Matthew Meszaros 06b8db5529 feat: report the one silent state a campaign had no words for, a mailbox pool that is part out of budget and part outside its own sending hours, which fell between the 'every mailbox is capped, sending resumes tomorrow' line and the deliberately unlogged closed-hours band and so left an active campaign sending nothing with an empty activity feed; give it its own line under the mailboxes_unavailable event the pool's other refusals already use, naming how many mailboxes are in each state and carrying the moment the pool comes back in metadata rather than promising a day, while leaving daily_cap_reached to mean exactly what it meant before, every usable mailbox spent and nothing coming back until tomorrow

2026-09-12 03:47:36 -07:00

..

behavior_test.go

feat: give each mailbox a human sending persona (randomized daily and hourly caps, send spacing, work start/end, lunch break and working weekdays, rolled once per local day in the mailbox's own timezone and applied across the campaign, warmup and smart-send schedulers), add campaign auto-pause guardrails that stop a campaign when its bounce, complaint or reply rate leaves the configured band, make mailbox rotation actually rotate for tag-resolved and all-mailbox campaigns, stop every scheduler from ever returning a slot in the past, and correct the mailbox min-gap field that stored seconds while labelling them minutes

2026-08-13 16:51:29 +02:00

behavior.go

feat: stop one un-sendable lead parking a whole campaign and stop the contact drawer's next-action time walking forward on every refresh (issue #437): route up to config.CampaignPlacementCandidates due leads per pass instead of one, classify a placement refusal that belongs to a single lead (ESP-strict finding no mailbox for that recipient's provider, a bound lead inside its own mailbox's minimum gap or waiting for it to reopen, a recipient's send-time-optimized hours) as the new ErrLeadDeferred so the pass moves to the lead behind them and only defers the campaign when every candidate is refused, log the ESP-strict deferral once a day rather than once per refused lead per tick, and make PreviewContactSend a pure read that answers unchanged state identically on every call by running placement with the even-distribution, jitter, conflict-resolution, distribution-curve and sub-minute layers off, taking a behaviour profile's gap at its floor instead of drawing it, picking the mailbox deterministically instead of re-rolling rotation, reporting the next sending day's first open minute instead of a jittered twenty-four-hours-from-now, and reporting a due step's time as the campaign chain's own stored wakeup

2026-09-12 02:58:48 -07:00

campaign_overdue_window_live_test.go

feat: stop a campaign sending outside its sending window, and past its end date, when a follow-up is overdue: the placer's schedule gates are asked about the moment a step became due, and nextScheduleSlot deliberately returns an instant that was already inside a window unchanged, so a step that came due at 2pm still read as due at 11pm and the task sent it there, while the end-date comparison found a candidate predating the end date; floor an overdue candidate at now so the window, the weekday mask and the end date are all asked about the send that is actually about to happen, with live tests for a closed window, a passed end date and the overdue step that must still go out promptly while the window is open

2026-09-12 03:29:33 -07:00

campaign_scheduler.go

feat: report the one silent state a campaign had no words for, a mailbox pool that is part out of budget and part outside its own sending hours, which fell between the 'every mailbox is capped, sending resumes tomorrow' line and the deliberately unlogged closed-hours band and so left an active campaign sending nothing with an empty activity feed; give it its own line under the mailboxes_unavailable event the pool's other refusals already use, naming how many mailboxes are in each state and carrying the moment the pool comes back in metadata rather than promising a day, while leaving daily_cap_reached to mean exactly what it meant before, every usable mailbox spent and nothing coming back until tomorrow

2026-09-12 03:47:36 -07:00

campaign_sender.go

feat: stop one un-sendable lead parking a whole campaign and stop the contact drawer's next-action time walking forward on every refresh (issue #437): route up to config.CampaignPlacementCandidates due leads per pass instead of one, classify a placement refusal that belongs to a single lead (ESP-strict finding no mailbox for that recipient's provider, a bound lead inside its own mailbox's minimum gap or waiting for it to reopen, a recipient's send-time-optimized hours) as the new ErrLeadDeferred so the pass moves to the lead behind them and only defers the campaign when every candidate is refused, log the ESP-strict deferral once a day rather than once per refused lead per tick, and make PreviewContactSend a pure read that answers unchanged state identically on every call by running placement with the even-distribution, jitter, conflict-resolution, distribution-curve and sub-minute layers off, taking a behaviour profile's gap at its floor instead of drawing it, picking the mailbox deterministically instead of re-rolling rotation, reporting the next sending day's first open minute instead of a jittered twenty-four-hours-from-now, and reporting a due step's time as the campaign chain's own stored wakeup

2026-09-12 02:58:48 -07:00

cold_ramp.go

feat: ease a graduating mailbox into cold volume (#231)

