Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

webkit_settings_set_enable_frame_flattening() deprecation #720

Closed
mcatanzaro opened this issue Aug 31, 2022 · 9 comments · Fixed by #747
Closed

webkit_settings_set_enable_frame_flattening() deprecation #720

mcatanzaro opened this issue Aug 31, 2022 · 9 comments · Fixed by #747

Comments

@mcatanzaro
Copy link

Hi, this is a heads-up that the webkit_settings_set_enable_frame_flattening() and the WebKitSettings:enable-frame-flattening property will break in WebKitGTK 2.40. You'll need to figure out how to stop using them soon. If your app breaks with this setting disabled, then you'll want to ensure updated versions of your application reach environments that you care about prior to the release of WebKitGTK 2.40.

For more information: https://bugs.webkit.org/show_bug.cgi?id=242883

Sorry for the trouble. :/

@stephancb
Copy link
Contributor

This seems to finally have bitten me (Fedora 37, webkit2gtk4.0-devel, astroid - v0.16-2-g7c2022f0).

In thread view the overflowing parts of expanded emails are hidden (i.e. not viewable). A workaround is to copy part.scss to ~/.config/astroid/ui/ as described here. Then comment the hiding of overflowing parts like

body {
...
  /* overflow-y: hidden; */
...
}

Restart astroid. Now overflowing emails are scrollable for viewing.

@kirschner
Copy link

I was also effected by this. Before seeing this report, I filled a bug report for the debian package and also found a temporary work around by downgrading libwebkit.

Thanks @stephancb for the temporary fix and to @mcatanzaro for the heads-up here.

@bertogg
Copy link

bertogg commented May 17, 2023

Does that fix work fine? If so, wouldn't it make sense to create a pull request with it?

@bitmeise
Copy link

It actually just promotes Astroid from unusable to barely usable. The view port is still very narrow.
At least on my machine.

@bertogg
Copy link

bertogg commented May 17, 2023

I see, is there anyone with knowledge on the Astroid code that can provide a better solution?

@thomas2p
Copy link

thomas2p commented May 30, 2023

@bertogg @bitmeise it works better if you also overwrite the height of the iframe containing the email body. In thread-view.css look for the iframe selector and add this rule to it: height: 70vh (Adjust the value to your liking)

Example:

iframe {
  border: none;
  width: 100%;
  height: 70vh;
  text-align: left;
  float: left;
  display: block;
}

Hoping someone will eventually find a proper fix for this :)

@ibuclaw
Copy link
Contributor

ibuclaw commented Jun 4, 2023

AFAICT, the content of the iframe is set here.

void AstroidExtension::set_iframe_src (ustring mid, ustring cid, ustring body) {
LOG (debug) << "set iframe src: " << mid << ", " << cid;
WebKitDOMDocument * d = webkit_web_page_get_dom_document (page);
GError *err;
WebKitDOMHTMLElement * body_container = WEBKIT_DOM_HTML_ELEMENT(webkit_dom_document_get_element_by_id (d, cid.c_str ()));
WebKitDOMHTMLElement * iframe =
DomUtils::select (WEBKIT_DOM_NODE(body_container), ".body_iframe");
/* by using srcdoc we avoid creating any requests that would have to be
* allowed on the main GUI thread. even if we run this function async there
* might be other sync calls to the webextension that cause blocking since
* most webextension funcs need to run on extension GUI thread in order to
* manipulate DOM tree */
/* according to: http://w3c.github.io/html/semantics-embedded-content.html#element-attrdef-iframe-src
* we need to escape quotation marks and amperands. it seems that by using
* this call webkit does this for us. this is critical since otherwise the
* content could break out of the iframe. */
/* it would probably be possible to mess up the style, but it should only affect the current frame content. this would anyway be possible. */
webkit_dom_element_set_attribute (WEBKIT_DOM_ELEMENT (iframe), "srcdoc",
ustring::compose (
"<STYLE>%1</STYLE>%2",
part_css,
body ).c_str (),
(err = NULL, &err));
g_object_unref (iframe);
g_object_unref (body_container);
g_object_unref (d);
}

Based from changes in other projects - looking at Evolution specifically - the height (scrollHeight?) of the message body needs to be used to dynamically set the iframe height (plus padding).

Reading the GTK docs to get sense of how things are pieced together, it does not look good that pretty much all the C++ API has been deprecated in favour of writing everything in Javascript though. :-)

https://webkitgtk.org/reference/webkitdomgtk/2.31.91/WebKitDOMHTMLIFrameElement.html#webkit-dom-html-iframe-element-set-height

ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option woruld be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
probably have to be done in anger at some point in the future anyway.

I gave up going down the pass of manipulating the iframe after setting
`srcdoc`, and instead replaced the iframe with a div instead. CSS has
been fixed up and theme version bumped as it would be a breaking change
for all existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
probably have to be done in anger at some point in the future anyway.

I gave up going down the pass of manipulating the iframe after setting
`srcdoc`, and instead replaced the iframe with a div instead. CSS has
been fixed up and theme version bumped as it would be a breaking change
for all existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the pass of manipulating the iframe after setting
`srcdoc`, and instead replaced the iframe with a div instead. CSS has
been fixed up and theme version bumped as it would be a breaking change
for all existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the path of manipulating the iframe after setting
`srcdoc`, and instead replaced the iframe with a div instead. CSS has
been fixed up and theme version bumped as it would be a breaking change
for all existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the path of manipulating the iframe after setting
`srcdoc`, and replaced the iframe with a div instead. CSS has been fixed
up and theme version bumped as it would be a breaking change for all
existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
ibuclaw added a commit to ibuclaw/astroid that referenced this issue Jun 6, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the path of manipulating the iframe after setting
`srcdoc`, and replaced the iframe with a div instead. CSS has been fixed
up and theme version bumped as it would be a breaking change for all
existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
Cynerd added a commit to Cynerd/myconfigs that referenced this issue Aug 3, 2023
@jorsn
Copy link
Member

jorsn commented Aug 20, 2023

In principle, this is not really difficult to fix, if (a) one uses the scrollHeight (maybe from some other DOM library if webkit's is deprecated?) and (b) there is one function which is always executed when the view is updated. However, I didn't find that one function, and I think there are different ones depending on whether the content is HTML or plain or some other MIME. At least, my fix worked only for html messages, not for an alternative part.

