diff --git a/src/base_core.h b/src/base_core.h
index 4496069..4899410 100644
--- a/src/base_core.h
+++ b/src/base_core.h
@@ -252,6 +252,18 @@ typedef unsigned __int128 U128;
 #define MemAppend(dst, src, len)    mempcpy(dst, src, len)
 #define MemCompare(first, second, len) memcmp(first, second, len)
 
+static INLINE U8 *MemFind(U8 *mem, U64 mem_len, U8 *needle, U64 needle_len)
+{
+  if (needle_len > mem_len) {
+      return NULL;
+  }
+  for (U64 i = 0 ; i + needle_len <= mem_len; i++) {
+      if (MemCompare(&mem[i], needle, needle_len) == 0) {
+        return &mem[i];
+      }
+  }
+  return NULL;
+}
 // atomics
 #define Atomic(type) _Atomic(type)
 #define AtomicSub(ptr, val) __atomic_fetch_sub((ptr), (val), __ATOMIC_SEQ_CST)
diff --git a/src/main.c b/src/main.c
index 863126a..63c6aa4 100644
--- a/src/main.c
+++ b/src/main.c
@@ -470,7 +470,35 @@ void string_test_misc(void)
     Str str4_trim = str_ascii_trim(str4);
     printf("str4_trim: '%.*s'\n", str4_trim.len, str4_trim.str);
 
+    S64 found = str_substr(str1, S("World"));
+    printf("1. found %ld\n", found);
+    found = str_substr(str_empty(), S("World"));
+    printf("2. found %ld\n", found);
+    found = str_substr(str_empty(), S(""));
+    printf("3. found %ld\n", found);
+    found = str_substr(str1, S(" "));
+    printf("4. found %ld\n", found);
+
 }
+void string_test_join(void)
+{
+    TRACEF("\n");
+    Arena arena;
+    arena_init(&arena, &ArenaOptions_Default);
+    Allocator allocator = arena_allocator(&arena);
+    Str strs[3] = {S("1"), S("two"), S("3-3")};
+    StrVec vec = STRVEC_STATIC_INITIALIZER(strs);
+
+    Str str = str_join(&allocator, vec, S("/"));
+    printf("str_join 1 '%.*s'\n", str.len, str.str);
+    str = str_join(&allocator, vec, S(""));
+    printf("str_join 2 '%.*s'\n", str.len, str.str);
+    vec.len = 0;
+    str = str_join(&allocator, vec, S("/"));
+    printf("str_join 2 '%.*s'\n", str.len, str.str);
+
+}
+
 int main(int argc, char *argv[])
 {
     arena_test_1();
@@ -486,6 +514,7 @@ int main(int argc, char *argv[])
     string_test_1();
     string_split_test_1();
     string_test_misc();
+    string_test_join();
 
 }
 
diff --git a/src/string.c b/src/string.c
index 8e0ce2e..d8a88d4 100644
--- a/src/string.c
+++ b/src/string.c
@@ -32,9 +32,7 @@ Str str_concat(Allocator *alloc, Str first, Str second)
 Str str_push(Allocator *alloc, Str str)
 {
     U8 *mem = (U8 *)allocator_push(alloc, str.len, 1);
-    if (!str_is_empty(str)) {
-        MemCopy(mem, str.str, str.len);
-    }
+    MemCopy(mem, str.str, str.len);
 
     return str_create(mem, str.len);
 }
@@ -80,8 +78,21 @@ Str str_ascii_trim(Str str)
     return new_str;
 }
 
-S64 str_substr(Str haystack, Str to_find)
+S64 str_substr(Str haystack, Str needle)
 {
+    if (needle.len == 0) {
+        return 0;
+    }
+
+    if (needle.len == 0) {
+        return str_find_byte(haystack, needle.str[0]);
+    }
+
+    U8 *found = MemFind(haystack.str, haystack.len, needle.str, needle.len);
+    if (found != NULL) {
+        return (S64)(found - haystack.str);
+    }
+
     return -1;
 }
 
@@ -112,3 +123,33 @@ void str_split(Str str, Str separators, StrVec *vec)
         vec->len++;
     }
 }
+
+Str str_join(Allocator *allocator, StrVec strings, Str separator)
+{
+
+    U32 str_len = 0;
+
+    for (U32 i = 0; i < strings.len; i++) {
+        str_len += strings.strings[i].len;
+    }
+    if (strings.len > 0) {
+        str_len += separator.len * (strings.len - 1);
+    }
+
+    U8 *joined_str = allocator_push(allocator, str_len, 1);
+    if (strings.len > 0) {
+        U32 pos = 0;
+        Str *str = &strings.strings[0];
+        MemCopy(&joined_str[pos], str->str, str->len);
+        pos += str->len;
+        for (U32 i = 1; i < strings.len; i++) {
+            str = &strings.strings[i];
+            MemCopy(&joined_str[pos], separator.str, separator.len);
+            pos += separator.len;
+            MemCopy(&joined_str[pos], str->str, str->len);
+            pos += str->len;
+        }
+    }
+
+    return str_create(joined_str, str_len);
+}
diff --git a/src/string.h b/src/string.h
index 33592c9..f9bb00c 100644
--- a/src/string.h
+++ b/src/string.h
@@ -9,7 +9,7 @@ struct Str {
 
 Str str_empty(void);
 Str str_slice(Str str, U32 start, U32 len);
-S64 str_substr(Str haystack, Str to_find);
+S64 str_substr(Str haystack, Str needle);
 
 Str str_concat(Allocator *alloc, Str first, Str second);
 Str str_push(Allocator *alloc, Str str);
@@ -99,12 +99,15 @@ struct StrVec {
     U32 len;
     U32 max_len;
 };
-#define STRVEC_STATIC_INITIALIZER(strings_) STRVEC_INITIALIZER((strings_), ArraySize(strings_))
-#define STRVEC_INITIALIZER(strings_, len_) {.strings = (strings_), .len = 0, .max_len = (len_)}
+#define STRVEC_STATIC_INITIALIZER(strings_) STRVEC_INITIALIZER((strings_), ArraySize(strings_), ArraySize(strings_))
+#define STRVEC_STATIC_INITIALIZER_EMPTY(strings_) STRVEC_INITIALIZER((strings_), 0, ArraySize(strings_))
+#define STRVEC_INITIALIZER(strings_, len_, max_len_) {.strings = (strings_), .len = (len_), .max_len = (max_len_)}
+
 #define STRVEC_DEFINE(var_name, len_)  \
     struct { \
         Str strings[len_];\
     } var_name##_strvec_array; \
-    StrVec var_name = STRVEC_STATIC_INITIALIZER(var_name##_strvec_array.strings)
+    StrVec var_name = STRVEC_STATIC_INITIALIZER_EMPTY(var_name##_strvec_array.strings)
 
 void str_split(Str str, Str separators, StrVec *vec);
+Str str_join(Allocator *allocator, StrVec strings, Str separator);