2026-08-28 10:47:03 -07:00

contact_preview_live_test.go

feat: stop one un-sendable lead parking a whole campaign and stop the contact drawer's next-action time walking forward on every refresh (issue #437): route up to config.CampaignPlacementCandidates due leads per pass instead of one, classify a placement refusal that belongs to a single lead (ESP-strict finding no mailbox for that recipient's provider, a bound lead inside its own mailbox's minimum gap or waiting for it to reopen, a recipient's send-time-optimized hours) as the new ErrLeadDeferred so the pass moves to the lead behind them and only defers the campaign when every candidate is refused, log the ESP-strict deferral once a day rather than once per refused lead per tick, and make PreviewContactSend a pure read that answers unchanged state identically on every call by running placement with the even-distribution, jitter, conflict-resolution, distribution-curve and sub-minute layers off, taking a behaviour profile's gap at its floor instead of drawing it, picking the mailbox deterministically instead of re-rolling rotation, reporting the next sending day's first open minute instead of a jittered twenty-four-hours-from-now, and reporting a due step's time as the campaign chain's own stored wakeup

2026-09-12 02:58:48 -07:00

contact_preview_test.go

feat: address Greptile on PR #260 by only linking meeting join_url values whose scheme is http or https in the contact timeline (both the meta Join link and the expanded detail), projecting the campaign's daily ramp advance in memory before the read-only next-send preview so a preview on a new UTC day budgets with the level the next scheduler pass will persist (with a unit test), and trimming the campaign-state model, preview and service comments to one-line invariants

2026-08-29 03:51:24 -07:00

contact_preview.go

feat: stop one un-sendable lead parking a whole campaign and stop the contact drawer's next-action time walking forward on every refresh (issue #437): route up to config.CampaignPlacementCandidates due leads per pass instead of one, classify a placement refusal that belongs to a single lead (ESP-strict finding no mailbox for that recipient's provider, a bound lead inside its own mailbox's minimum gap or waiting for it to reopen, a recipient's send-time-optimized hours) as the new ErrLeadDeferred so the pass moves to the lead behind them and only defers the campaign when every candidate is refused, log the ESP-strict deferral once a day rather than once per refused lead per tick, and make PreviewContactSend a pure read that answers unchanged state identically on every call by running placement with the even-distribution, jitter, conflict-resolution, distribution-curve and sub-minute layers off, taking a behaviour profile's gap at its floor instead of drawing it, picking the mailbox deterministically instead of re-rolling rotation, reporting the next sending day's first open minute instead of a jittered twenty-four-hours-from-now, and reporting a due step's time as the campaign chain's own stored wakeup

2026-09-12 02:58:48 -07:00

daily_budget_live_test.go

feat: pick a closed-hours mailbox that stays closed for the next ten minutes so the scheduling pass cannot run after it opens, and assert the mixed-pool deferral lands on the reopening from both sides instead of accepting any earlier time

2026-09-04 02:59:00 -07:00

domain_auth_test.go

feat: enforce the persisted SPF/DKIM/DMARC state as a real cold-send and warmup gate behind a 72h grace clock and an operator toggle, after first fixing the DMARC organizational-domain fallback in dnsauth so a dedicated sending subdomain covered by its parent's record stops reading as unauthenticated, adding auth_state to the four mailbox loaders that never selected it (which would have made the gate dead code), stamping auth_failing_since on entry to failing so a resolver hiccup can never stop a campaign, notifying the org on that transition, and reporting an all-gated pool as ErrDomainAuthFailing instead of a message about sending windows (#160)

2026-08-22 09:37:26 -07:00

email_scheduler.go

feat: give each mailbox a human sending persona (randomized daily and hourly caps, send spacing, work start/end, lunch break and working weekdays, rolled once per local day in the mailbox's own timezone and applied across the campaign, warmup and smart-send schedulers), add campaign auto-pause guardrails that stop a campaign when its bounce, complaint or reply rate leaves the configured band, make mailbox rotation actually rotate for tag-resolved and all-mailbox campaigns, stop every scheduler from ever returning a slot in the past, and correct the mailbox min-gap field that stored seconds while labelling them minutes

2026-08-13 16:51:29 +02:00

entry_delay_live_test.go

feat: carry the unexcluded routing pass's own re-check time back when the daily new-lead cap is spent, so a campaign whose only remaining leads are still inside the entry delay defers to their due time instead of reporting itself completed, with a live test that fails without it

2026-09-08 05:13:13 -07:00

errors.go

feat: stop one un-sendable lead parking a whole campaign and stop the contact drawer's next-action time walking forward on every refresh (issue #437): route up to config.CampaignPlacementCandidates due leads per pass instead of one, classify a placement refusal that belongs to a single lead (ESP-strict finding no mailbox for that recipient's provider, a bound lead inside its own mailbox's minimum gap or waiting for it to reopen, a recipient's send-time-optimized hours) as the new ErrLeadDeferred so the pass moves to the lead behind them and only defers the campaign when every candidate is refused, log the ESP-strict deferral once a day rather than once per refused lead per tick, and make PreviewContactSend a pure read that answers unchanged state identically on every call by running placement with the even-distribution, jitter, conflict-resolution, distribution-curve and sub-minute layers off, taking a behaviour profile's gap at its floor instead of drawing it, picking the mailbox deterministically instead of re-rolling rotation, reporting the next sending day's first open minute instead of a jittered twenty-four-hours-from-now, and reporting a due step's time as the campaign chain's own stored wakeup

