| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | NETWARE: Only use the path context from the stat cache if the request | bnicholes | 2003-01-06 | 2 | -12/+28 |
| * | Added a new function, apr_filepath_encoding, to determine the character | brane | 2003-01-03 | 2 | -0/+22 |
| * | OS/2: Return APR_TIMEUP from apr_file_read() when a pipe times out. | bjh | 2003-01-02 | 1 | -1/+6 |
| * | OS/2: Fill in finfo->fname, was being left uninitialized. | bjh | 2003-01-02 | 1 | -0/+2 |
| * | Update copyright notices to 2003. | thommay | 2003-01-01 | 38 | -38/+38 |
| * | We can't optimize away this test! Just because a path didn't contain | wrowe | 2002-12-31 | 1 | -1/+1 |
| * | Fix uninitialized variable dereference from a recent commit | brianp | 2002-12-30 | 1 | -1/+1 |
| * | Close the last testdir emit by always calling FindFirstFile on open/rewind | wrowe | 2002-12-29 | 1 | -3/+44 |
| * | apr_dir_make_recursive is now implemented. The unix code is a little | wrowe | 2002-12-29 | 1 | -1/+45 |
| * | Get us building on unix again. | rbb | 2002-12-29 | 1 | -1/+1 |
| * | Defer allocation of the apr_file_t until the file is successfully opened | wrowe | 2002-12-29 | 1 | -33/+38 |
| * | Add an internal Win32 apr_file_open flag APR_OPENINFO to allow APR itself | wrowe | 2002-12-29 | 2 | -13/+28 |
| * | Grudgingly accept that folks may want to merge paths containing wildcard | wrowe | 2002-12-28 | 1 | -1/+8 |
| * | Introduce APR_EPATHWILD to designate that a given apr_file_[l]stat() | wrowe | 2002-12-28 | 1 | -1/+4 |
| * | Several StdHandles may be defined as the same handle, handle that case. | wrowe | 2002-12-28 | 2 | -10/+15 |
| * | Remove the segfault from testfile.c and introduce new breakage. | wrowe | 2002-12-28 | 1 | -0/+15 |
| * | OS/2: Fix file open flag manipulation so that APR_EXCL actually works. This | bjh | 2002-12-26 | 1 | -1/+1 |
| * | Use getcwdpath() rather than getcwd() in apr_filepath_get() to make sure that | bnicholes | 2002-12-20 | 1 | -3/+13 |
| * | Based on Philip Martin's research, we need to ignore any d_type results | wrowe | 2002-12-18 | 1 | -8/+20 |
| * | Reverted my last change for now. | brane | 2002-12-18 | 1 | -2/+2 |
| * | Reverse the tests for DIRENT_INODE and DIRENT_TYPE that were added to | brane | 2002-12-18 | 1 | -2/+2 |
| * | Brian, for your review. Start catching those scenarios which are UNKFILE | wrowe | 2002-12-17 | 1 | -0/+7 |
| * | Brad, please review. Adopt the alternate 'UNKFILE' if we failed | wrowe | 2002-12-17 | 1 | -6/+8 |
| * | Untested... but Ryan's assertion was correct, this should be static. | wrowe | 2002-12-15 | 1 | -2/+2 |
| * | Solaris is a bit strange, if there are no more entries in the | rbb | 2002-12-15 | 1 | -0/+11 |
| * | Fix a compile break on Solaris, and a warning on Linux. If the function | rbb | 2002-12-15 | 1 | -1/+2 |
| * | fix dependency on a nonexistent function, and remove an unused variable | brianp | 2002-12-15 | 1 | -4/+1 |
| * | Introduce detection of dirent->d_fileno (or ->d_ino) and dirent->d_type | wrowe | 2002-12-14 | 1 | -12/+52 |
| * | Use the new APR_UNKFILE when appropriate. | wrowe | 2002-12-14 | 1 | -1/+1 |
| * | Always use overlapped pipes on NT-based machines. This allows pipes to | rbb | 2002-12-12 | 1 | -43/+1 |
| * | Fix of my braindead error, submitted by Karl Fogel <kfogel@newton.ch.collab... | wrowe | 2002-12-12 | 1 | -3/+3 |
| * | Since Jeff indicates we trip on FIFO... here should be a legitmiate | wrowe | 2002-12-12 | 1 | -0/+16 |
| * | Fix rootpath detection for Win9X so we don't go stat'ing the volume. | wrowe | 2002-12-12 | 1 | -2/+2 |
| * | Comments for legibility. | wrowe | 2002-12-12 | 1 | -2/+4 |
| * | switch {case} and default: are probably better for handling this case. | wrowe | 2002-12-12 | 1 | -17/+21 |
| * | Dueling patches. Replace FirstBill's strcasecmp patch with a patch that | wrowe | 2002-12-12 | 1 | -66/+109 |
| * | Add a stub function for apr_dir_make_recursive. This doesn't do anything, | rbb | 2002-12-11 | 1 | -0/+8 |
| * | IIUC, users expect APR_EAGAIN() to respond TRUE if a lock has contention | wrowe | 2002-12-10 | 1 | -5/+0 |
| * | Allow for the path length to change while resolving each segment | bnicholes | 2002-11-26 | 1 | -2/+2 |
| * | Win32: Fix apr_stat() fooness. When passing in a filename like c:/foo/<a>, | stoddard | 2002-11-22 | 1 | -0/+16 |
| * | *) Renames done (deprecated functions wrapped): | thommay | 2002-11-20 | 1 | -2/+3 |
| * | Need to default to the standard open() call for NetWare when the returned | bnicholes | 2002-11-14 | 1 | -2/+4 |
| * | Store and use the path context to each directory and file rather than relying | bnicholes | 2002-11-06 | 3 | -147/+0 |
| * | Always lock the file on append, it's a good idea | wrowe | 2002-10-29 | 1 | -9/+5 |
| * | Comment a not so obvious tidbit. | stoddard | 2002-10-29 | 1 | -0/+5 |
| * | If opened append and overlapped, use apr_file_seek -first- to recompute | wrowe | 2002-10-29 | 1 | -4/+4 |
| * | Fix comments | stoddard | 2002-10-29 | 1 | -4/+2 |
| * | Win32: Get APR_APPEND file i/o working correctly. | stoddard | 2002-10-29 | 3 | -42/+56 |
| * | * apr/file_io/unix/copy.c (apr_file_transfer_contents): Fix comment. | kfogel | 2002-10-25 | 1 | -1/+1 |
| * | Getting ready for an API change in the NetWare LIBC library that will speed up | bnicholes | 2002-10-23 | 3 | -8/+206 |
