Repository navigation
Add 158 doc-sourced samples for verified-sources batch 1 (19 providers) - #69
Merged
Merged
Conversation
Copied from hookdeck/webhook-registry samples-doc/ at 7e60bde1 (rows added
in hookdeck/webhook-registry#101): agentmail, brevo, crusoe, green-api,
guesty, loop, nash, openmeter, pipedream, pleo, recall-ai, revenuecat,
rubrik, segment, stitch, taskrabbit, telegram, uphold, yoco. The other 17
rows in that PR have no doc sample, so they get no directory.
Sample files are unchanged. Two index.json files differ from the export:
- revenuecat: files under 1.0/ with latest_version "1.0" rather than
latest/. Every body carries "api_version": "1.0", which RevenueCat
documents as the webhook format version, and the README says a payload
api_version is the value.
- telegram: topic_identifier is left out. The export wrote a prose
description ("key-presence: ..."); the registry row has no path because
the update type is the name of the key besides update_id, which the
resolver cannot express.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01D7eqaQdKAUzQXRbGtym1PF
- nash: null. The event is composed from type + "." + event, so "type" named a capture by family only (courier_location.updated became courier_location), and different events would overwrite each other. This follows the strava and facebook convention for composed keys. - brevo: ["event", "msg_status", "event_name"], the registry's email, SMS and coupon family carriers. This resolves 8 of 8 samples, where "event" alone resolved 0. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01D7eqaQdKAUzQXRbGtym1PF
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
This branch was successfully deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Requested by Gareth · project thread
Before: hookdeck/webhook-registry#101 (verified sources batch 1, merged) added 36 provider rows and exported doc examples for 19 of them to the registry's
samples-doc/tree. None of those 19 had a directory here.After: 19 new provider directories with 158 doc-sourced samples.
providers.jsongoes from 155 to 174 providers. The other 17 rows have no published example body, so they get no directory: ada, azure-logic-apps, browserbase, dealhub, fathom, fillout, freshdesk, google-workspace, gorgias, instantly, n8n, nango, sendoso, sovos, vanta, wetravel and zapier. No existing file is changed. This touches only new directories, so it doesn't conflict with the pilot or batch 2 PRs.topic_identifierlatestevent_typelatesteventlatestnull(discriminator unknown)latesttypeWebhooklatesteventlatesteventTypelatesttype(composed:type+.+event)latesttypelatesteventlatesteventTypelatestevent1.0event.typelatestclasslatesttypelatesttypelatesteventlatestlatesttypelatestevent_typeHow: I copied
samples-doc/providers/<p>/from webhook-registrymain(7e60bde1) intoproviders/, the same way #65 to #67 did. Every sample file is byte-identical to the export. Twoindex.jsonfiles differ, and those changes are listed first below. Every othertopic_identifierequals the registry row's discriminator path. I checked each one with a port ofresolveTopicagainst every sample. Each file carriessource.type: vendor-documentationand its page URL, and each version is markedsourced_via: docs.yarn compilepasses (155 → 174 providers, compiled topic count equals file count for all 19).yarn testpasses 14/14.Worth a look:
1.0/, notlatest/. All 17 bodies carry"api_version": "1.0", which RevenueCat documents as the webhook format version. The README rule says a payloadapi_versionis the value (Quo'sapiVersionset the same precedent). The export usedlatest/, so the files were moved andlatest_versionand the provenance key were changed to match.topic_identifier. The export wrote a prose string ("key-presence: the Update field besides update_id"). The resolver would have read that as a key name. The registry row haspath: nullbecause the update type is the name of the one key besidesupdate_id, which no README form can express. I found no precedent for key-presence, so it's left out. Each file'stopicstill names it.type. The event istype+.+event(courier_location.updated), so a capture would be named by family only (courier_location). This follows the sentry precedent (sentry: add 16 doc-sourced samples #62). The composed rows in batch 2 (modern-treasury, stytch, webex) were exported asnull, following the strava and facebook precedent. Both are established here; it's a one-line change if you want them aligned.eventresolves none of its 8 samples.eventis the email family's carrier, but the published examples are 7 SMS bodies (msg_status) and 1 coupon body (event_name). The README's list form (["event", "msg_status", "event_name"]) would resolve all 8. I kept the registry's primary path, as the export did.typeresolves 3 of 7. The GraphQL-shaped family (payment, payment.confirmation, refund, disbursement) has no type field at all, per the registry row.BACKUP(the EventType enum), but the body'sclassisBackup, so a capture would be namedBackup.null: the registry discriminator isunknown, following the deepgram convention.Update (9b838ee): nash's
topic_identifieris nownull. Withtype, a capture was named by family only (courier_location.updatedbecamecourier_location), so different events would overwrite each other. brevo's is now["event", "msg_status", "event_name"], the registry's email, SMS and coupon carriers. It resolves 8/8 samples (was 0/8). The nash and brevo notes under "Worth a look" are superseded.yarn compilepasses andyarn testpasses 14/14.Generated by Claude Code