Fix ascii_only? flag in strio_write (#77)

tompng · web-flow · commit b31a538576a9 · 2024-01-18T12:42:15.000+09:00
Followup of #79 `rb_str_resize()` was changed by ruby/ruby@b0b9f72 . ```c rb_str_resize(string, shorter) // clear ENC_CODERANGE in some case rb_str_resize(string, longer) // does not clear ENC_CODERANGE anymore ``` ```c // rb_str_resize in string.c if (slen > len && ENC_CODERANGE(str) != ENC_CODERANGE_7BIT) { ENC_CODERANGE_CLEAR(str); } ``` I think this change is based on an assumption that appending null bytes will not change flag `ascii_only?`. `strio_extend()` will make the string longer if needed, and update the flags correctly for appending null bytes. Before `memmove()`, we need to `rb_str_modify()` because updated flags are not updated for `memmove()`.
diff --git a/ext/stringio/stringio.c b/ext/stringio/stringio.c
@@ -915,9 +915,6 @@ strio_extend(struct StringIO *ptr, long pos, long len)
 	if (pos > olen)
 	    MEMZERO(RSTRING_PTR(ptr->string) + olen, char, pos - olen);
     }
-    else {
-	rb_str_modify(ptr->string);
-    }
 }
 
 /*
@@ -1464,14 +1461,9 @@ strio_write(VALUE self, VALUE str)
 	}
     }
     else {
-	int cr0 = ENC_CODERANGE(ptr->string);
-	int cr = ENC_CODERANGE_UNKNOWN;
-	if (rb_enc_asciicompat(enc) && rb_enc_asciicompat(enc2)) {
-	    cr = ENC_CODERANGE_AND(cr0, ENC_CODERANGE(str));
-	}
 	strio_extend(ptr, ptr->pos, len);
+	rb_str_modify(ptr->string);
 	memmove(RSTRING_PTR(ptr->string)+ptr->pos, RSTRING_PTR(str), len);
-	if (cr != cr0) ENC_CODERANGE_SET(ptr->string, cr);
     }
     RB_GC_GUARD(str);
     ptr->pos += len;
diff --git a/test/stringio/test_stringio.rb b/test/stringio/test_stringio.rb
@@ -961,6 +961,12 @@ def test_coderange_after_overwrite
     assert_predicate(s.string, :ascii_only?)
     s.write "\u{431 43e 433 443 441}"
     assert_not_predicate(s.string, :ascii_only?)
+
+    s = StringIO.new("\u{3042}")
+    s.rewind
+    assert_not_predicate(s.string, :ascii_only?)
+    s.write('aaaa')
+    assert_predicate(s.string, :ascii_only?)
   end
 
   def assert_string(content, encoding, str, mesg = nil)

Original file line number	Diff line number	Diff line change
`@@ -915,9 +915,6 @@ strio_extend(struct StringIO *ptr, long pos, long len)`
`915`	`915`	`if (pos > olen)`
`916`	`916`	`MEMZERO(RSTRING_PTR(ptr->string) + olen, char, pos - olen);`
`917`	`917`	`}`
`918`		`- else {`
`919`		`- rb_str_modify(ptr->string);`
`920`		`- }`
`921`	`918`	`}`
`922`	`919`
`923`	`920`	`/*`
`@@ -1464,14 +1461,9 @@ strio_write(VALUE self, VALUE str)`
`1464`	`1461`	`}`
`1465`	`1462`	`}`
`1466`	`1463`	`else {`
`1467`		`- int cr0 = ENC_CODERANGE(ptr->string);`
`1468`		`- int cr = ENC_CODERANGE_UNKNOWN;`
`1469`		`- if (rb_enc_asciicompat(enc) && rb_enc_asciicompat(enc2)) {`
`1470`		`- cr = ENC_CODERANGE_AND(cr0, ENC_CODERANGE(str));`
`1471`		`- }`
`1472`	`1464`	`strio_extend(ptr, ptr->pos, len);`
	`1465`	`+ rb_str_modify(ptr->string);`
`1473`	`1466`	`memmove(RSTRING_PTR(ptr->string)+ptr->pos, RSTRING_PTR(str), len);`
`1474`		`- if (cr != cr0) ENC_CODERANGE_SET(ptr->string, cr);`
`1475`	`1467`	`}`
`1476`	`1468`	`RB_GC_GUARD(str);`
`1477`	`1469`	`ptr->pos += len;`