[html5] r3250 - [e] (0) make some notes of v2 suggestions.

whatwg at whatwg.org whatwg at whatwg.org
Sun Jun 14 18:10:15 PDT 2009


Author: ianh
Date: 2009-06-14 18:10:13 -0700 (Sun, 14 Jun 2009)
New Revision: 3250

Modified:
   index
   source
Log:
[e] (0) make some notes of v2 suggestions.

Modified: index
===================================================================
--- index	2009-06-15 00:42:34 UTC (rev 3249)
+++ index	2009-06-15 01:10:13 UTC (rev 3250)
@@ -5188,6 +5188,10 @@
 
   <!-- IE does [Callable] with magic that calls item() or namedItem() as appropriate as if it were a regular [[Get]],
        so we might need to change this for compat. However, WebIDL doesn't support that for now. -->
+  <!-- heycam says we could just say something like "In the ECMAScript binding, host objects that implement interface
+       <code>HTMLCollection</code> have a [[Call]] method that that behaves as follows: 1. If called with 0 or more than 1
+       argument, throw a TypeError. 2. Let arg be the single argument passed. 3. Invoke [[Get]] on the object with ToString(argument)
+       as the property name. 4. Return Result(3)." (but we'd have to verify that that's right for 0 and >1 args) -->
   <pre class=idl>[Callable=<a href=#dom-htmlcollection-nameditem title=dom-HTMLCollection-namedItem>namedItem</a>]
 interface <dfn id=htmlcollection-0>HTMLCollection</dfn> {
   readonly attribute unsigned long <a href=#dom-htmlcollection-length title=dom-HTMLCollection-length>length</a>;
@@ -16431,6 +16435,23 @@
    </dd>
   </dl><p>An <code><a href=#the-img-element>img</a></code> element represents an image.</p>
 
+  <!-- v2 ideas for <img>:
+
+     * Maps sites would like to know which images are already cached,
+       so that they can use images that are vaguely suitable while
+       they wait for the most appropriate image to download.
+
+       Almost like lowsrc="", except that many images might be
+       appropriate.
+
+       Slight hitch: their images are at a different origin, and we
+       don't want to allow arbitrary cross-origin inspection (privacy
+       leak risk).
+
+     * See note at rel=noreferrer.
+
+  -->
+
   <p>The image given by the <dfn id=attr-img-src title=attr-img-src><code>src</code></dfn> attribute is the
   embedded content, and the value of the <dfn id=attr-img-alt title=attr-img-alt><code>alt</code></dfn> attribute is the
   <code><a href=#the-img-element>img</a></code> element's <a href=#fallback-content>fallback content</a>.</p>
@@ -17780,7 +17801,30 @@
     Object specification. <a href=#refsWINDOW>[WINDOW]</a></p> <!--
     XXX -->
    </dd>
-  </dl><p>The <code><a href=#the-iframe-element>iframe</a></code> element <a href=#represents>represents</a> a
+  </dl><!-- v2 doc="" idea (awaiting implementation experience with seamless="" first):
+
+       http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2008-May/014874.html
+
+       doc="" would take a string which would then be interpreted as
+       the source document markup of an HTML document, much like the
+       above; it would override src="" if it was present, allowing
+       src="" to be used for legacy UAs:
+
+          <iframe seamless sandbox="allow-scripts allow-forms" doc="
+            <!DOCTYPE HTML>
+            <title></title>
+            Welcome to my blog!
+            </sandbox>
+            <a href='#' onclick='alert(document.cookie)'>Click here</a>
+          "></iframe>
+
+       (There are things we can do to make this better, e.g. make the
+       <!DOCTYPE HMTL> and <title></title> bits implicit, maybe
+       introducing type="" to say whether it's HTML or XML instead of
+       only supporting HTML, maybe saying that if src="" and doc=""
+       are both specified they must have identical data, etc.)
+
+  --><p>The <code><a href=#the-iframe-element>iframe</a></code> element <a href=#represents>represents</a> a
   <a href=#nested-browsing-context>nested browsing context</a>.</p>
 
   <p>The <dfn id=attr-iframe-src title=attr-iframe-src><code>src</code></dfn> attribute
@@ -48548,6 +48592,17 @@
 
   <h3 id=offline><span class=secno>6.9 </span>Offline Web applications</h3>
 
+  <!-- v2 ideas for appcache:
+
+     * A way to limit what gets download when the user agent is
+       updating the application cache and it turns out the server has
+       changed EVERY page because every page has a dynamic "site last
+       modified" date on it.
+
+       http://groups.google.com/group/gears-users/browse_thread/thread/efbd808325df607a/c73adb34f9b63cf7?hl=en&q=whatwg#c73adb34f9b63cf7
+
+  -->
+
   <h4 id=introduction-4><span class=secno>6.9.1 </span>Introduction</h4>
 
