aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
...
* remove HAVE_STDARG_HJulian Orth2014-03-31
|
* remove HAVE_STDINTJulian Orth2014-03-31
|
* remove HAVE_INTTYPESJulian Orth2014-03-31
|
* remove HAVE_MATH_HJulian Orth2014-03-31
|
* remove HAVE_MEMSEToni-link2014-03-31
|
* remove HAVE_MEMCMPoni-link2014-03-31
|
* remove HAVE_BCMPJulian Orth2014-03-31
|
* remove HAVE_ERRNOJulian Orth2014-03-31
|
* remove HAVE_SETJMP_HJulian Orth2014-03-31
|
* Remove feature HAVE_QSORToni-link2014-03-31
| | | | qsort conforms to C99, so we don't need our own version.
* alloc: xmalloc and xrealloc can't return nullNicolas Hillegeer2014-03-31
| | | | | With this annotation/function attribute, we can inform gcc that these functions never return null, which can save a lot of redundant null checks.
* func-attr: add some extra handy function attrsNicolas Hillegeer2014-03-31
| | | | | | | | | | | The attributes in question are: - nonnull: specify whether a function argument cannot/may not be null - returns_nonnull: specify whether a function will not return a null pointer (example: xmalloc can't return null, so it should be annotated as such). Only available from gcc 4.9 onwards. Currently these attributes are only supported by gcc.
* func-attr: clang 3.5 doesn't support alloc_sizeNicolas Hillegeer2014-03-31
| | | | | | | The attribute was removed in commit c047507 in the clang repository as it was never properly implemented anyway. This fixes compiling with clang 3.5. Fixes issue #429
* Rename ga_init2() to ga_init()John Schmidt2014-03-31
|
* Inline ga_init()John Schmidt2014-03-31
|
* Fix shell.{c,h} comments to match the style guideThiago de Arruda2014-03-31
|
* Add doxygen comments to the functions in shell.cThiago de Arruda2014-03-31
|
* Rewrite `show_shell_mess` as a `ShellOpts` flagThiago de Arruda2014-03-31
|
* Update `call_shell` signature to use `ShellOpts`Thiago de Arruda2014-03-31
|
* Refactor SHELL_* defines into enum typedefThiago de Arruda2014-03-31
| | | | | | The SHELL_* defines are the bitflags that can be passed to `mch_call_shell`. The enum is defined in 'os/shell.h', where all shell-related functions will eventually be defined.
* Refactor/cleanup argument parsing functionsThiago de Arruda2014-03-31
|
* Extract `shell_build_argv` from `mch_call_shell`Thiago de Arruda2014-03-31
|
* Make `extra_shell_arg` a `mch_call_shell` parameterThiago de Arruda2014-03-31
|
* Extract `shell_count_argc` from `mch_call_shell`Thiago de Arruda2014-03-31
|
* Extract `shell_skip_word` from `mch_call_shell`Thiago de Arruda2014-03-31
|
* Split code for counting and parsing argumentsThiago de Arruda2014-03-31
|
* Add doxygen-style comments to functions from os/fs.c.Thomas Wienecke2014-03-31
|
* Add basic Doxyfile.Thomas Wienecke2014-03-31
|
* xrealloc(): similar to xmalloc()Felipe Oliveira Carvalho2014-03-31
| | | | | Replaced all calls to realloc by xrealloc. All `== NULL` tests can be removed and the code within `!= NULL` tests can be unwrapped.
