114 Commits

Author	SHA1	Message	Date
lilleman	57f53d56f1	Give the link's liveness one owner	2026-09-28 21:55:46 +02:00
lilleman	8721491cd5	Drop the store operator from the README's personas until the store ships	2026-09-28 21:05:05 +02:00
lilleman	2142c2e3bb	Spell out the hand-wired ESME's bind, and where a getter-only assignment throws	2026-09-28 04:08:53 +02:00
lilleman	4bdb1e64b0	Take bound()'s bind type as a string, name the fix in its refusals, and show the ESME's spelling	2026-09-28 04:07:44 +02:00
lilleman	2ac72fb3f4	Refuse a bind bound() cannot record, and document it for a 0.4.0 consumer	2026-09-28 04:01:51 +02:00
lilleman	b70f0d6ade	Let the session own its bind state, written by bound() alone	2026-09-28 03:55:40 +02:00
lilleman	dd09a27dc6	Export isTlvName, so a string can be narrowed into tlvs	2026-09-27 15:39:43 +02:00
lilleman	cda8e9b112	Relay a parsed PDU's TLVs as they are, refuse only a tagId that disagrees with its key, and refuse a number for an octet TLV	2026-09-27 15:29:31 +02:00
lilleman	939eda7269	Key a TLV input like the read side, drop tagId, copy TLV octets at parse, and export only Tlvs, TlvInputs and Tlv	2026-09-27 15:18:08 +02:00
lilleman	75d4794522	Type each known TLV's value by its tag, on read and on write	2026-09-27 15:12:09 +02:00
lilleman	42b1239f3d	Name the retry status by direction, and correct what the held-message decision claims	2026-09-26 14:53:09 +02:00
lilleman	f2977638f2	Stop refusing only once the held messages are down to half, so a full window warns once, and name the octets in the warning	2026-09-26 14:44:00 +02:00
lilleman	b9167073d4	Warn once at the held-message bound, tell receivers sendResp() is what releases a message, and pin that a refused segment joins no group	2026-09-26 14:41:10 +02:00
lilleman	fb6b125aab	Give the dumbclient peers a namespace owner that outlives them, count window 2000's throttled messages, and correct what the findings blamed on the peer	2026-09-26 14:37:20 +02:00
lilleman	855ccbe75f	Refuse a message past the held-message bound with ESME_RTHROTTLED, and answer a full reassembly buffer the same way	2026-09-26 13:18:56 +02:00
lilleman	c3ebd6d6bc	Cap the heap unanswered messages hold, and detach them from the chunk they arrived in	2026-09-25 19:22:31 +02:00
lilleman	72b400e661	Name the per-occurrence charge exactly	2026-09-25 19:06:20 +02:00
lilleman	f1ef6b6643	State the octet cap's charge as a rule a consumer can size with	2026-09-25 19:05:48 +02:00
lilleman	01210e734a	Correct the doc claims the prose pass found, and move two decisions into the log	2026-09-24 00:31:49 +02:00
lilleman	11ff349333	Lead the changelog with the read-side break, and show each repeatable TLV's shape	2026-09-24 00:27:01 +02:00
lilleman	a885913ecb	Link AGENTS.md absolutely, as npm renders the README	2026-09-24 00:19:28 +02:00
lilleman	befa648991	Point goal 1 at the defect table where it lives	2026-09-24 00:19:23 +02:00
lilleman	cbfd6e0fbe	Count ten goals, and point at where the hard rules live	2026-09-24 00:19:19 +02:00
lilleman	10877c9f5f	Read every occurrence of a repeatable TLV, and drop the unread tlvMap	2026-09-24 00:13:38 +02:00
lilleman	228b81aa5e	Say where a failed send's taken segments report, and that an unanswered one fails it too	2026-09-23 23:29:59 +02:00
lilleman	e725f62b95	Make a failed send the whole message's answer, with no merged report after it	2026-09-23 23:29:08 +02:00
lilleman	0ee7bbbaf0	Make goal 2 cover a value we could not send as the caller wrote it	2026-09-23 23:06:35 +02:00
lilleman	bdebef849c	Correct three claims a reader would have built on	2026-09-21 10:39:54 +02:00
lilleman	36a1c6ea7b	Cut the prose the code, the error and the record already carry	2026-09-21 08:25:39 +02:00
lilleman	a7eb4923f5	Say what is true for a TLV too, and cover the to half of the check	2026-09-21 08:20:59 +02:00
lilleman	dc690b912f	Name the option a refused address came from, and the data a 0.5.0 consumer stored	2026-09-21 08:14:08 +02:00
lilleman	f870b7530f	Cover the guarded address writers, and cut what the code already says	2026-09-21 08:08:42 +02:00
lilleman	c1e0407942	Refuse a NULL inside a C-Octet String, which ends the field on the peer	2026-09-21 07:59:08 +02:00
lilleman	7db242e375	Read and write every text field on the wire as latin1	2026-09-21 07:54:25 +02:00
lilleman	c5ed25f915	Make throughput goal 6, with floors a release has to clear	2026-09-20 23:23:08 +02:00
lilleman	a24f350b7a	Prove the default bound is wired, and name the option on the failure a consumer reads	2026-09-20 20:41:45 +02:00
lilleman	1416615ca3	Bound every connect attempt at 10 s by default, with false the way to opt out	2026-09-20 20:34:53 +02:00
lilleman	8f416c2bc4	Name the peer and the stalled phase on a connect timeout, and quote an untyped value	2026-09-20 18:17:39 +02:00
