aboutsummaryrefslogtreecommitdiff
path: root/toys/posix/sed.c
AgeCommit message (Collapse)Author
2019-02-23Bumper typo cleanup.Elliott Hughes
Inspired by some of the small patches that have gone by recently. Limited to only things found in `generated/help.h`, plus a wider cleanup for the more common "milisecond" typo.
2019-02-13sed: fix endless loop in "b loop" test.Elliott Hughes
Very few places actually check for errors from emit, and I actually see the same endless loop from "sed (GNU sed) 4.4" on current Debian, so I'm not sure this isn't Broken As Designed, but an endless loop spewing "short write" (or saying nothing, in the case of GNU sed) really doesn't feel like useful behavior in face of EPIPE, which really isn't going to fix itself. Certainly not being able to run the sed tests to completion is pretty annoying --- which is why, unless we remove this test as invalid, we should probably also add a SKIP_HOST=1 to the "b loop" test. Note that even with this fix you'll see the error twice: sed: short write: Broken pipe sed: short write: Broken pipe Once from the first = command to fail, and then another from the !FLAG(n) flush of the pattern space.
2019-02-12sed: fix substitution of empty capturing groups.Elliott Hughes
The test for \N where N was larger than the number of capturing groups in the regular expression was incorrect, and firing for cases such as matching __(ARM_)?NR_([a-z]*) against __NR_read, where the first group is empty (because it failed to match) but the second group did match "read". Use regex_t's re_nsub for the error check, and treat rm_so == -1 as a signal to just copy nothing into the result. (Found trying to build minijail in AOSP.)
2019-01-31Sed assumed FLAG(r) was 1 but didn't document it above the NEWTOY line. (oops)Rob Landley
2019-01-25OpenEmbedded needs commands installed in specific places.Rob Landley
2019-01-12sed: add -z.Elliott Hughes
Used to construct SELinux policies in the AOSP build. I left loopfiles_lines with its hard-coded '\n' because although cut(1) also has a -z option, I can't find any case where it's used in any of the codebases searchable by me. (And fmt(1), the other user, doesn't even have the option.) YAGNI. Bug: http://b/122744241
2019-01-11Fix b ending with }Rob Landley
For a definition of "fix" that's even _more_ of a deviation from posix, but matches what debian does...
2018-12-15Add sed -iEXT to keep backup files.Rob Landley
2018-12-04Add FLAG(x) macro, expanding to (toys.optflags & FLAG_##x)Rob Landley
2018-12-04Clean up some --help formatting.Elliott Hughes
Be consistent about upper versus lower case. (Upper seems to have the majority, so I went with that, though I'm happy to provide the opposite patch as long as we're consistent!) Be consistent about using \t. (Though saving a few bytes seems like it might be better done in the code that generates help.h rather than directly in the source, since tabs make careful ASCII art layout hard enough that we regularly have things misaligned.) Remove trailing periods (most of which seem to have been added by me). Always use the US "human readable" rather than my British "human-readable", and be more consistent about declaring whether we're showing multiples of 1000 or 1024. Just say "verbose" rather than adding a useless "mode" or "output".
2018-08-25Convert more argument variables in GLOBALS() to new style.Rob Landley
2018-06-14Change do_lines() semantics to end with a callback(0,0) to indicate EOF,Rob Landley
adjusting existing users.
2017-05-26Be more consistent about periods in help text.Elliott Hughes
2017-05-08Now that xopen() no longer returns stdio, !fd doesn't need strcmp("-") to verifyRob Landley
2016-10-01Move --version handling next to --help handling, so it applies to all commands.Rob Landley
Say "toybox" before version string. Tweak sed to preserve lie-to-autoconf.
2016-09-06Teach sed s/// how to handle [:space:] type sequences.Rob Landley
Or more accurately, s@[[:space:]@]@replace@ which can't treat the @ in [] as a delimiter but has to know about nested [[]] to make that decision.
2016-09-05Replace loopfiles' failok with WARN_ONLY open flag.Rob Landley
2016-09-03Move genericish function to lib.Rob Landley
2016-08-04Make xopen() skip stdin/stdout/stderr, add xopen_stdio() if you want stdout,Rob Landley
add xopenro() that takes one argument and understands "-" means stdin, and switch over lots of users.
2016-07-26Move regexec0 into lib (regexec that takes length and matches after NUL).Rob Landley
2016-07-23Usischev Yury pointed out a use after free error.Rob Landley
2016-06-29Remove unused argument from do_lines()Rob Landley
2016-04-16Rename gratuitous references to Roger Zelazny's amber series, add comments.Rob Landley
2016-04-11sed -f - should read from stdin.Rob Landley
2016-03-24Debian bug 635570 did something horribly nonstandard (depending on a sideRob Landley
effect of "sed -e 'a\'" with no trailing line). But there's an actual user, and it's not hard to implement, and it's not hard to implement, and there isn't an obvious _other_ way to do it, so add the behavior and a test for it. Fix some bad/missing comments while I was there, and add a couple TODOs.
