aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2019-12-21Tighten up unquote_file()Rob Landley
2019-12-21Very basic patch smoketestRob Landley
2019-12-20Making sh single build work broke other single builds. Fix them again.Rob Landley
2019-12-20patch: support [FILE [PATCH]] arguments.Elliott Hughes
POSIX only mentions -i/stdin, but GNU patch -- and Larry Wall's patch 1.3, found via https://en.wikipedia.org/wiki/Patch_(Unix) -- also support supplying the name of the file to patch and the name of the patch file as optional arguments. The AOSP build makes use of this syntax to patch snakeyaml to remove references to java.beans.* stuff.
2019-12-15Nothing in pending should default y.Rob Landley
2019-12-14Any substitute EVAL should supply its own --Rob Landley
2019-12-14Teach "make sh" to make the multiplexer and builtins, even though it's "single".Rob Landley
2019-12-14Bugfix: xgetline() wasn't returning NULL at EOF.Rob Landley
2019-12-14dirtree.c: avoid spurious EINVAL warnings.Elliott Hughes
An Android engineer complained that they were seeing this when not running as root: $ adb shell ls ls: ./postinstall: Invalid argument ls: ./init: Permission denied ls: ./data_mirror: Invalid argument ls: ./init.environ.rc: Invalid argument ls: ./metadata: Invalid argument acct adb_keys apex From strace, it was here: newfstatat(4, "adb_keys", 0x7fc67eca88, AT_SYMLINK_NOFOLLOW) = -1 EACCES (Permission denied) readlinkat(4, "adb_keys", 0x5e843c7720, 4095) = -1 EINVAL (Invalid argument) So stop looking at st.st_mode (and then deciding to do a readlinkat()) if we didn't actually successfully stat().
2019-12-13fix loopback device mount failwayling
When we "losetup" success need mount loop device. Found this issue on AndroidQ
2019-12-12Next round of toysh work, with associated lib plumbing.Rob Landley
2019-12-06Squash a warning.Rob Landley
2019-12-06du.test: skip on Mac.Elliott Hughes
None of the current tests are relevant on the Mac because small symlinks are inlined into inodes, as are empty directories, so everything's using zero blocks.
2019-12-06printf.test: work around bash 3.2 on the Mac.Elliott Hughes
2019-12-06stat: fix filesystem type on macOS.Elliott Hughes
(Also show unknown values on Linux in hex rather than just "unknown".)
2019-12-06main.c: fall back to user's locale.Elliott Hughes
We still prefer C.UTF-8 for predictability, but macOS doesn't have that.
2019-12-04Fix getopt optstringE5ten
Move longopts after their corresponding shortopts instead of before
2019-12-04ifconfig.test: deflake.Elliott Hughes
::2 can legitimately be part of an IPv6 address, causing the test to fail because the grep matches an unrelated part of the output. Be more explicit about what we're actually searching for (as the previous test already was) to avoid flakiness.
2019-11-24bc: fix a comparison bugGavin Howard
2019-11-23macos_miniconfig: add cp, mktemp, and mv.Elliott Hughes
All three have been in use for the Android build for a while now, and all the tests pass.
2019-11-23file.test: fix /dev/zero major/minor for Mac.Elliott Hughes
2019-11-23Fix iconv and tests on Mac.Elliott Hughes
The Mac iconv_open(3) doesn't follow Unicode TR#22 rules for charset alias matching that bionic and glibc do (and, strictly, POSIX doesn't say you have to even though it's obviously a good idea), so we have to say exactly "UTF-8" rather than "utf8". Additionally, the 2006-era bash 3.2 on current versions of macOS (because it was the last GPLv2 bash) seems to have bugs that cause it to mangle UTF-8 input, so we can't reliably echo a UTF-8 sequence into a file. Use \x in the tests to work around this.
2019-11-23id.test: fix tests on Mac.Elliott Hughes
BSDs call the root group "wheel" instead. Also remove the duplicated "id id" from the progress output.
2019-11-23Fix dev_minor()/dev_major()/dev_makedev() for Mac.Elliott Hughes
Major/minor device encoding is not portable. No two BSDs agree with each other, and Darwin is different again. Everyone does agree on having major()/minor()/makedev() macros, but they disagree whether they should be in <sys/types.h> (the BSDs including Darwin, and old versions of glibc) or <sys/sysmacros.h> (glibc >= 2.26 and bionic). This fixes `ls -l /dev/zero` and `stat /dev/zero` on Mac.
2019-11-23du: use FLAG().Elliott Hughes
2019-11-23Don't use an internal implementation detail of skipnot in a test file.Rob Landley
2019-11-23cat.test: fix tests for Mac.Elliott Hughes
/proc/self/exe doesn't exist, but $C already gives us the path to the binary. /dev/full doesn't have any equivalent afaik, so skip that test if /dev/full is missing.
