From 8fbff9dc2e8f3f444437c82835f22ba5d6a73367 Mon Sep 17 00:00:00 2001
From: Rob Landley <rob@landley.net>
Date: Thu, 8 Feb 2024 05:43:52 -0600
Subject: [PATCH] Add wc -L and tests.

---
 tests/wc.test   | 18 ++++++++++++++++++
 toys/posix/wc.c | 32 +++++++++++++++++++++-----------
 2 files changed, 39 insertions(+), 11 deletions(-)

diff --git a/tests/wc.test b/tests/wc.test
index 66cf48be..c751cce8 100755
--- a/tests/wc.test
+++ b/tests/wc.test
@@ -21,6 +21,7 @@ testing "standard input -cl" "wc -cl" "      1       5\n" "" "a b\nc"
 testing "-c" "wc -c file1" "26 file1\n" "" ""
 testing "-l" "wc -l file1" "4 file1\n" "" ""
 testing "-w" "wc -w file1" "5 file1\n" "" ""
+testing "-L" "wc -L file1" "18 file1\n" "" ""
 NOSPACE=1 testing "one file" "wc file1" "4 5 26 file1\n" "" ""
 testing "multiple files" "wc input - file1" \
         "      1       2       3 input\n      0       2       3 -\n      4       5      26 file1\n      5       9      32 total\n" "a\nb" "a b"
@@ -33,3 +34,20 @@ testing "-m 2" 'cat "$FILES/utf8/test2.txt" | wc -m' "169\n" "" ""
 echo -n " " > file1
 NOSPACE=1 testing "-mlw" "wc -mlw input" "1 2 11 input\n" "hello, 世界!\n" ""
 rm file1
+
+# Corner cases for "line length". Not QUITE fold because \b doesn't subtract
+
+testcmd '-L 2' '-L' '1\n' '' 'a\n'
+testcmd '-L 3' '-L' '1\n' '' 'a'
+testcmd '-L 4' '-L' '6\n' '' 'first\rsecond\n'
+testcmd '-L 5' '-L' '9\n' '' '\ta\n'
+testcmd '-L 6' '-L' '8\n' '' '\t\b\b\b\n'
+testcmd '-L 7' '-L' '9\n' '' 'abc\td\n'
+testcmd '-L 8' '-L' '4\n' '' 'abc\bd\n'
+testcmd '-L 9' '-L' '0\n' '' '\x01\n'
+testcmd '-L 10' '-L' '1\n' '' 'w\xcc\x88\n'
+NOSPACE=1 testcmd '-Lm incoherence' '-Lm' '2 0\n' '' '\x01\n'
+
+# 25 wide chars, also testing field order
+NOSPACE=1 testcmd '-cmlLw wide' '-cmlLw' '0 1 25 75 50\n' '' \
+  "$(<"$FILES"/utf8/japan.txt)"
diff --git a/toys/posix/wc.c b/toys/posix/wc.c
index 0fd9b550..1762d879 100644
--- a/toys/posix/wc.c
+++ b/toys/posix/wc.c
@@ -4,16 +4,17 @@
  *
  * See http://opengroup.org/onlinepubs/9699919799/utilities/wc.html
 
-USE_WC(NEWTOY(wc, "mcwl", TOYFLAG_USR|TOYFLAG_BIN))
+USE_WC(NEWTOY(wc, "Lcmwl", TOYFLAG_USR|TOYFLAG_BIN))
 
 config WC
   bool "wc"
   default y
   help
-    usage: wc [-lwcm] [FILE...]
+    usage: wc [-Llwcm] [FILE...]
 
     Count lines, words, and characters in input.
 
+    -L	Show max line length
     -l	Show lines
     -w	Show words
     -c	Show bytes
@@ -28,7 +29,7 @@ config WC
 #include "toys.h"
 
 GLOBALS(
-  unsigned long totals[4];
+  unsigned long totals[5];
 )
 
 static void show_lengths(unsigned long *lengths, char *name)
@@ -45,12 +46,13 @@ static void show_lengths(unsigned long *lengths, char *name)
   if (!(!toys.optc && !(toys.optflags & (toys.optflags-1))) && toys.optc!=1)
     space = 7;
 
-  for (i = 0; i<4; i++) {
+  for (i = 0; i<ARRAY_LEN(TT.totals); i++) {
     if (toys.optflags&(1<<i)) {
       printf(" %*ld"+first, space, lengths[i]);
       first = 0;
     }
-    TT.totals[i] += lengths[i];
+    if (i==4) TT.totals[i] = maxof(TT.totals[i], lengths[i]);
+    else TT.totals[i] += lengths[i];
   }
   if (*toys.optargs) printf(" %s", name);
   xputc('\n');
@@ -59,15 +61,15 @@ static void show_lengths(unsigned long *lengths, char *name)
 static void do_wc(int fd, char *name)
 {
   int len = 0, clen = 1, space = 0;
-  unsigned long word = 0, lengths[] = {0,0,0,0};
+  unsigned long word = 0, lengths[ARRAY_LEN(TT.totals)] = {0}, line = 0;
 
-  // Speed up common case: wc -c normalfile is file length.
+  // fast path: wc -c normalfile is file length.
   if (toys.optflags == FLAG_c) {
     struct stat st;
 
     // On Linux, files in /proc often report their size as 0.
     if (!fstat(fd, &st) && S_ISREG(st.st_mode) && st.st_size) {
-      lengths[2] = st.st_size;
+      lengths[3] = st.st_size;
       goto show;
     }
   }
@@ -82,8 +84,8 @@ static void do_wc(int fd, char *name)
 
     for (pos = 0; pos<len; pos++) {
       if (toybuf[pos]=='\n') lengths[0]++;
-      lengths[2]++;
-      if (FLAG(m)) {
+      lengths[3]++;
+      if (FLAG(m)||FLAG(L)) {
         // If we've consumed next wide char
         if (--clen<1) {
           // next wide size, don't count invalid, fetch more data if necessary
@@ -91,7 +93,14 @@ static void do_wc(int fd, char *name)
           if (clen == -1) continue;
           if (clen == -2 && !done) break;
 
-          lengths[3]++;
+          lengths[2]++;
+          line += maxof(wcwidth(wchar), 0);
+          if (wchar=='\t') line += 8-(line&7);
+          else if (wchar=='\n' || wchar=='\r') {
+            if (line>lengths[4]) lengths[4] = line;
+            line = 0;
+          }
+
           space = iswspace(wchar);
         }
       } else space = isspace(toybuf[pos]);
@@ -106,6 +115,7 @@ static void do_wc(int fd, char *name)
     if (pos != len) memmove(toybuf, toybuf+pos, len-pos);
     len -= pos;
   }
+  if (line>lengths[4]) lengths[4] = line;
 
 show:
   show_lengths(lengths, name);
-- 
2.39.2