* Document xmalloc() and deprecate lalloc()Felipe Oliveira Carvalho2014-03-31
|
* Avoid a seg fault when exiting after OOM errorFelipe Oliveira Carvalho2014-03-31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I'm debugging OOM behavior using http://www.nongnu.org/failmalloc/ on Linux. gdb environment: ``` set env LD_PRELOAD=libfailmalloc.so set env FAILMALLOC_SPACE=0xF00000 ``` SEGV was happening like this: ``` Starting program: /home/felipe/code/neovim/build/bin/nvim [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Vim: Error: Out of memory. Program received signal SIGSEGV, Segmentation fault. 0x00000000004d3719 in getout (exitval=1) at /home/felipe/code/neovim/src/main.c:836 836 if (*p_viminfo != NUL) (gdb) ``` After the fix it works as expected: ``` (gdb) set environment LD_PRELOAD=libfailmalloc.so (gdb) set environment FAILMALLOC_SPACE=0xF00000 (gdb) r Starting program: /home/felipe/code/neovim/build/bin/nvim [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Vim: Error: Out of memory. Vim: Finished. [Inferior 1 (process 21505) exited with code 01] (gdb) ```
* xmalloc() that succeeds or gracefully abortsFelipe Oliveira Carvalho2014-03-31
|
* update readme: "progress" section; formattingJustin M. Keyes2014-03-31
|
* Force the `cmake` target to re-run the CMake configuration.John Szakmeister2014-03-31
| | | | | This makes it more handy when files are added or removed from the source tree. A simple `make cmake` will re-configure and pick up the change.
* Use the clang 3.4 release tarball instead of apt for clang 3.4.John Szakmeister2014-03-30
| | | | | | | | | | It appears the llvm.org/apt/ repository isn't always reliable. So let's use the release tarball instead. Also, make using 3.4 conditional, so we can use the clang 3.3 if things still manage to go awry in the future. Note: using 3.3 means that we won't get leak detection. I left the logic for using llvm.org/apt/, just in case we want try using it again sometime.
* Re-integrate FEAT_SIGNS code, close #383Justin M. Keyes2014-03-29
| | | | | - omit FEAT_NETBEANS_INTG and FEAT_SIGN_ICONS - omit FEAT_GUI blocks
* use function attributes on allocation functionsNicolas Hillegeer2014-03-29
| | | | | This should allow gcc/clang/icc to generate better code and clarifies function intent.
* add func_attr.h to be able to use func attribsNicolas Hillegeer2014-03-29
| | | | | Currently enables these for every gnu-like compilers, i.e.: gcc, clang and icc.
* Move remove_duplicates to garray.cJohn Schmidt2014-03-29
|
* Extract path.c from misc1.cJohn Schmidt2014-03-29
|
* Drop the deb-src for clang.John Szakmeister2014-03-29
| | | | | The llvm.org/apt repo currently has an issue with the source packages, and we don't need it on Travis anyways.
* Append to CMAKE_PREFIX_PATH instead setting it outright.John Szakmeister2014-03-28
| | | | | Also, don't set it in the FindXxx cmake scripts--it's unnecessary, and it resets the value.
* Address clint warnings and other style issues.Thomas Wienecke2014-03-27
| | | | | | | * Add const. * Fix conditions (move && from end to start of line). * Use int32_t instead of long. * Use //-style comments.
* Rename mch_* functions to os_* in os module.Thomas Wienecke2014-03-27
|
* Fix prepare watcher teardown in the event loopThiago de Arruda2014-03-27
|
* Re-integrate FEAT_FILTERPIPE codeThiago de Arruda2014-03-27
| | | | | | | | | | | | | This feature was accidentally removed when doing the initial import from vim. It makes vim use pipes instead of temporary files for filtering buffers through shell commands. I found that this was missing when looking for references of SHELL_READ/SHELL_WRITE outside mch_call_shell`. When `mch_call_shell` is reimplemented on top of libuv process management facilities, pipes will always be used for communication with child processes so it makes sense to enable the feature permanently.
* Make `EventType` enum private to input.cThiago de Arruda2014-03-27
| | | | | This enum doesn't need to be public since `event_poll` is only interested in user input(but other events may be handled by libuv callbacks).
* Don't use alias in travis.sh.John Szakmeister2014-03-27
| | | | | We don't want to clobber anyone's make alias when running travis.sh directly.
* More cleanupJulian Orth2014-03-27
|
* remove leftover c++ headersJulian Orth2014-03-27
|
generated by cgit (git 2.34.1) at 2025-06-29 00:13:44 +0000