SShortSingh.
Back to feed

Figma-to-Penpot test keeps objects, text and components but shifts text bounds

0
·13 views

A November 2026 test moved a small Figma fixture of ten objects into Penpot via exporter archive and cloud import. The horizontal auto-layout reversed in Penpot's representation, yet visible order and the 16 px gap were preserved. Native Inter text remained editable, with editing and component-fill propagation confirmed in the test file. Text bounds shifted slightly, so the migration was not pixel-identical. A PNG preview export via the MCP interface returned a server timeout, although the archive import itself succeeded.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Bareship enforces its React Native architecture through ESLint and dependency-cruiser rules

Bareship records its architecture as lint errors rather than in a wiki, so violations are caught automatically. ESLint enforces that imports flow in one direction across five layers and that features never import one another, while dependency-cruiser rechecks the full import graph in CI. Additional rules target specific past bugs, such as hard-coded user-facing text, left/right styles that do not mirror in right-to-left languages, and calls the React Compiler caches. Each error message explains the fix and links to the decision record behind the rule, letting developers or AI agents correct code without asking a colleague. The article is Part 3 of the Building Bareship series.

0
ProgrammingDEV Community ·

EU deforestation rules push coffee traceability down to plot-level data

The EU Deforestation Regulation requires importers to file statements with the geolocation of every plot their coffee came from, starting 30 December 2026 for larger operators and 30 June 2027 for micro and small ones. Plot location data is increasingly available, but linking it to the lots that factories, mills and warehouses handle is where systems tend to fail. The author proposes modelling plots as first-class database entities, enforcing geometry rules such as point precision and polygons for plots over 4 hectares, and recording lot composition as explicit edges at the time of pooling or bulking. Production periods are to be derived from delivery dates and stored with each lot. A recursive query over these edges can then trace a graded lot back to every upstream delivery and plot.

0
ProgrammingDEV Community ·

How to self-host a Framer site with Docker on your own server

A guide explains that Framer sites cannot be self-hosted directly since Framer does not export files, but a converted project can be run as a standard Node.js app. The recommended approach is to convert the published site into a code project, build it into a Docker image, run the container on a Linux server, and place a reverse proxy such as Caddy or nginx in front for HTTPS. Next.js is the most common target, with TanStack Start, Astro and React Router following similar steps using different start commands. The guide includes a sample Dockerfile, notes on host binding and Node.js version requirements, and advice on serving assets from the user's own server. It also states that the tool used is independent and unaffiliated with Framer, and that conversions should only be done for sites the user owns or has permission to convert.

0
ProgrammingDEV Community ·

Privacy-First Calculator Estimates the Time and Money Cost of Religious Belief

Senad Dizdarević, a journalist and founder of Evidence-Based Strong Atheism, built a web tool to quantify the cumulative cost of religious participation. The Cost of Religious Belief Calculator runs entirely in the user's browser, so no personal data is collected, stored, or sent over a network. It estimates direct financial contributions, time commitments, and the opportunity cost of those resources over several years. The creator also says it attempts to capture emotional and cognitive toll, though that dimension is harder to measure. The app is hosted at god-doesntexist.com and is presented as a free resource for people questioning their religious commitments.