This simplifies and further optimizes text-width computation by computing the entire width table in advance, and serializing it in the style of QuickTextMeasurer (#1390). This entirely removes the need for PDFKit at runtime. This has the advantage of fixing #1305 – more generally: producing the same result everywhere – without having to deploy a copy of Verdana.
The lifting is delegated to these three libraries, which are housed in a monorepo: https://github.com/metabolize/anafanafo
I'd be happy to move it into the badges org if folks want to collaborate on maintaining them.
QuickTextMeasurer took kerning pairs into account, whereas this implementation does not. I was thinking kerning would be a necessary refinement, though this seems to work well enough.
I dropped in a binary-search package to traverse the data structure, in part to conserve space. This causes a moderate performance regression, though there is ample room for improving on that: https://github.com/badges/shields/pull/2311#issuecomment-439182704
* move gh-badges files out of /lib
As far as possible, this is just moving files
around and updating paths however there are 2
functional changes in this commit:
- remove use of lib/register-chai-plugins.spec
in badge-cli.spec.js
- remove use of starRating()
in text-measurer.spec.js
* update service tests that use colorscheme.json
* split package.json in two
* clean up import
* don't hard-code path
* start a changelog
* put a license file in the package dir
* re-organise documentation 📚
* don't pack test files
* remove favicon from Makefile
* give package its own test command
* link the docs better in README
* Basic process metrics
* Enable Prometheus by an environment variable
* Code formatting
* Documentation for Prometheus metrics
* Link from README to documentation of Prometheus
* Link from README to documentation of Prometheus
* Link from README to documentation of Prometheus
* Separate module for metrics + tests
* Metrics limited by IP
* Metrics are forbidded for all requets by default
* Code refactoring
* allowedIps passed as a string to PrometheusMetrics
* Handle missing config
* METRICS_PROMETHEUS_ALLOWED_IPS added to documentation
* Log info about enabled metrics
* Unused code removed
* package-lock.json updated
* prom-client updated to 11.1.2
* Code refactoring
* Do not read IP address from X-Forwarder-For header
This is a fairly simple addition of a Redis-backed TokenPersistence. When GithubConstellation is initialized, it will create a FsTokenPersistence or a RedisTokenPersistence based on configuration. Have added tests of the Redis backend as an integration test, and ensured the server starts up correctly when a `REDIS_URL` is configured.
Ref: #1848
Make a clear distinction between programmer errors ("internal errors") and runtime errors, and allow configuring the server to let the programmer errors bubble up in development and unit testing. This saves a huge amount of time because it generates ordinary stack traces when things go wrong. And, if these errors occur in production, we'll catch them, and display **shields | internal error** which is the equivalent of a 500 error.
Ref: #1379
This takes a naive approach to font-width computation, the most compute-intensive part of rendering badges.
1. Add the widths of the individual characters.
- These widths are measured on startup using PDFKit.
2. For each character pair, add a kerning adjustment
- The difference between the width of each character pair, and the sum of the characters' separate widths.
- These are computed for each character pair on startup using PDFKit.
3. For a string with characters outside the printable ASCII character set, fall back to PDFKit.
This branch averaged 0.041 ms in `makeBadge`, compared to 0.144 ms on master, a speedup of 73%. That was on a test of 10,000 consecutive requests (using the `benchmark-performance.sh` script, now checked in).
The speedup applies to badges containing exclusively printable ASCII characters. It wouldn't be as dramatic on non-ASCII text. Though, we could add some frequently used non-ASCII characters to the cached set.
- Periodically log github auth information
- Tokens are hashed which reduces the security risk inherent in the logs
- A consistent hash is used so tokens can be correlated across the three data structures and across the three servers
- Add an admin endpoint for github auth information
- Tokens are returned as-is to enable troubleshooting (e.g. comparing our reqRemaining to github’s)
- Support single-server testing and a local dev server (like Next) that is on a different port from the shields server
- Refactor config schema
With this change, the suggestions work locally in #1273.
Paul MelnikowandGitHub
chris48sandGitHub
Marcin MielnickiandGitHub
Thaddee Tyl