lilleman	17447bbe3a	Give up a connect after connectTimeout instead of waiting the OS out	2026-09-20 00:42:09 +02:00
lilleman	c90b855958	Drop the anti-mention from the audience	2026-09-19 00:59:29 +02:00
lilleman	0f13951e8c	Move the goals to README.md and state the audience	2026-09-19 00:48:21 +02:00
lilleman	793839d0e0	Point the package and README at Gitea	2026-09-13 17:57:10 +02:00
lilleman	7310634d7d	Rewrite the README task-first for 1.0.0 and move the 0.4.0 guide to MIGRATION.md (#102)	2026-09-13 12:05:13 +02:00
lilleman	13e3ff8f1f	Say undefined where the SMSC named no id for a segment it took (#101) ... * Cover the ids a send gets back where the SMSC named only some of them * Say undefined where the SMSC named no id for a segment it took * Record why the merger takes the wider type, and de-duplicate the interop narrowing * Guard the numbering hole the merger must refuse, and correct the phase record	2026-09-12 22:37:55 +02:00
lilleman	ca1a7473ed	Declare the alphabet a GSM message is actually written in (#100) ... * Regression tests for the alphabet a GSM message declares * Declare the alphabet a GSM message is actually written in * Reflow the composing-by-hand paragraph * Drop the ASCII alias SMPP's flat table shares with the option's own * Bound the receipt test's wait and tighten the notes around it	2026-09-09 21:30:55 +02:00
lilleman	989eb01763	Name the refusal rule, and refuse a period the encoder cannot spell (#99) ... * Refuse a relative validity period the SMPP time format cannot hold * Export unencodableText(), so a caller can word a refusal as the library does * Name the refusal rule in the README, and correct what it claims * Word the period refusal by what the encoder spells, and keep unencodableText() internal * Correct the decision bullet premise about what the relative format holds * Close the stability nits on the time guard and its tests	2026-09-09 20:36:13 +02:00
lilleman	ef13290230	Refuse a body the PDU's own data_coding cannot carry (#98) ... * Regression tests for a body the PDU's own data_coding cannot carry * Refuse a body the PDU's own data_coding cannot carry, in the codec both send paths build through * Record the codec's body refusal, and drop two todo entries it and the interop run close * Regression tests for the aliased body TLV and the data_coding short_message settles * Find the body TLV by tag id, leave data_coding to the field that will be read, and correct the record * Record the two architecture findings this change does not close * Regression tests for a duplicated body tag and a short_message the wire never carries * Resolve every body TLV against the field that will be read, and move TLV writing to its table * Regression tests for a body only the TLV carries and an empty short_message * Let only octets the command writes settle the alphabet, and stop shadowing the TLV table	2026-09-09 19:33:36 +02:00
lilleman	c06cc0648b	Refuse what we cannot send: an alphabet that cannot carry the message, and a time nobody can read (#97) ... * Regression tests for a named alphabet and a time that cannot carry the send * Refuse a message the named alphabet cannot carry, and a time nobody can read * Close the architecture review: a builder that cannot drop a time, and docs that hold * Refuse a time of no known kind, and clamp a relative period to what the format holds * Reflow the decision record and name the clamp in its test	2026-09-09 18:16:08 +02:00
lilleman	7d6039796d	Fit every concatenated segment in an SMS, and close the encoding helpers' throw (#96) ... * Regression tests for the concatenated segment budget and the encoding-name door * Budget unpacked segments at 134 octets, and export isEncodingName * Name the length a Latin-1 message starts costing more segments at * Fold the totality clause into hard rule 1, and log the unrepresentable-alphabet finding * Make the maxSegments regression discriminate, and drop the narrating comments * State the per-alphabet capacities accurately, and log the invalid-date path	2026-09-09 17:21:51 +02:00
lilleman	e80b07167c	Read the GSM 03.38 message class where the spec puts it (#95) ... * Regression tests for the GSM 03.38 message class behind sms.flash * Read the GSM 03.38 message class where the spec puts it, and refuse a flash message no alphabet can carry * Keep the code span in the message-class decision on one line * Stop offering the flash message class as an alphabet, and check the encoding option by name * Settle every refusable send option in one check, and record what the review left open * Say what the flash refusal and the encoding option actually do * Keep the README off the alphabet whose long messages do not fit	2026-09-09 16:40:32 +02:00