CHANGELOG.md

@@ -10,17 +10,14 @@
   `source_addr` of `Kaffeé` previously reached the application as `Kaffei`, because the codec wrote
   the octet and then masked bit 7 reading it back; `destination_addr`, `system_id`, `message_id`,
   `service_type` and the C-Octet String TLVs were affected the same way. A character past `U+00FF`
-  in one of those fields is now refused, where it used to go out as its low octet — which for `一`,
+  in one of those fields is now refused, where it used to go out as its low octet.
-  `　` and most emoji is `0x00`, ending the field there.
 
   **Check what you stored before you roll this out.** Values your application persisted under 0.5.0
   were read with bit 7 masked, so an address or a `message_id` carrying an octet above `0x7F` is
   spelled differently now: a stored id will not match the receipt it belongs to, and a stored address
   will not match the sender it came from. Ids most SMSCs issue are digits or hex and are unaffected.
 - A `U+0000` inside a C-Octet String — `source_addr`, `message_id`, `system_id` and the rest — is
-  refused. The peer reads such a field to its first NULL, so one sent inside the value shifted every
+  refused. An Octet String carries a NULL as before.
-  mandatory field behind it while `command_length` still counted the whole string. An Octet String
-  carries a NULL as before; its length octet is what ends it.
 
 ## 0.5.0
 

README.md

@@ -288,11 +288,11 @@ await session.sendSms({
 ```
 
 **Addresses.** `sourceAddrTon` and `destinationAddrTon` default to 5 for an alphanumeric address
-and 1 for a numeric one; the NPI fields default to 0. An address is latin1: `Kaffeé` goes out as
+and 1 for a numeric one; the NPI fields default to 0. An address is latin1, so `é` is one octet on
-the six octets that spell it, `4B 61 66 66 65 E9`, and an address you received always sends back.
+the wire and an address you received always sends back. One outside `/^[\u0001-\u00FF]*$/` is
-One outside `/^[\u0001-\u00FF]*$/` is refused, naming the character and its index — strip or
+refused, naming the character and its index — strip or transliterate it first. An SMSC may still
-transliterate it first. An SMSC may still refuse a non-ASCII sender of its own accord, which
+refuse a non-ASCII sender of its own accord, which reaches you as a refusal such as
-reaches you as a refusal such as `ESME_RINVSRCADR`.
+`ESME_RINVSRCADR`.
 
 **Encoding.**
 
@@ -614,8 +614,7 @@ if (isCommand(pduObj, 'submit_sm')) {
   names, detected from the text where you name none. One that alphabet cannot carry is refused,
   naming the character, its code point and where it is.
 - Every text field is latin1: addresses, `system_id`, `message_id`, `service_type` and the C-Octet
-  String TLVs. A character past `U+00FF` is refused, as is a `U+0000` in a C-Octet String, which the
+  String TLVs. A character past `U+00FF` is refused, as is a `U+0000` in a C-Octet String.
-  peer reads as the end of the field.
 - A `Buffer` goes out exactly as given under any `data_coding`: binary payloads, hand-built user
   data headers, deliberately malformed bodies.
 - `session.send()` and `session.sendReturn()` build through the same codec and refuse the same bodies.

docs/decisions.md

@@ -488,12 +488,10 @@ rule and an index of the titles below.
   Goal 2 settles the refusals, both of them a `size()` that would have agreed with a `write()` that
   put something else on the wire: a character past `U+00FF` written as its low octet, and a caller's
   own `U+0000`, which a mandatory field's reader takes as the end of the field. Goal 4 settles them
-  twice over: for a great many characters that low octet is `0x00`, and the PDU went out malformed
+  twice over: for one character in every 256 that low octet is `0x00`, and the PDU went out malformed
   on the operator's parser. `wantText()` and `wantCstringText()` are the only two places that decide
-  it, which is why the `dest_address` and
+  it. Rejected: reading latin1 and leaving the write spelled ASCII, which leaves two halves agreeing
-  `unsuccess_sme` structures write their embedded addresses through `cstring.write()` rather than
+  only by accident. Rejected: refusing the upper half on send
-  reaching past it into `writeCstring()`. Rejected: reading latin1 and leaving the write spelled
-  ASCII, which leaves two halves agreeing only by accident. Rejected: refusing the upper half on send
   to stay strict to 3.4's ASCII, which would be a new restriction taking away traffic this library
   already sends and operators already accept, on no defect. Rejected: refusing `U+0000` in every
   text field, which would buy one spelling by taking a legitimate octet away from the

test/unsendable.test.ts

@@ -366,7 +366,7 @@ describe('an address the field cannot carry', () => {
 			assert.deepEqual(sent.smsIds, [], address);
 		}
 
-		assert.deepEqual(smsc.octets, [], 'an address the field cannot carry never reaches the socket');
+		assert.deepEqual(smsc.octets, []);
 	});
 });