:
- Fix memory leak reported by Carl Worth (thanks Carl!).
-
-2005-05-22 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Permit unions to have fixed-length list fields.
-
-2005-05-21 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- In function parameters and structure fields, line up field names in
- the same starting column.
-
-2005-05-17 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Make header files self-sufficient, by including xcb.h in the
- generated headers rather than the generated sources.
-
- * src/xcbext.h:
- Make xcbext.h self-sufficient, by including xcb.h.
-
-2005-05-17 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Line up function parameters with opening parenthesis.
-
-2005-05-17 Jamey Sharp <jamey@minilop.net>
-
- * src/c-client.xsl:
- Improve whitespace output on function declarations.
-
-2005-04-23 Jeremy Kolb <jkolb@brandeis.edu>
- * src/c-client.xsl:
- Added list support for typedefs.
- GLX now compiles.
-
-2005-04-23 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Allow protocol descriptions to explicitly qualify types with a
- namespace, which should be the value of the header attribute from
- the protocol description containing the desired type.
-
-2005-04-10 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h, src/xcb_in.c:
- New public function XCBGetRequestRead: returns the last sequence
- number that the server is known to have processed. This function
- enables apps to determine whether forcing a cookie is going to
- block.
-
-2005-04-09 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_ext.c:
- Implement XCBPrefetchExtensionData. An extension can now be in
- one of three states in a particular connection's cache: not
- queried; query cookie available; and query reply available.
-
-2005-04-09 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_ext.c, src/xcbint.h:
- Assign a display-independent unique integer to each XCBExtension
- structure. Use a map instead of a list for the extension cache.
- Key the map on the assigned integer.
-
-2005-04-06 Jeremy Kolb <jkolb@brandeis.edu>
-
- * src/c-client.xsl:
- Added support for c types float and double.
- Needed for GLX extension.
-
-2005-04-02 Jamey Sharp <jamey@minilop.net>
-
- * src/c-client.xsl:
- Workaround for XSLT processors that can't understand attribute
- values of the form '{{ "{...}" }}'. (What? That looks completely
- clear to me. *grin*)
-
-2005-04-01 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_list.c:
- Make the _destroy operations no-op on null pointers.
-
-2005-03-31 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h, src/xcb_conn.c, src/xcb_out.c, src/xcbint.h:
- Lazily check for BIG-REQUESTS support. The first call to
- XCBGetMaximumRequestLength takes the two-round-trip penalty for
- discovering BIG-REQUESTS support, and caches the result for
- subsequent calls. Also XCBSendRequest is now careful to only ask
- for the true maximum request length if the length from the
- connection setup data isn't sufficient for the current request.
-
-2005-03-31 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c:
- Fix X error handling to, uh, not hang.
-
-2005-03-30 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c, src/xcbint.h:
- Accept zero or more replies to any request. Eliminates, for the
- moment, nearly all run-time error checking regarding the number
- of replies expected for a particular request. Also breaks the
- unexpected reply callback, which is OK because I was going to
- remove it anyway, and now I've done so.
-
-2005-03-30 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_list.c, src/xcbint.h:
- Add a map ADT implemented using the list ADT.
-
-2005-03-28 Vincent Torri <torri@iecn.u-nancy.fr>
-
- * doc/tutorial/index.html, doc/tutorial/xcb.css:
- Add a tutorial
-
-2005-03-27 Jeremy Kolb <jkolb@brandeis.edu>
-
- * src/Makefile.am:
- Added xf86dri (xcb-proto/src/extensions/xf86dri.h).
- Removed compilation of record as it's broken.
-
-2005-03-25 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c:
- Instead of storing a pending flag per reply record, look in the
- readers list to find out whether there's a reader waiting for
- the reply we've just been asked to wait for.
-
-2005-03-25 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_list.c, src/xcbint.h:
- Add _xcb_queue_is_empty.
-
-2005-03-25 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c, src/xcb_list.c, src/xcbint.h:
- Add a queue ADT. Currently implemented using the list ADT, but
- analysis will probably suggest a better data structure. The
- split is intended to help that analysis.
-
-2005-03-25 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c, src/xcb_list.c, src/xcbint.h:
- Remove _xcb_list_length. (Two callers in xcb_in didn't actually
- care about the returned value beyond "empty" or "non-empty",
- which they can get other ways.) _xcb_in_events_length became
- obsolete in the process, and I took _xcb_list_clear out of
- xcbint.h while I was at it because only _xcb_list_delete calls
- it these days.
-
-2005-03-25 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h, src/xcb_in.c:
- Remove XCBEventQueueLength and XCBEventQueueClear: they're not
- part of the new API, and they interfere with analysis of the
- _xcb_list implementation.
-
-2005-03-24 Josh Triplett <josh@freedesktop.org>
-
- * configure.ac:
- Change pdx.freedesktop.org to lists.freedesktop.org in package
- email address.
-
-2005-03-23 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_conn.c, src/xcb_in.c:
- avn@daemon.hole.ru pointed out two memory leaks in XCB, which
- this update fixes. I also introduced a third a couple of days
- ago, which this fixes as well.