2019-11-23Elliott pointed out a codepath that could use uninitialized data.Rob Landley
(If DIRTREE_SYMFOLLOW returns ENOENT, skipping the second fstatat() would also skip the else goto error)
2019-11-23Don't test for specific error message output, they vary by libc version.Rob Landley
Check that the filename we fed in is output to stderr.
2019-11-22Sigh. Third attempt to fix trailing slashes in singlemake PREFIX.Rob Landley
(So much jetlag.)
2019-11-22Add getopt(1).Elliott Hughes
Includes new tests.
2019-11-22env.test: be sure to call toybox env.Elliott Hughes
This was failing on macOS because it was calling the system env(1).
2019-11-22mac: build fallocate and stat by default.Elliott Hughes
fallocate is fairly new, but I'm not sure why we didn't have stat on the list already. Having it fixes the truncate tests.
2019-11-20Ahem: work when PREFIX hasn't got trailing / OR when PREFIX is blank.Rob Landley
2019-11-18Work when PREFIX hasn't got trailing /Rob Landley
2019-11-14xargs: distinguish the two "argument too long" error cases.Elliott Hughes
One is really "the command is too long for me to ever call it given other constraints", so leave "argument too long" for the case where it's actually an argument causing the issue.
2019-11-13Pedantic constant annotations to shut up older toolchains.Rob Landley
2019-11-13Old gcc versions are confused by __has_include(), so nest the #ifdefsRob Landley
and #define/#undef a second symbol for the else case.
2019-11-12Don't rm prompt for nonexistent, just warn.Denys Nykula
2019-11-12Replace a test I took out last commit because it seemed impossible to trigger,Rob Landley
but the problem is some vertical sort arrangements are impossible, and that's what it was testing for. For example, showing 29 entries in 9 columns with horizontal sort requires 4 rows: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 29 29 x x x x x x x But with vertical sort that would be: 1 5 9 13 17 21 25 29 x 2 6 10 14 18 22 26 x x 3 7 11 15 19 23 27 x x 4 8 12 16 20 24 28 x x It still doesn't fit in 3 rows (3x9=27) but with 4 rows the 7 leftover spaces eats a whole column, so you _can't_ have 9 columns with vertical sort.
2019-11-08ls cleanup: use FLAG() macros, simplify next_column(), fix off by one errorsRob Landley
in whitespace accounting, eliminate lastcol, same sort[next] in dt, don't count trailing whitespace on last entry in row.
2019-11-05find.test: add missing tests for "accidentally" fixed bug.Elliott Hughes
Bug #69 was fixed recently by commit 0b2cfcb8fdea9673f3c2e0940f1b16d5825e16ea. Add the two tests from the bug so we can close it out. Fixes #69.
2019-11-05id: fix Android issues.Elliott Hughes
glibc doesn't set errno when getpw* fails, so the perror_exit() looked fine. bionic sets ENOENT and the trailing "No such file or directory" looks silly, so switch to error_exit(). Additionally, the default format tests fail on Android because of SELinux (but for a different reason than usual!). There's no id --no-context flag, so use sed to just throw away any SELinux context.
2019-11-05id.c: fix -nG.Elliott Hughes
The previous patch broke -nG, so move the -G code back to showone() which handles -n. Add the missing tests for the various uses of -n. Also refactor the code to avoid the need to test optflags directly.
2019-11-05Switch -w tests to check boundary conditions, fix code to pass tests,Rob Landley
variable declarations go at the start of blocks, and remove specific people's names from todo items (anybody can do any todo).
2019-11-04id: various fixes.Elliott Hughes
Handle unknown groups (fixes #117). Fix -G to show *all* groups, not just all supplementary groups. Fix -Z output to not include "context=".
2019-11-03id: support numeric lookup.Elliott Hughes
Fall back to converting the "name" to an integer and calling getpwuid(). We need to update `username` for the later call to getgrouplist(). Also fix the separator printing logic to avoid a trailing ',' on `id 0`. Switch to FLAG() and move some declarations down to where they can be initialized, both for clarity. Also add simple tests. Sadly, there's no always-present user that is in multiple groups.
2019-11-03ip: remove get_line().Elliott Hughes
2019-10-31ls: Add tests for `-C` and `-x` optionsAndrew Ilijic
These tests ensure we follow the behavior of other `ls` commands, in the basic case.
2019-10-31ls: Remove trailing whitespace so that tests passAndrew Ilijic
When in modes `-C` and `-x` we need to remove the trailing whitespace on each line. This is the behavior of other `ls` commands. Other `ls` commands will print the last filename and then print a newline. Prior to this patch we would print the last filename, followed by two spaces, and then print a newline. Previously, we would get to the end of the loop and print the padding. I couldn't figure out a way to determine when the program had reached the end of a line. So I piggybacked off of the newline code.