basic/string-util: add a convenience function to cescape mostly-ascii fields

It's not supposed to be the most efficient, but instead fast and simple to use.

I kept the logic in ellipsize_mem() to use unicode ellipsis even in non-unicode
locales. I'm not quite convinced things should be this way, especially that with
this patch it'd actually be simpler to always use "…" in unicode locale and "..."
otherwise, but Lennart wanted it this way for some reason.

diff --git a/src/basic/string-util.c b/src/basic/string-util.c
index 498e31122588689fb0d400d52f3781d572366e92..f241a33767cea049099290de8e6088700ad91be3 100644 (file)
--- a/src/basic/string-util.c
+++ b/src/basic/string-util.c
@@ -14,6 +14,7 @@
 #include <string.h>
 
 #include "alloc-util.h"
+//#include "escape.h"
 #include "gunicode.h"
 //#include "locale-util.h"
 #include "macro.h"
@@ -457,6 +458,20 @@ bool string_has_cc(const char *p, const char *ok) {
         return false;
 }
 
+static int write_ellipsis(char *buf, bool unicode) {
+        if (unicode || is_locale_utf8()) {
+                buf[0] = 0xe2; /* tri-dot ellipsis: … */
+                buf[1] = 0x80;
+                buf[2] = 0xa6;
+        } else {
+                buf[0] = '.';
+                buf[1] = '.';
+                buf[2] = '.';
+        }
+
+        return 3;
+}
+
 static char *ascii_ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigned percent) {
         size_t x, need_space;
         char *r;
@@ -505,17 +520,7 @@ static char *ascii_ellipsize_mem(const char *s, size_t old_length, size_t new_le
         assert(x <= new_length - need_space);
 
         memcpy(r, s, x);
-
-        if (is_locale_utf8()) {
-                r[x+0] = 0xe2; /* tri-dot ellipsis: … */
-                r[x+1] = 0x80;
-                r[x+2] = 0xa6;
-        } else {
-                r[x+0] = '.';
-                r[x+1] = '.';
-                r[x+2] = '.';
-        }
-
+        write_ellipsis(r + x, false);
         memcpy(r + x + 3,
                s + old_length - (new_length - x - need_space),
                new_length - x - need_space + 1);
@@ -600,23 +605,56 @@ char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigne
         */
 
         memcpy(e, s, len);
-        e[len + 0] = 0xe2; /* tri-dot ellipsis: … */
-        e[len + 1] = 0x80;
-        e[len + 2] = 0xa6;
-
+        write_ellipsis(e + len, true);
         memcpy(e + len + 3, j, len2 + 1);
 
         return e;
 }
 
 char *ellipsize(const char *s, size_t length, unsigned percent) {
-
         if (length == (size_t) -1)
                 return strdup(s);
 
         return ellipsize_mem(s, strlen(s), length, percent);
 }
 
+char *cellescape(char *buf, size_t len, const char *s) {
+        /* Escape and ellipsize s into buffer buf of size len. Only non-control ASCII
+         * characters are copied as they are, everything else is escaped. The result
+         * is different then if escaping and ellipsization was performed in two
+         * separate steps, because each sequence is either stored in full or skipped.
+         *
+         * This function should be used for logging about strings which expected to
+         * be plain ASCII in a safe way.
+         *
+         * An ellipsis will be used if s is too long. It was always placed at the
+         * very end.
+         */
+
+        size_t i;
+        const char *t = s;
+
+        assert(len > 4 + 4 + 1); /* two chars and the terminator */
+
+        for (i = 0; i < len - 9; t++) {
+                if (!*t)
+                        goto done;
+                i += cescape_char(*t, buf + i);
+        }
+
+        /* We have space for one more char and terminating nul at this point */
+        if (*t) {
+                if (*(t+1))
+                        i += write_ellipsis(buf + i, false);
+                else
+                        i += cescape_char(*t, buf + i);
+        }
+
+ done:
+        buf[i] = '\0';
+        return buf;
+}
+
 bool nulstr_contains(const char *nulstr, const char *needle) {
         const char *i;
 

diff --git a/src/basic/string-util.h b/src/basic/string-util.h
index 52a1d7f2bee377ae9ff715a4fb3255c8dc412a00..99b51556979d7ba9cbf2e0a3a646b9c745b7ecd8 100644 (file)
--- a/src/basic/string-util.h
+++ b/src/basic/string-util.h
@@ -165,6 +165,7 @@ bool string_has_cc(const char *p, const char *ok) _pure_;
 
 char *ellipsize_mem(const char *s, size_t old_length_bytes, size_t new_length_columns, unsigned percent);
 char *ellipsize(const char *s, size_t length, unsigned percent);
+char *cellescape(char *buf, size_t len, const char *s);
 
 bool nulstr_contains(const char *nulstr, const char *needle);
 

diff --git a/src/test/test-string-util.c b/src/test/test-string-util.c
index a7c863c8958f8626c88be92a8b863b423e87c103..8c7226e5f8f683a717539f604f290a455030c09c 100644 (file)
--- a/src/test/test-string-util.c
+++ b/src/test/test-string-util.c
@@ -6,6 +6,7 @@
 ***/
 
 #include "alloc-util.h"
+//#include "locale-util.h"
 #include "macro.h"
 #include "string-util.h"
 #include "strv.h"
@@ -79,6 +80,29 @@ static void test_ascii_strcasecmp_nn(void) {
 }
 #endif // 0
 
+static void test_cellescape(void) {
+        char buf[40];
+
+        assert_se(streq(cellescape(buf, 10, "1"), "1"));
+        assert_se(streq(cellescape(buf, 10, "12"), "12"));
+        assert_se(streq(cellescape(buf, 10, "123"), is_locale_utf8() ? "1…" : "1..."));
+
+        assert_se(streq(cellescape(buf, 10, "1\011"), "1\\t"));
+        assert_se(streq(cellescape(buf, 10, "1\020"), "1\\020"));
+        assert_se(streq(cellescape(buf, 10, "1\020x"), is_locale_utf8() ? "1…" : "1..."));
+
+        assert_se(streq(cellescape(buf, 40, "1\020"), "1\\020"));
+        assert_se(streq(cellescape(buf, 40, "1\020x"), "1\\020x"));
+
+        assert_se(streq(cellescape(buf, 40, "\a\b\f\n\r\t\v\\\"'"), "\\a\\b\\f\\n\\r\\t\\v\\\\\\\"\\'"));
+        assert_se(streq(cellescape(buf, 10, "\a\b\f\n\r\t\v\\\"'"), is_locale_utf8() ? "\\a…" : "\\a..."));
+        assert_se(streq(cellescape(buf, 11, "\a\b\f\n\r\t\v\\\"'"), is_locale_utf8() ? "\\a…" : "\\a..."));
+        assert_se(streq(cellescape(buf, 12, "\a\b\f\n\r\t\v\\\"'"), is_locale_utf8() ? "\\a\\b…" : "\\a\\b..."));
+
+        assert_se(streq(cellescape(buf, sizeof buf, "1\020"), "1\\020"));
+        assert_se(streq(cellescape(buf, sizeof buf, "1\020x"), "1\\020x"));
+}
+
 static void test_streq_ptr(void) {
         assert_se(streq_ptr(NULL, NULL));
         assert_se(!streq_ptr("abc", "cdef"));
@@ -432,6 +456,7 @@ int main(int argc, char *argv[]) {
         test_ascii_strcasecmp_n();
         test_ascii_strcasecmp_nn();
 #endif // 0
+        test_cellescape();
         test_streq_ptr();
         test_strstrip();
         test_strextend();