2016-03-10Fix segfault in sed -e 'c\'.Andy Chu
Found by afl-fuzz.
2016-02-17The perl build's attempt to escape spaces and such in LD_LIBRARY_PATH is _SAD_.Rob Landley
It uses a sed expression that assumes you can escape - to use it as a literal (you can't, it has to be first or last char of the range), and assumes you have to escape delimiters in sed [] context (you don't), and/or that non-printf escapes become the literal character (they don't, the backslash is preserved as a literal), meaning it winds up doing "s/[\-\]//" which is a length 1 range, which is officially undefined behavior according to posix, and regcomp errors out. But if we don't accept it (like other implementations do) the perl build breaks. So collapse [A-A] into just [A]. Testcae taken from perl 5.22.0 file Makefile.SH line 8. (While we're at it, remove an unused argument from a function.)
2016-02-11removed unread assignment in sed.clovelycuppatea
clang scan-build flags up this line as being unread, so clearing from code.
2016-01-28Bugfix I forgot to checkin, plus a wrapper function.Rob Landley
2015-11-05Fix sed bug where any ] right after [ was skipped, not just first one in range.Rob Landley
2015-10-29Fix sed bug reported by Isabella Parakiss, where sed -e "/x/c\" -e "y" addedRob Landley
an extra newline because the test for whether we have an existing string to append a newline to was checking if struct step had data appended to it, and the /x/ regex is data appended to it. Change test to check for null terminator at ->arg1 offset.
2015-03-28Fix sed bug David Halls hit trying to compile libiconv.Rob Landley
2015-03-06sed depends on -r to be rightmost flag to simplify the REG_EXTENDED test.Rob Landley
(If a flag is 1, you can multiply it by the value you want and get that value else zero without a branch or conditional assignment.) So move -E one to the left so FLAG_r is 1 again, and add a [+Er] suffix instead of testing FLAG_r | FLAG_E in the users.
2015-03-02On 64 bit, subtracting two pointers produces a long result. On 32 bit, it's ↵Rob Landley
an int. Even though long _is_ 32 bits on a 32 bit systems, gcc warns about it because reasons. Also, the warning being that "expects int, but type is wchar_t"... no, type is not wchar_t. Type is probably long. Specify the ACTUAL TYPE, not the random typedef alias for it. If the translated type _did_ match, there wouldn't be a warning! (This is why c89 promoted all arguments to int, precisely so this wasn't a problem.)
2015-02-28BSD sed uses -E instead of -r to enable ERE.Isabella Bosia
GNU sed silently supports -E as an alias for -r too.
2015-01-22Fix sed s//\[newline]/ line continuations.Rob Landley
The problem was that readline() was returning a newline at the end of each string, which wasn't getting stripped in the parser and thus \ wasn't at the end of a line for -f, it was escaping a literal newline, so the continuation logic didn't trigger. Remove some redundant null checks while we're at it, and don't bother terminating a string we don't return (yes we leak memory in an error path, but it's about to error_exit() anyway).
2015-01-13sed bugfix: N or n at end of script would save the terminating NULL as the ↵Rob Landley
resume position, so the script would restart from beginning.
2015-01-13sed s/// can have line continuations in the replacement part, with or ↵Rob Landley
without a \ escaping the newline.
2015-01-04Fix sed backslash parsing in square bracket pattern sections.Rob Landley
2014-12-22sed 'r' didn't work right.Rob Landley
2014-12-21Another sed bug. (The e2fsprogs build uses multiple line continuations on ↵Rob Landley
the same command.)
2014-12-21Promote sed to posix.Rob Landley
2013-02-22Move some unfinished commands to the "pending" directory.Rob Landley
2012-11-13Reindent to two spaces per level. Remove vi: directives that haven't worked ↵Rob Landley
right in years (ubuntu broke its' vim implementation). Remove trailing spaces. Add/remove blank lines. Re-wordwrap in places. Update documentation with new coding style. The actual code should be the same afterward, this is just cosmetic refactoring.
2012-10-08New build infrastructure to generate FLAG_ macros and TT alias, #define ↵Rob Landley
FOR_commandname before #including toys.h to trigger it. Rename DEFINE_GLOBALS() to just GLOBALS() (because I could never remember if it was DECLARE_GLOBALS). Convert existing commands to use new infrastructure, and replace optflag constants with FLAG_ macros where appropriate.
2012-08-25Regularize command headers, update links to standards documents.Rob Landley
2012-08-25Move commands into "posix", "lsb", and "other" menus/directories.Rob Landley