-
-2005-03-23 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c:
- Quit storing an "error" flag: it's redundant with the contents
- of the packet that it was stored with.
-
-2005-03-22 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c, src/xcbint.h:
- My recent Xlib patch stressed XCB's pending reply queue in ways
- it has never been stressed before, with thousands of entries.
- Profiling revealed that wake_up_next_reader was taking most of
- the cycles of my test app with its linear search of the reply
- queue. So I gave it its own 'readers' queue, which can have at
- most one entry per thread, and which can be no more than the
- number of pending replies. In the current implementation the
- readers queue uses only constant-time operations, too.
-
-2005-03-20 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_in.c:
- Wait for the last known read sequence number to pass the request
- we're waiting for, instead of checking whether data has arrived
- for the current request. Causes null to be returned if somebody
- screwed up by expecting a reply when none can come; relied on by
- new Xlib/XCB bits that intentionally and routinely screw up this
- way because the API doesn't give them enough information.
-
- * src/xcb_list.c, src/xcbint.h:
- Add _xcb_list_peek_head so Xlib doesn't have to repeatedly
- remove and re-insert the item at the head of its new
- pending_requests queue. Profiling showed that was a significant
- performance hit for the above-mentioned patch and with this
- three-line function apps are actually usable again.
-
-2004-11-26 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- In do-request processing, use the existing num-parts variable
- rather than recalculating the same expression.
-
-2004-11-26 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- * Add a list of known core types, and don't search all the
- protocol descriptions for types in that list. This
- significantly shortens the processing time.
- * Search for all definitions of a non-core type, not just the
- first, in order to check for duplicate definitions.
- * Add an error message for non-core types with no definitions
- found.
-
-2004-11-10 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Don't create a separate local variable for exprfields, and change
- an explicit xsl:for-each to an xsl:apply-templates.
-
-2004-11-10 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Prefix variables in request functions with "xcb_" to avoid
- collisions with field names.
-
-2004-11-09 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- * src/xcb_out.c:
- * src/xcbext.h:
- More API changes: Put constant request-specific data into a static
- const structure in each request function.
-
-2004-11-09 Josh Triplett <josh@freedesktop.org>
-
- * src/c-client.xsl:
- Add support for an <indent> tag in a function body, which indents
- the <l>s it contains.
-
- * debian/.cvsignore:
- debian/copyright is a generated file, so add it to
- debian/.cvsignore
-
-2004-11-08 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_conn.c:
- Make the __thread-based assertion a compile-time
- option, defaulting to off. It breaks on many systems
- and isn't critical to XCB's functionality.
-
-2004-11-07 Josh Triplett <josh@freedesktop.org>
-
- * c-client.xsl:
- Update XSLT to deal with change from extension element to
- extension-name and extension-xname attributes on xcb element.
-
-2004-11-06 Josh Triplett <josh@freedesktop.org>
-
- * xcb.h:
- * xcb_auth.c:
- * xcb_conn.c:
- * xcb_ext.c:
- * xcb_in.c:
- * xcb_list.c:
- * xcb_out.c:
- * xcb_util.c:
- * xcb_xid.c:
- * xcbext.h:
- * xcbint.h:
- Include license in every file rather than referencing COPYING.
- Update copyright notices to 2004.
-
- * COPYING:
- Remove non-applicable licenses from COPYING file.
-
-2004-11-06 Josh Triplett <josh@freedesktop.org>
-
- * configure.ac:
- * src/.cvsignore:
- * src/Makefile.am:
- * src/xcb.h:
- Merged changes for XML-XCB project.
-
- * src/c-client.xsl:
- * src/xcb.h:
- Add XSLT binding generator and non-M4 version of xcb.h.
-
- * src/client-c.xcb:
- * src/xcb.h.m4:
- Remove M4 binding generator and M4 version of xcb.h.
-
- * debian/changelog:
- * debian/control:
- * debian/copyright.debian:
- * debian/rules:
- Updated Debian packaging.
-
-2004-09-30 Jamey Sharp <jamey@minilop.net>
-
- * src/xcbext.h:
- * src/xcb_in.c:
- * src/client-c.xcb:
- Renamed XCBWaitReply to XCBWaitForReply for new API.
-
-2004-09-30 Jamey Sharp <jamey@minilop.net>
-
- * src/Makefile.am:
- Fix build failure when compiling tests.
-
-2004-09-29 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h.m4:
- * src/xcb_in.c:
- Added XCBPollForEvent function from new API.
-
-2004-09-29 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h.m4:
- * src/xcbext.h:
- * src/client-c.xcb:
- * src/xcb_ext.c:
- Update the extension cache for its new API. The
- optimizations that this API enables are not yet
- implemented, but the current implementation should
- at least function correctly.
-
-2004-09-29 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_util.c:
- * src/xcbext.h:
- Oops, XCBPopcount was supposed to change argument
- types too.
-
-2004-09-29 Jamey Sharp <jamey@minilop.net>
-
- * src/client-c.xcb:
- * src/xcb_util.c:
- * src/xcbext.h:
- Renamed XCBOnes to XCBPopcount as per new API spec.
