-
Accommodate for bug in Chrome 31+ that causes
assetsChanged
to always return true. (#278)Andrew Volozhanin + Nick Reed
-
Change URL after fetching page.
Marek Labos
-
Fix compatibility with
link_to :back
.Marek Labos
-
Send correct referer after asset change detected.
Marek Labos
-
Add the
page:before-change
event, triggered when a Turbolinks-enabled link is clicked. Can be used to cancel the Turbolinks process.Nick Reed
-
Add Turbolinks.pagesCached() to the public API for getting and setting the size of the page cache.
Nick Reed
-
Handle 5xx responses
Marek Labos
-
Add the ability to not execute scripts on turbolinks page loads by specifying
data-turbolinks-eval=false
on the<script>
tag. For example:<script type="text/javascript" data-turbolinks-eval=false>
Mario Visic
-
Workaround for WebKit history state bug with regards to the handling of 4xx responses.
Yasuharu Ozaki
-
Extracted
fetchReplacement()
onload response-handling logic out intovalidateResponse()
.Nick Reed
-
Changed response header name from
X-XHR-Current-Location
toX-XHR-Redirected-To
. The header will only be sent if there has been a redirect. Fixes compatibility issue when usingredirect_to
with options such as anchors or a trailing slash by storing the redirect location in a session variable and then using that value to set the response header.Yasuharu Ozaki
-
Escape URLs when checking for cross-origin redirects.
Nick Reed
-
Improve performance of
constrainPageCacheTo
,executeScriptTags
, andremoveNoscriptTags
by not gathering and returning the results of the loop comprehensions.Tim Ruffles
-
Change page without Turbolinks when Turbolinks.visit is called in Chrome on iOS
Frank Showalter
-
Maintain the latest CSRF authenticity token in the
<meta name="csrf-token">
head tag, if it exists.Nick Reed
-
Added Turbolinks::XDomainBlocker module with after_filter to detect cross-domain redirects, returning 403 Forbidden so that the client will reissue the request without Turbolinks. (XSS Protection)
Mala
-
Remove hash when checking for non-HTML links. (XSS Protection)
Nick Reed + Mala
-
Check Content-Type response header. Fall back to non-Turbolinks request unless the header is either
text/html
,application/xhtml+xml
, orapplication/xml
. (XSS Protection)Nick Reed + Mala
-
Add a
page:receive
event, triggered the moment the ajax request returns, before any processing is done.Ben Weintraub
-
Explicitly set
useCapture
flag to default (false) on addEventListener and removeEventListener calls. Removes errors on older browsers where the flag was required.Matthieu Aussaguel
-
Copy
<noscript>
tag list by slicing so thatremoveNoscriptTags
works with multiple<noscript>
tags.Lion Vollnhals*
-
Copy
<script>
tag list by slicing so thatexecuteScriptTags
works in situations where a script removes itself from the DOM.Nick Reed
-
Add link to Turbolinks Compatibility project to README.
Nick Reed
-
Reflect X-XHR-Referer in
request.referer
Nick Reed
-
Add
createDocumentUsingDOM
method to avoid DOMParser exceptions on certain browsers.Nick Reed
-
Add jquery.turbolinks note to README
Ry Walker
-
Abort XHR on popstate / Maintain history when aborting XHR
Nick Reed
-
Disable Turbolinks on Chrome for iOS
David Estes
-
Disable Turbolinks after non-GET requests. Adds a cookie named
request_method
.Nick Reed
-
Abort XHR request if the user clicks another link before it finishes.
David Estes
-
Fall back to using
document.location.href = url
when there is an error so the applicationCache can be used in offline mode.R. Potter
-
Remove hash from XHR url to fix IE 10 bug.
Nick Reed
-
Scroll to anchor link id if link has an anchor tag.
Yasuharu Ozaki
-
Allow Turbolinks to function in non-Rails environments.
Yasuharu Ozaki
-
Optimize script tag execution by using script injection instead of
window.eval
. Deprecates the use of thedata-turbolinks-evaluated
attribute.Nick Reed + John-David Dalton
-
Delay existing asset check until it's time to compare them to the fetched page's assets.
Nick Reed
-
Only track assets that have a
data-turbolinks-track
attribute.Yasuharu Ozaki
-
Fix issue where anchors are being dropped from the URL when changing pages
Nick Reed
-
Update
extractLink
to safeguard against links that are removed from the DOM when clicked.Gleb Mazovetskiy
-
Improve
executeScriptTags
by handling case wheresrc
attribute is empty and usingwindow.eval
instead ofeval
.itzki
-
Add compatibility section to README
DHH
-
Add support for executing external script tags in the body
Nick Reed
-
Add note to README about dynamically added scripts
Manuel Meurer
-
Prevent issue with ActionController::Live by moving
set_xhr_current_location
from an an after_filter to a before_filter.Kentaro Kuribayashi
-
Only execute script tags on XHR request load, not from history fetch.
David Estes
-
Use
document.location.reload()
instead ofdocument.location.href = url
if the assets have changed.Yasuharu Ozaki
-
Remove the
samePageLink
method.Yasuharu Ozaki
-
Change the
pageCache
from an array to a hash to fix back button handling when window.history.length is > 0 on page load.Steven Bristol
-
Handle case where a node, after being clicked, is removed from the DOM before
extractLink
can climb it's tree.Steven Bristol
-
Ignore links with a
target
attribute.Steven Bristol
-
Use timestamp to initialize history state position.
Yasuharo Ozaki
-
Detect additional asset changes.
David Estes
-
Add
coffee-rails
as a dependency.Rafael Mendonça França
-
Remember assets as soon as possible to prevent dynamic scripts from being checked and added.
David Estes
-
Remove
cloneNode
since there is no need to reset the node's events.David Estes
-
Improve
createDocument
to only determine correct parser once.David Estes
-
Only execute script tags that contain Javascript.
Nick Reed
-
Issue a full page load if the assets change.
DHH
-
Detect asset changes.
David Estes