   <p><em>This section is non-normative.</em></p>
@@ -52556,7 +52611,10 @@
 
   </div>
 
+  <!-- v2: Would be nice to apply this to other elements too,
+  e.g. letting <img> or CSS or <video> hide the referrer -->
 
+
   <h5 id=link-type-pingback><span class=secno>6.12.3.12 </span>Link type "<dfn title=rel-pingback><code>pingback</code></dfn>"</h5>
 
   <p>The <code title=rel-pingback><a href=#link-type-pingback>pingback</a></code> keyword may be

Modified: source
===================================================================
--- source	2009-06-15 00:42:34 UTC (rev 3249)
+++ source	2009-06-15 01:10:13 UTC (rev 3250)
@@ -4847,6 +4847,10 @@
 
   <!-- IE does [Callable] with magic that calls item() or namedItem() as appropriate as if it were a regular [[Get]],
        so we might need to change this for compat. However, WebIDL doesn't support that for now. -->
+  <!-- heycam says we could just say something like "In the ECMAScript binding, host objects that implement interface
+       <code>HTMLCollection</code> have a [[Call]] method that that behaves as follows: 1. If called with 0 or more than 1
+       argument, throw a TypeError. 2. Let arg be the single argument passed. 3. Invoke [[Get]] on the object with ToString(argument)
+       as the property name. 4. Return Result(3)." (but we'd have to verify that that's right for 0 and >1 args) -->
   <pre class="idl">[Callable=<span title="dom-HTMLCollection-namedItem">namedItem</span>]
 interface <dfn>HTMLCollection</dfn> {
   readonly attribute unsigned long <span title="dom-HTMLCollection-length">length</span>;
@@ -17450,6 +17454,23 @@
 
   <p>An <code>img</code> element represents an image.</p>
 
+  <!-- v2 ideas for <img>:
+
+     * Maps sites would like to know which images are already cached,
+       so that they can use images that are vaguely suitable while
+       they wait for the most appropriate image to download.
+
+       Almost like lowsrc="", except that many images might be
+       appropriate.
+
+       Slight hitch: their images are at a different origin, and we
+       don't want to allow arbitrary cross-origin inspection (privacy
+       leak risk).
+
+     * See note at rel=noreferrer.
+
+  -->
+
   <p>The image given by the <dfn
   title="attr-img-src"><code>src</code></dfn> attribute is the
   embedded content, and the value of the <dfn
@@ -18902,6 +18923,31 @@
    </dd>
   </dl>
 
+  <!-- v2 doc="" idea (awaiting implementation experience with seamless="" first):
+
+       http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2008-May/014874.html
+
+       doc="" would take a string which would then be interpreted as
+       the source document markup of an HTML document, much like the
+       above; it would override src="" if it was present, allowing
+       src="" to be used for legacy UAs:
+
+          <iframe seamless sandbox="allow-scripts allow-forms" doc="
+            <!DOCTYPE HTML>
+            <title></title>
+            Welcome to my blog!
+            </sandbox>
+            <a href='#' onclick='alert(document.cookie)'>Click here</a>
+          "></iframe>
+
+       (There are things we can do to make this better, e.g. make the
+       <!DOCTYPE HMTL> and <title></title> bits implicit, maybe
+       introducing type="" to say whether it's HTML or XML instead of
+       only supporting HTML, maybe saying that if src="" and doc=""
+       are both specified they must have identical data, etc.)
+
+  -->
+
   <p>The <code>iframe</code> element <span>represents</span> a
   <span>nested browsing context</span>.</p>
 
@@ -55296,6 +55342,17 @@
 
   <h3 id="offline">Offline Web applications</h3>
 
+  <!-- v2 ideas for appcache:
+
+     * A way to limit what gets download when the user agent is
+       updating the application cache and it turns out the server has
+       changed EVERY page because every page has a dynamic "site last
+       modified" date on it.
+
+       http://groups.google.com/group/gears-users/browse_thread/thread/efbd808325df607a/c73adb34f9b63cf7?hl=en&q=whatwg#c73adb34f9b63cf7
+
+  -->
+
   <h4>Introduction</h4>
 
   <p><em>This section is non-normative.</em></p>
@@ -61611,7 +61668,10 @@
 
   </div>
 
+  <!-- v2: Would be nice to apply this to other elements too,
+  e.g. letting <img> or CSS or <video> hide the referrer -->
 
+
   <h5>Link type "<dfn title="rel-pingback"><code>pingback</code></dfn>"</h5>
 
   <p>The <code title="rel-pingback">pingback</code> keyword may be




More information about the Commit-Watchers mailing list