-
-2004-09-29 Jamey Sharp <jamey@minilop.net>
-
- * src/xcbext.h:
- * src/xcb.h.m4:
- * src/Makefile.am:
- * src/client-c.xcb:
- * src/xcb_in.c:
- * src/xcb_out.c:
- * src/xcb_util.c:
- * src/xcb_xid.c:
- Moved declarations that are only useful to protocol
- implementations into a separate header file,
- xcbext.h.
-
-2004-09-28 Jamey Sharp <jamey@minilop.net>
-
- * src/xcbint.h:
- * src/xcb_conn.c:
- Add internal sanity check macros for sequence
- numbers, and invoke them every time _xcb_conn_wait
- is entered.
-
-2004-09-28 Jamey Sharp <jamey@minilop.net>
-
- * src/client-c.xcb:
- Allow LISTPARAMs to omit the length argument. When
- generating C code, a LOCALPARAM is automatically
- generated for the length if the length argument is
- omitted.
-
-2004-08-13 Jamey Sharp <jamey@minilop.net>
-
- * configure.ac:
- * src/Makefile.am:
- * src/.cvsignore:
- * src/check_all.c:
- * src/check_suites.h:
- * src/check_public.c:
- Add the first unit tests for XCB. This adds a build
- dependency on 'check'; I hope somebody will make it
- more of a build recommendation.
-
-2004-08-13 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_conn.c:
- Add a sanity check: assert() that _xcb_conn_wait
- will not be re-entered within the same thread. This
- turns out not to be true in some current buggy code.
- Xlib re-enters from _XAsyncReplyHandler in
- src/xcl/display.c; and the X Test Suite seems to
- catch SIGABRT and continue running.
-
-2004-08-10 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_out.c:
- I misread the BIG-REQUESTS specification. The long
- length field needs to count itself too.
-
-2004-07-24 Jamey Sharp <jamey@minilop.net>
-
- * src/client-c.xcb:
- Renamed functions that return new iterators to
- *Iter, to avoid conflict with similar functions that
- return arrays.
-
- Deprecating ARRAYFIELD in favor of LISTFIELD, which
- now generates both array and iterator accessors. You
- get an array accessor as long as the element type is
- fixed length; you get an interator accessor as long
- as the base type has an iterator. You always get a
- 'Length' function.
-
-2004-07-24 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_auth.c:
- Bart cleaned up some code that I made ugly, and
- I'm fixing a warning that resulted from a static
- function being created but not necessarily used.
-
-2004-07-07 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_xid.c:
- OK, it might be a good idea to commit something
- that compiles. I thought I'd tested this change...
-
-2004-07-06 Jamey Sharp <jamey@minilop.net>
-
- * src/xcbint.h:
- * src/xcb_xid.c:
- Added support for the XC-Misc extension's ability
- to request more XIDs from the server when the
- client runs out of them.
-
-2004-07-06 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h.m4:
- * src/xcb_auth.c:
- * src/xcb_util.c:
- Removed XCBNextNonce, and the nonce parameter of
- XCBGetAuthInfo, from the API. XCB can compute
- that just fine without help from the application.
-
-2004-07-06 Jamey Sharp <jamey@minilop.net>
-
- * src/Makefile.am:
- Importing part of patches, adding SYNC, by
- Mikko Markus Torni <mikko.markus.torni@iki.fi>.
- This needed better padding computation in XCB
- before it could be implemented cleanly.
-
-2004-07-02 Jamey Sharp <jamey@minilop.net>
-
- * src/Makefile.am:
- Importing part of patches, adding XC-MISC, by
- Mikko Markus Torni <mikko.markus.torni@iki.fi>.
-
-2004-07-02 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb.h.m4:
- * src/xcb_in.c:
- Remove XCBEventQueueRemove and XCBEventQueueFind.
- Their existence in the API constrains the
- implementation, and they're not useful for real
- code, like toolkits, anyway.
-
-2004-07-02 Jamey Sharp <jamey@minilop.net>
-
- * src/client-c.xcb:
- * src/xcb.h.m4:
- XCB headers are now compatible with C++ source.
-
-2004-07-02 Jamey Sharp <jamey@minilop.net>
-
- * src/client-c.xcb:
- * src/xcb.h.m4:
- Keith's fix for padding didn't quite work, despite
- looking quite sensible at the time. Replaced
- XCB_TYPE_CEIL with XCB_TYPE_PAD and made all the
- iterators, lists, and arrays count the number of
- bytes from the beginning of the current request.
-
-2004-07-02 Jamey Sharp <jamey@minilop.net>
-
- * src/xcb_out.c:
- * src/xcbint.h:
- Force a response from the server if we haven't seen
- one in 65530 requests. If you don't do this, XCB
- gets confused the next time it waits for a reply.
-
-2004-06-16 Keith Packard <keithp@keithp.com>
-
- * src/client-c.xcb:
- * src/xcb.h.m4:
- Must round datatype lengths to 4-byte boundary for
- NEXTFIELD and End macros. Created new XCB_TYPE_CEIL
- macro for this purpose and used it in a couple of
- places. It is to be hoped that this represents all
- of the locations that need changing...
projects / free-sw / xcb / libxcb / commitdiff