2026-09-12 02:58:48 -07:00

helpers.go

feat: bind a campaign lead to the mailbox that sends its first email so every follow-up leaves from the same address, holding a lead back while its mailbox is merely out of budget or outside its hours and moving it to another mailbox only when that one can no longer send for the campaign at all

2026-09-09 08:51:00 -07:00

humanize_test.go

test: cover the gmail/graph message mapping (read-state inversion, spam/category labels, unpadded base64url bodies, warmup and classification header surfacing), the imap header-flag parser, the graph MIME builder, and the human-behavior timing (sub-minute randomisation, daily volume factor, night-deferred and heavy-tailed engagement)

2026-07-04 11:45:01 +02:00

live_integration_test.go

feat: bind a campaign lead to the mailbox that sends its first email so every follow-up leaves from the same address, holding a lead back while its mailbox is merely out of budget or outside its hours and moving it to another mailbox only when that one can no longer send for the campaign at all

2026-09-09 08:51:00 -07:00

no_eligible_mailbox_test.go

feat: stop a freshly connected mailbox being silently excluded from every campaign send, by making an unset mailbox timezone representable as the empty string the campaign scheduler already checks for, since email_accounts.timezone defaulted to 'UTC' while campaigns.timezone defaults to 'Europe/London' and nothing in the OAuth or SMTP onboarding paths ever set either, so a brand new mailbox looked deliberately placed in UTC, was compared against the differing campaign zone and dropped by the hardcoded 8am-8pm business-hours gate whenever the current UTC hour fell outside it, emptying the candidate pool and failing the campaign start, adding a migration that changes the column default and converts existing 'UTC' rows because until now no API field, dashboard control or onboarding path could set that column at all so every such row is the old default rather than a choice, adding the missing Timezone field to UpdateEmail with IANA validation so the setting the sending-behaviour UI already tells people to change is finally reachable and an unloadable zone is rejected instead of being silently coerced to UTC by the scheduler, and replacing the misleading 'no active email accounts found for campaign's email tags' response for a pool that exists but is entirely gated out with a distinct message naming the real cause, via an ErrNoEligibleMailbox that wraps ErrNoEmailAccounts so the three callers that pause a campaign on it are unaffected (#126)

2026-08-16 07:46:37 +02:00

pacing_test.go

feat: bind a campaign lead to the mailbox that sends its first email so every follow-up leaves from the same address, holding a lead back while its mailbox is merely out of budget or outside its hours and moving it to another mailbox only when that one can no longer send for the campaign at all

2026-09-09 08:51:00 -07:00

parse_time_test.go

feat: make send-time optimization actually schedule sends (#226)

2026-08-28 09:46:39 -07:00

send_time_test.go

feat: make send-time optimization actually schedule sends (#226)

2026-08-28 09:46:39 -07:00

send_time.go

feat: make send-time optimization actually schedule sends (#226)

2026-08-28 09:46:39 -07:00

service.go

feat: give a cold mailbox a rotation lifecycle so a tired one can rest (#237)

2026-08-28 12:08:52 -07:00

tenancy_live_test.go

feat: scope the tag, category and folder registries and unibox conversation labels to the organization instead of the creating user, so a teammate sees and can edit the labels the owner made, splitting a label two workspaces shared into one copy each and guarding every label write against ids from another workspace (#457)

2026-09-12 03:37:31 -07:00

warmup_pool_type_test.go

feat: keep a restricted or suspended workspace out of the premium warmup pool by checking the organization's risk posture before the stored warmup_pool_type in resolveWarmupPoolType (tasks and email services), which is never empty so the restricted branch below it was dead code, make the scheduler's recipient-capacity count risk-aware the same way, wire the org risk repository into the email service, and refuse the premium membership move in UpdateEmailAccountWarmupPoolType while the owning organization is restricted so a worker rebalance cannot readmit a risky tenant

2026-08-28 21:26:57 -07:00

warmup_scheduler_test.go

feat: close the warmup ramp loop with an early-signal hold (#227)

2026-08-28 10:06:19 -07:00

warmup_scheduler.go

feat: address the Greptile review on the pool link: require https for the cloud URL (loopback exempt for local development) since the instance token and mailbox passwords travel on it, remove the cloud copy when the local enrollment row cannot be written so a mailbox never warms in both places, delete the local enrollment row before the cloud one and restore it if the cloud call fails so a mailbox is never left with no warmup anywhere, and trim the new multi-line comments to the one-line style

2026-08-29 07:17:10 -07:00