util-lib: add truncation based on printable width to utf8_escape_non_printable()

3 files changed, 95 insertions, 14 deletions

diff --git a/src/basic/utf8.c b/src/basic/utf8.c
index 090c69d140..e12876962d 100644
--- a/src/basic/utf8.c
+++ b/src/basic/utf8.c
@@ -196,42 +196,88 @@ char *utf8_escape_invalid(const char *str) {
         return p;
 }
 
-char *utf8_escape_non_printable(const char *str) {
-        char *p, *s;
+static int utf8_char_console_width(const char *str) {
+        char32_t c;
+        int r;
+
+        r = utf8_encoded_to_unichar(str, &c);
+        if (r < 0)
+                return r;
+
+        /* TODO: we should detect combining characters */
+
+        return unichar_iswide(c) ? 2 : 1;
+}
+
+char *utf8_escape_non_printable_full(const char *str, size_t console_width) {
+        char *p, *s, *prev_s;
+        size_t n = 0; /* estimated print width */
 
         assert(str);
 
-        p = s = malloc(strlen(str) * 4 + 1);
+        if (console_width == 0)
+                return strdup("");
+
+        p = s = prev_s = malloc(strlen(str) * 4 + 1);
         if (!p)
                 return NULL;
 
-        while (*str) {
+        for (;;) {
                 int len;
+                char *saved_s = s;
+
+                if (!*str) /* done! */
+                        goto finish;
 
                 len = utf8_encoded_valid_unichar(str, (size_t) -1);
                 if (len > 0) {
                         if (utf8_is_printable(str, len)) {
+                                int w;
+
+                                w = utf8_char_console_width(str);
+                                assert(w >= 0);
+                                if (n + w > console_width)
+                                        goto truncation;
+
                                 s = mempcpy(s, str, len);
                                 str += len;
+                                n += w;
+
                         } else {
-                                while (len > 0) {
+                                for (; len > 0; len--) {
+                                        if (n + 4 > console_width)
+                                                goto truncation;
+
                                         *(s++) = '\\';
                                         *(s++) = 'x';
                                         *(s++) = hexchar((int) *str >> 4);
                                         *(s++) = hexchar((int) *str);
 
                                         str += 1;
-                                        len--;
+                                        n += 4;
                                 }
                         }
                 } else {
-                        s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER);
+                        if (n + 1 > console_width)
+                                goto truncation;
+
+                        s = mempcpy(s, UTF8_REPLACEMENT_CHARACTER, strlen(UTF8_REPLACEMENT_CHARACTER));
                         str += 1;
+                        n += 1;
                 }
+
+                prev_s = saved_s;
         }
 
-        *s = '\0';
+ truncation:
+        /* Try to go back one if we don't have enough space for the ellipsis */
+        if (n + 1 >= console_width)
+                s = prev_s;
+
+        s = mempcpy(s, "…", strlen("…"));
 
+ finish:
+        *s = '\0';
         return p;
 }
 
@@ -519,15 +565,15 @@ size_t utf8_console_width(const char *str) {
         /* Returns the approximate width a string will take on screen when printed on a character cell
          * terminal/console. */
 
-        while (*str != 0) {
-                char32_t c;
+        while (*str) {
+                int w;
 
-                if (utf8_encoded_to_unichar(str, &c) < 0)
+                w = utf8_char_console_width(str);
+                if (w < 0)
                         return (size_t) -1;
 
+                n += w;
                 str = utf8_next_char(str);
-
-                n += unichar_iswide(c) ? 2 : 1;
         }
 
         return n;
diff --git a/src/basic/utf8.h b/src/basic/utf8.h
index 6df70921db..62e99b7280 100644
--- a/src/basic/utf8.h
+++ b/src/basic/utf8.h
@@ -22,7 +22,10 @@ bool utf8_is_printable_newline(const char* str, size_t length, bool newline) _pu
 #define utf8_is_printable(str, length) utf8_is_printable_newline(str, length, true)
 
 char *utf8_escape_invalid(const char *s);
-char *utf8_escape_non_printable(const char *str);
+char *utf8_escape_non_printable_full(const char *str, size_t console_width);
+static inline char *utf8_escape_non_printable(const char *str) {
+        return utf8_escape_non_printable_full(str, (size_t) -1);
+}
 
 size_t utf8_encode_unichar(char *out_utf8, char32_t g);
 size_t utf16_encode_unichar(char16_t *out, char32_t c);
diff --git a/src/test/test-utf8.c b/src/test/test-utf8.c
index 343cbf4ced..b5c4e3dc34 100644
--- a/src/test/test-utf8.c
+++ b/src/test/test-utf8.c
@@ -111,6 +111,37 @@ static void test_utf8_escape_non_printable(void) {
         assert_se(utf8_is_valid(p6));
 }
 
+static void test_utf8_escape_non_printable_full(void) {
+        log_info("/* %s */", __func__);
+
+        for (size_t i = 0; i < 20; i++) {
+                _cleanup_free_ char *p;
+
+                p = utf8_escape_non_printable_full("goo goo goo", i);
+                puts(p);
+                assert_se(utf8_is_valid(p));
+                assert_se(utf8_console_width(p) <= i);
+        }
+
+        for (size_t i = 0; i < 20; i++) {
+                _cleanup_free_ char *p;
+
+                p = utf8_escape_non_printable_full("\001 \019\20\a", i);
+                puts(p);
+                assert_se(utf8_is_valid(p));
+                assert_se(utf8_console_width(p) <= i);
+        }
+
+        for (size_t i = 0; i < 20; i++) {
+                _cleanup_free_ char *p;
+
+                p = utf8_escape_non_printable_full("\xef\xbf\x30\x13", i);
+                puts(p);
+                assert_se(utf8_is_valid(p));
+                assert_se(utf8_console_width(p) <= i);
+        }
+}
+
 static void test_utf16_to_utf8(void) {
         const char16_t utf16[] = { htole16('a'), htole16(0xd800), htole16('b'), htole16(0xdc00), htole16('c'), htole16(0xd801), htole16(0xdc37) };
         static const char utf8[] = { 'a', 'b', 'c', 0xf0, 0x90, 0x90, 0xb7 };
@@ -189,6 +220,7 @@ int main(int argc, char *argv[]) {
         test_utf8_encoded_valid_unichar();
         test_utf8_escape_invalid();
         test_utf8_escape_non_printable();
+        test_utf8_escape_non_printable_full();
         test_utf16_to_utf8();
         test_utf8_n_codepoints();
         test_utf8_console_width();