Picturefill 3.0.3

JavaScript module sizes and how they're ludicrous is a hot topic on the internet these days, eh? Well, we here at Picturefill are not ones to miss out on a good trend. Thanks to @gpittarelli we've reduced the installed size of Picturefill by about 98%.

For those of you concerned that this means 98% of your responsive images won't be polyfilled, fear not. We're only being polite to your disk space and not installing a full copy of the examples directory (and various other small fried fish, by comparison) each time you install Picturefill. These examples are, of course, available on the wider interwebs, should you need them.

Official Release Notes ™:

• add .npmignore file (via #674)

3.0.2

• documented polyfill plugins
• force FF to re-select image candidate on window.resize (including srcset only)
• proper sizes detection for Safari 9.x

Picturefill 3.0.1

• Fixes a bug where smaller images would sometimes be downloaded after larger ones were already cached.

Picturefill 3.0.0

• Spec-accurate parsers for srcset and sizes
• Close matching of native behavior in terms of source selection
• A number of fixes for first-wave implementation issues, such as https://bugzilla.mozilla.org/show_bug.cgi?id=1135812
• A significantly more robust test suite

3.0.0 Release Candidate 1

Barring the discovery of any major issues, this is a preview of the release version of Picturefill 3.0.0.

Help us test, and let us know how it goes!

Picturefill 3.0.0 Beta 1

Picturefill 3.0 boasts a deadly accurate set of parsers, and matches a number of native behaviors that Picturefill 2.x didn’t—but also meant a near-complete rewrite.

That means we need testers.

Give the Picturefill 3.0.0 beta a try, and let us know if you run into any issues!

2.3.1

This release fixes a critical issue with Picturefill’s currentSrc support test. This issue caused errors in both the Spartan and WebKit nightlies.

Picturefill 2.3.0

Changes from Picturefill 2.3 Beta:

Spec Matching

• Invalid sizes values now fall back to 100vw default (f379528)

Bugs

• Fixes a potential race condition when using type (3fba68d)
• Prevents a fatal error in IE8 when assigning an undefined width (d1f5a1f)
• Prevent IE10/11 from setting an inherent width of zero on SVGs (742c66c)
• Fixes an error that prevented Picturefill from being handled as a named AMD module (a5859fb)

Picturefill 2.3.0 Beta

Along with assorted bugfixes and performance improvements, the following major issues have been addressed with this release:

Added support detection for the following image types: jpxr, jpg2000, and apng.
Image format support can now be handled using extensions. SVG and WebP remain a part of the Picturefill core, so as not to break existing implementations.

Support for intrinsic image dimensions
Picturefill now matches native behavior in terms of an image’s inherent size: when served at “2x,” the image is displayed at half of its source’s inherent size. Note that this can and must be overrideable via CSS.

Picturefill is now a named AMD module, and uses AMD/CommonJS module factory also in native supporting browsers.

Picturefill has a team now
Everyone join me in welcoming the team responsible for the lion’s share of work that went into this release, and many more to come.

Did someone just whisper “Picturefill 3?”
I thought I just heard someone whisper “Picturefill 3.” Weird.

Picturefill 2.2.0

Downloads, documentation, and examples:
http://picturefill.responsiveimages.org

Assorted bugfixes, improvements to compatibility in AMD environments, minor improvements to performance and spec compliance.