sed: add -z.

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

4 files changed, 21 insertions, 14 deletions

diff --git a/lib/lib.c b/lib/lib.c
index 2ba009f9..b074a92b 100644
--- a/lib/lib.c
+++ b/lib/lib.c
@@ -656,7 +656,7 @@ void loopfiles(char **argv, void (*function)(int fd, char *name))
 static void (*do_lines_bridge)(char **pline, long len);
 static void loopfile_lines_bridge(int fd, char *name)
 {
-  do_lines(fd, do_lines_bridge);
+  do_lines(fd, '\n', do_lines_bridge);
 }
 
 void loopfiles_lines(char **argv, void (*function)(char **pline, long len))
@@ -1356,7 +1356,7 @@ char *getgroupname(gid_t gid)
 // the line pointer if they want to keep it, or 1 to terminate processing,
 // otherwise line is freed. Passed file descriptor is closed at the end.
 // At EOF calls function(0, 0)
-void do_lines(int fd, void (*call)(char **pline, long len))
+void do_lines(int fd, char delim, void (*call)(char **pline, long len))
 {
   FILE *fp = fd ? xfdopen(fd, "r") : stdin;
 
@@ -1364,7 +1364,7 @@ void do_lines(int fd, void (*call)(char **pline, long len))
     char *line = 0;
     ssize_t len;
 
-    len = getline(&line, (void *)&len, fp);
+    len = getdelim(&line, (void *)&len, delim, fp);
     if (len > 0) {
       call(&line, len);
       if (line == (void *)1) break;
diff --git a/lib/lib.h b/lib/lib.h
index 3e6838ea..578a99c9 100644
--- a/lib/lib.h
+++ b/lib/lib.h
@@ -253,7 +253,7 @@ int regexec0(regex_t *preg, char *string, long len, int nmatch,
   regmatch_t pmatch[], int eflags);
 char *getusername(uid_t uid);
 char *getgroupname(gid_t gid);
-void do_lines(int fd, void (*call)(char **pline, long len));
+void do_lines(int fd, char delim, void (*call)(char **pline, long len));
 long environ_bytes();
 long long millitime(void);
 char *format_iso_time(char *buf, size_t len, struct timespec *ts);
diff --git a/tests/sed.test b/tests/sed.test
index a7d5a6eb..f2ff8fd7 100755
--- a/tests/sed.test
+++ b/tests/sed.test
@@ -165,4 +165,6 @@ testing "bonus backslashes" \
 testing "end b with }" "sed -n '/START/{:a;n;/END/q;p;ba}'" "b\nc\n" \
   "" "a\nSTART\nb\nc\nEND\nd"
 
+testing '-z' 'sed -z "s/\n/-/g"' "a-b-c" "" "a\nb\nc"
+
 # -i with $ last line test
diff --git a/toys/posix/sed.c b/toys/posix/sed.c
index bab6fb00..175e05d6 100644
--- a/toys/posix/sed.c
+++ b/toys/posix/sed.c
@@ -11,13 +11,13 @@
  *       What's the right thing to do for -i when write fails? Skip to next?
  * test '//q' with no previous regex, also repeat previous regex?
 
-USE_SED(NEWTOY(sed, "(help)(version)e*f*i:;nEr[+Er]", TOYFLAG_USR|TOYFLAG_BIN|TOYFLAG_LOCALE|TOYFLAG_NOHELP))
+USE_SED(NEWTOY(sed, "(help)(version)e*f*i:;nErz(null-data)[+Er]", TOYFLAG_USR|TOYFLAG_BIN|TOYFLAG_LOCALE|TOYFLAG_NOHELP))
 
 config SED
   bool "sed"
   default y
   help
-    usage: sed [-inrE] [-e SCRIPT]...|SCRIPT [-f SCRIPT_FILE]... [FILE...]
+    usage: sed [-inrzE] [-e SCRIPT]...|SCRIPT [-f SCRIPT_FILE]... [FILE...]
 
     Stream editor. Apply one or more editing SCRIPTs to each line of input
     (from FILE or stdin) producing output (by default to stdout).
@@ -27,8 +27,9 @@ config SED
     -i	Edit each file in place (-iEXT keeps backup file with extension EXT)
     -n	No default output (use the p command to output matched lines)
     -r	Use extended regular expression syntax
-    -E	Alias for -r
+    -E	POSIX alias for -r
     -s	Treat input files separately (implied by -i)
+    -z	Use \0 rather than \n as the input line separator
 
     A SCRIPT is a series of one or more COMMANDs separated by newlines or
     semicolons. All -e SCRIPTs are concatenated together as if separated
@@ -159,11 +160,11 @@ config SED
 
       #  Comment, ignore rest of this line of SCRIPT
 
-    Deviations from posix: allow extended regular expressions with -r,
-    editing in place with -i, separate with -s, printf escapes in text, line
-    continuations, semicolons after all commands, 2-address anywhere an
-    address is allowed, "T" command, multiline continuations for [abc],
-    \; to end [abc] argument before end of line.
+    Deviations from POSIX: allow extended regular expressions with -r,
+    editing in place with -i, separate with -s, NUL-separated input with -z,
+    printf escapes in text, line continuations, semicolons after all commands,
+    2-address anywhere an address is allowed, "T" command, multiline
+    continuations for [abc], \; to end [abc] argument before end of line.
 */
 
 #define FOR_sed
@@ -181,6 +182,7 @@ GLOBALS(
   long nextlen, rememberlen, count;
   int fdout, noeol;
   unsigned xx;
+  char delim;
 )
 
 // Linked list of parsed sed commands. Offset fields indicate location where
@@ -639,7 +641,7 @@ static void do_sed_file(int fd, char *name)
     for (command = (void *)TT.pattern; command; command = command->next)
       command->hit = 0;
   }
-  do_lines(fd, sed_line);
+  do_lines(fd, TT.delim, sed_line);
   if (FLAG(i)) {
     if (TT.i && *TT.i) {
       char *s = xmprintf("%s%s", name, TT.i);
@@ -998,6 +1000,8 @@ void sed_main(void)
   struct arg_list *al;
   char **args = toys.optargs;
 
+  if (!FLAG(z)) TT.delim = '\n';
+
   // Lie to autoconf when it asks stupid questions, so configure regexes
   // that look for "GNU sed version %f" greater than some old buggy number
   // don't fail us for not matching their narrow expectations.
@@ -1022,7 +1026,8 @@ void sed_main(void)
 
   for (al = TT.e; al; al = al->next) parse_pattern(&al->arg, strlen(al->arg));
   parse_pattern(0, 0);
-  for (al = TT.f; al; al = al->next) do_lines(xopenro(al->arg), parse_pattern);
+  for (al = TT.f; al; al = al->next)
+    do_lines(xopenro(al->arg), TT.delim, parse_pattern);
   dlist_terminate(TT.pattern);
   if (TT.nextlen) error_exit("no }");