Add -t, deviations from posix, redo loop to have fewer duplicate tests.

1 files changed, 46 insertions, 27 deletions

diff --git a/toys/posix/strings.c b/toys/posix/strings.c
index d2154e79..911fc452 100644
--- a/toys/posix/strings.c
+++ b/toys/posix/strings.c
@@ -3,23 +3,29 @@
  * Copyright 2014 Kyung-su Kim <kaspyx@gmail.com>
  * Copyright 2014 Kyungwan Han <asura321@gmail.com>
  *
- * No Standard
+ * See http://pubs.opengroup.org/onlinepubs/9699919799/utilities/strings.html
+ *
+ * Deviations from posix: we don't readahead to the end of the string to see
+ * if it ends with NUL or newline before printing. Add -o. We always do -a
+ * (and accept but don't document the flag), but that's sort of conformant.
+ * Posix' STDOUT section says things like "%o %s" and we support 64 bit offsets.
+ *
  * TODO: utf8 strings
- * TODO: posix -t
 
-USE_STRINGS(NEWTOY(strings, "an#=4<1fo", TOYFLAG_USR|TOYFLAG_BIN))
+USE_STRINGS(NEWTOY(strings, "t:an#=4<1fo", TOYFLAG_USR|TOYFLAG_BIN))
 
 config STRINGS
   bool "strings"
   default y
   help
-    usage: strings [-fo] [-n LEN] [FILE...]
+    usage: strings [-fo] [-t oxd] [-n LEN] [FILE...]
 
     Display printable strings in a binary file
 
-    -f	Precede strings with filenames
+    -f	Show filename
     -n	At least LEN characters form a string (default 4)
-    -o	Precede strings with decimal offsets
+    -o	Show offset (ala -t d)
+    -t	Show offset type (o=octal, d=decimal, x=hexadecimal)
 */
 
 #define FOR_strings
@@ -27,38 +33,51 @@ config STRINGS
 
 GLOBALS(
   long num;
+  char *t;
 )
 
 static void do_strings(int fd, char *filename)
 {
   int nread, i, wlen = TT.num, count = 0;
   off_t offset = 0;
-  char *string = xzalloc(wlen + 1);
+  char *string = 0, pattern[8];
+
+  if (TT.t) if (!(string = strchr("oxd", *TT.t))) error_exit("-t needs oxd");
+  sprintf(pattern, "%%7ll%c ", string ? *string : 'd');
 
-  for (;;) {
-    nread = read(fd, toybuf, sizeof(toybuf));
-    if (nread < 0) perror_msg_raw(filename);
-    if (nread < 1) {
-      if (count == wlen) xputc('\n');
-      break;
+  // input buffer can wrap before we have enough data to output, so
+  // copy start of string to temporary buffer until enough to output
+  string = xzalloc(wlen+1);
+
+  for (i = nread = 0; ;i++) {
+    if (i >= nread) {
+      nread = read(fd, toybuf, sizeof(toybuf));
+      i = 0;
+      if (nread < 0) perror_msg_raw(filename);
+      if (nread < 1) {
+        if (count) goto flush;
+        break;
+      }
     }
-    for (i = 0; i < nread; i++, offset++) {
-      if (((toybuf[i] >= 32) && (toybuf[i] <= 126)) || (toybuf[i] == '\t')) {
-        if (count == wlen) fputc(toybuf[i], stdout);
-        else {
-          string[count++] = toybuf[i];
-          if (count == wlen) {
-            if (toys.optflags & FLAG_f) printf("%s: ", filename);
-            if (toys.optflags & FLAG_o)
-              printf("%7lld ",(long long)(offset + 1 - wlen));
-            printf("%s", string);
-          }
+
+    offset++;
+    if ((toybuf[i]>=32 && toybuf[i]<=126) || toybuf[i]=='\t') {
+      if (count == wlen) fputc(toybuf[i], stdout);
+      else {
+        string[count++] = toybuf[i];
+        if (count == wlen) {
+          if (toys.optflags & FLAG_f) printf("%s: ", filename);
+          if (toys.optflags & (FLAG_o|FLAG_t))
+            printf(pattern, (long long)(offset - wlen));
+          printf("%s", string);
         }
-      } else {
-        if (count == wlen) xputc('\n');
-        count = 0;
       }
+      continue;
     }
+flush:
+    // End of previous string
+    if (count == wlen) xputc('\n');
+    count = 0;
   }
   xclose(fd);
   free(string);