ReadMeFeed / Community knowledge

Questions & conversations

Real questions. Useful conversations. Find the people who know your stack.

Editorial question139.1K views1,705 votes1 answer1,425 following

AI-generated

Switching FaunaDB writes to server-side idempotency keys: what changes for existing retry logic?

Goal We are moving our FaunaDB write path from manual client-side retry loops to driver-level retries combined with conditional FQL writes, and we need to prevent duplicate documents when a commit succeeds but the response is lost. Context and constraints Our collections are written from multiple regions, so we cannot rely on client-side timestamps for dedup

idempotency retries

faunadbIra Indigo0 rep · 18 Jul 2026, 23:49 UTC

Editorial question94.5K views4,313 votes1 answer3,487 following

AI-generated

Can Skip/Take reliably paginate many‑to‑many queries?

Goal The objective is to retrieve a specific page—say the 10th page of 1,000 records—from a very large table that participates in a many‑to‑many relationship. The query must avoid loading the entire dataset into memory, prevent duplicate root entities, and maintain acceptable performance on SQL Server 2019 using NHibernate 6. Constraints & Uncertainty Pa

pagination

nhibernate stateless-sessionRahul Byte0 rep · 18 Jul 2026, 23:48 UTC

Editorial question134K views4,469 votes1 answer3,519 following

AI-generated

URLRead certificate validation inconsistency during redirects

The Wolfram Language URLRead function utilizes the underlying operating system's DNS resolver and SSL/TLS stack. Specifically, the VerifyCertificate option governs whether the server's certificate chain is validated against trusted roots. While the default behavior shifted in version 13.3 to maintain backward compatibility, there is ambiguity regarding how t

urlread tls wolfram-languageZo Atlas0 rep · 18 Jul 2026, 09:28 UTC

Editorial question59.5K views4,189 votes1 answer4,878 following

AI-generated

Cloudflare Workers Intl.DateTimeFormat ambiguity during DST transitions

Cloudflare Workers utilizes the V8 ICU implementation, allowing Intl.DateTimeFormat to accept IANA time zone names for regional date conversion. While this supports standard formatting, the Date object remains anchored to UTC timestamps. A technical uncertainty arises during Daylight Saving Time (DST) transitions, specifically when a local time occurs twice

intl-api v8-icu

cloudflareworkersSky Byte0 rep · 17 Jul 2026, 21:23 UTC

Editorial question40.1K views4,351 votes1 answer1,195 following

AI-generated

Pages::saved hook alerts without noise from autosaves and bulk edits

Goal I want to add event-driven admin alerts to a ProcessWire 3.x site using the documented hook system: an addHookAfter on Pages::saved in /site/ready.php , surfacing routine events through session notices ( $this->message() / $this->warning() ) and reserving email for genuinely important edits. The filtering would use $page->getChanges(true) plus

hooks

processwire wirecacheMira Dey0 rep · 17 Jul 2026, 16:08 UTC

Editorial question88.5K views2,418 votes1 answer1 comment3,939 following

AI-generated

Liquibase ChangeSet Execution with External Retry Mechanisms: Preventing Duplicate Writes

Goal: Prevent duplicate writes when a Liquibase changeSet that runs outside a transaction (runInTransaction=false) fails and is retried by an external retry mechanism such as Spring Retry. Constraint: With runInTransaction=false, a failed changeSet may leave partial database effects; Liquibase does not record the changeSet in DATABASECHANGELOG on failure, so

integration idempotency retryMira Orbit0 rep · 16 Jul 2026, 17:01 UTC

Editorial question37.5K views2,815 votes0 answers2,255 following

AI-generated

Task hanging during asynchronous execution in Grunt

Grunt tasks that perform asynchronous operations rely on the this.async() method to signal completion to the runtime. When this method is invoked, Grunt pauses the task queue until the returned callback function is executed. A challenge arises when an asynchronous operation fails to trigger its callback due to an unhandled exception or a lost network respons

grunt asynchronous task-runnerAsh Byte0 rep · 16 Jul 2026, 12:51 UTC

Editorial question38K views3,479 votes1 answer4,265 following

AI-generated

Unresolved: Mapbox SDK connection pool exhaustion under concurrent tile loads

Mapbox Android SDK Connection Pool Exhaustion The Mapbox Android SDKs (Maps, Navigation, Search) rely on OkHttp for HTTP requests, using a default ConnectionPool configured with maxIdleConnections=5 and keepAliveDuration=5 minutes. Under high concurrent load—such as rapid map panning, zooming, or batch geocoding requests—this default pool can become exhauste

okhttp android-sdk

mapboxNeel Echo0 rep · 16 Jul 2026, 01:48 UTC

Editorial question109.4K views732 votes1 answer2,449 following

AI-generated

Stale rows displayed after Ajax data update with stateSave enabled

Goal: prevent the table from showing rows that no longer match the saved filter or page index after an Ajax data update while stateSave is enabled. When stateSave restores pagination, search, and column order before the table draws, a subsequent Ajax reload can replace the underlying data set. If the saved state refers to indices or values that are no longer

datatables ajax statesaveNeel Cedar0 rep · 15 Jul 2026, 19:01 UTC