From f9c3f741172374599a6bb4749d188eeae52a42c3 Mon Sep 17 00:00:00 2001
From: Enno Rehling <enno.rehling@gmail.com>
Date: Sun, 7 Jun 2020 12:16:23 +0200
Subject: [PATCH] Bug 2670 trim &nbsp; and others from strings.

---
 src/util/unicode.c      |  2 +-
 src/util/unicode.test.c | 37 +++++++++++++++++++++++++++++++++++--
 2 files changed, 36 insertions(+), 3 deletions(-)

diff --git a/src/util/unicode.c b/src/util/unicode.c
index e87c27fc3..c7fada910 100644
--- a/src/util/unicode.c
+++ b/src/util/unicode.c
@@ -25,7 +25,7 @@
 #define B00000001 0x01
 
 static bool char_trimmed(wint_t wc) {
-    if (wc >= 0x2000 && wc <= 0x200f) {
+    if (wc == 0xa0 || wc == 0x202f || (wc >= 0x2000 && wc <= 0x200f)) {
         /* only weird stuff here */
         return true;
     }
diff --git a/src/util/unicode.test.c b/src/util/unicode.test.c
index 8a5d77488..f746c77da 100644
--- a/src/util/unicode.test.c
+++ b/src/util/unicode.test.c
@@ -154,7 +154,7 @@ static void test_unicode_compare(CuTest *tc)
 }
 
 static void test_unicode_trim_zwnj(CuTest *tc) {
-    const char zwnj[] = { 0xe2, 0x80, 0x8c, 0x00 };
+    const char zwnj[] = { 0xe2, 0x80, 0x8c, 0 };
     char name[64];
     char expect[64];
     snprintf(name, sizeof(name), "%sA%sB%s  ", zwnj, zwnj, zwnj);
@@ -163,8 +163,38 @@ static void test_unicode_trim_zwnj(CuTest *tc) {
     CuAssertStrEquals(tc, expect, name);
 }
 
+static void test_unicode_trim_nbsp(CuTest *tc) {
+    const char code[] = { 0xc2, 0xa0, 0 };
+    char name[64];
+    char expect[64];
+    snprintf(name, sizeof(name), "%sA%sB%s  ", code, code, code);
+    snprintf(expect, sizeof(expect), "A%sB", code);
+    CuAssertIntEquals(tc, 6, unicode_utf8_trim(name));
+    CuAssertStrEquals(tc, expect, name);
+}
+
+static void test_unicode_trim_nnbsp(CuTest *tc) {
+    const char code[] = { 0xe2, 0x80, 0xaf, 0 };
+    char name[64];
+    char expect[64];
+    snprintf(name, sizeof(name), "%sA%sB%s  ", code, code, code);
+    snprintf(expect, sizeof(expect), "A%sB", code);
+    CuAssertIntEquals(tc, 8, unicode_utf8_trim(name));
+    CuAssertStrEquals(tc, expect, name);
+}
+
+static void test_unicode_trim_figure_space(CuTest *tc) {
+    const char code[] = { 0xe2, 0x80, 0x87, 0 };
+    char name[64];
+    char expect[64];
+    snprintf(name, sizeof(name), "%sA%sB%s  ", code, code, code);
+    snprintf(expect, sizeof(expect), "A%sB", code);
+    CuAssertIntEquals(tc, 8, unicode_utf8_trim(name));
+    CuAssertStrEquals(tc, expect, name);
+}
+
 static void test_unicode_trim_ltrm(CuTest *tc) {
-    const char ltrm[] = { 0xe2, 0x80, 0x8e, 0x00 };
+    const char ltrm[] = { 0xe2, 0x80, 0x8e, 0 };
     char name[64];
     char expect[64];
     snprintf(name, sizeof(name), "%sBrot%szeit%s  ", ltrm, ltrm, ltrm);
@@ -188,6 +218,9 @@ CuSuite *get_unicode_suite(void)
     CuSuite *suite = CuSuiteNew();
     SUITE_ADD_TEST(suite, test_unicode_trim);
     SUITE_ADD_TEST(suite, test_unicode_trim_zwnj);
+    SUITE_ADD_TEST(suite, test_unicode_trim_nbsp);
+    SUITE_ADD_TEST(suite, test_unicode_trim_nnbsp);
+    SUITE_ADD_TEST(suite, test_unicode_trim_figure_space);
     SUITE_ADD_TEST(suite, test_unicode_trim_ltrm);
     SUITE_ADD_TEST(suite, test_unicode_trim_emoji);
     SUITE_ADD_TEST(suite, test_unicode_utf8_to_other);