Does anybody know all places at which one has to add the relevant code?

@rothn
Copy link
Contributor

rothn commented Sep 8, 2023

Rebasing onto https://github.com/ibuclaw/astroid/tree/remove_iframes fixes the breakage for me

rothn pushed a commit to rothn/astroid that referenced this issue Sep 9, 2023
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the path of manipulating the iframe after setting
`srcdoc`, and replaced the iframe with a div instead. CSS has been fixed
up and theme version bumped as it would be a breaking change for all
existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
jorsn added a commit to jorsn/astroid that referenced this issue Sep 16, 2023
yzhs pushed a commit to yzhs/astroid that referenced this issue Mar 30, 2024
As of WebKitGTK 2.40, enable-frame-flattening is no longer supported,
the property does nothing, resulting in all messages in the thread
viewer being cut off at around 150px, making all long messages
unreadable.

I initially had a look into seeing whether the iframe height could be
set programmatically with `webkit_dom_html_iframe_element_set_height`
from the content of the iframe via `webkit_dom_element_get_scroll_height`.
However nothing worked at page load time of an email thread, the content
scroll height remained at 150 throughout the entire
`ThreadView_on_load_changed` pass (because iframes are rendered lazily
after the main window thread has finished?  Can't think of any other
reason why this was observed). The scroll height was only found to
update after the thread had finished loading, and triggering a hide/show
toggle on each message.

Another option would be to use the JavaScriptCore API, and have all the
logic to manipulate the height of the iframe in JS. Two possible
problems with this: Firstly `enable-javascript` is explicitly disabled
in Astroid's WebKit settings; second, potentially all of the
`webextension` code may need to be rewritten in JS - given the huge
amount of deprecation warnings when building Astroid, this likely will
have to be done in anger at some point in the future anyway.

I gave up going down the path of manipulating the iframe after setting
`srcdoc`, and replaced the iframe with a div instead. CSS has been fixed
up and theme version bumped as it would be a breaking change for all
existing users that override the theme of the UI.

I can't see any concernable difference between before and after when
viewing both text/html and text/plain messages, though it is difficult
to gauge as my starting point is broken messages.

The `AstroidExtension::reload_images` code path is the least tested of
all the changes. The minimal check I did to see whether the logic is
still fine was to send myself an email with a CID attachment, and
observe it render correctly after pressing C-i in thread view.

As far as this change is concerned, this is one way to fix astroidmail#720 that
suits my use of Astroid. I'm not convinced that it is the correct way to
go about it though, as I have no reason to dismiss the rationale for
using iframes in the first place - see comments delete by this patch
that make reference to style issues, deadlocks and security concerns by
not having the content contained to its own iframe.
yzhs pushed a commit to yzhs/astroid that referenced this issue Mar 30, 2024
jorsn added a commit to jorsn/astroid that referenced this issue May 28, 2024
- fixes astroidmail#720
- alternative to removing iframes as in astroidmail#740 or astroidmail#732
  or the semi-working astroidmail#743

This does not allow the document (message content) to execute
javascript, as can be verified with the following email,
which displays "no scripts" by default, but "scripts active" when js is
active.

```
Date: Tue, 1 January 1970 00:00:00 +0000
From:
Subject:
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="=-TUjG03Ah9OcLtWE56Ucm"

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<div id=3D"jscontent">
  no scripts
</div>

<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html>
<html xmlns=3D"http://www.w3.org/1999/xhtml" lang xml:lang>
<head>
  <meta charset=3D"utf-8" />
  <meta name=3D"generator" content=3D"pandoc" />
  <meta name=3D"viewport" content=3D"width=3Ddevice-width, initial-scale=3D=
1.0, user-scalable=3Dyes" />
  <style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
span.underline{text-decoration: underline;}
div.column{display: inline-block; vertical-align: top; width: 50%;}
</style>
</head>
<body>
<div id=3D"jscontent">
<p>no scripts</p>
</div>
<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>
</body>
</html>

--=-TUjG03Ah9OcLtWE56Ucm--
```
jorsn added a commit to jorsn/astroid that referenced this issue Jun 3, 2024
- fixes astroidmail#720
- alternative to removing iframes as in astroidmail#740 or astroidmail#732
  or the semi-working astroidmail#743

This does not allow the document (message content) to execute
javascript, as can be verified with the following email,
which displays "no scripts" by default, but "scripts active" when js is
active.

```
Date: Tue, 1 January 1970 00:00:00 +0000
From:
Subject:
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="=-TUjG03Ah9OcLtWE56Ucm"

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<div id=3D"jscontent">
  no scripts
</div>

<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html>
<html xmlns=3D"http://www.w3.org/1999/xhtml" lang xml:lang>
<head>
  <meta charset=3D"utf-8" />
  <meta name=3D"generator" content=3D"pandoc" />
  <meta name=3D"viewport" content=3D"width=3Ddevice-width, initial-scale=3D=
1.0, user-scalable=3Dyes" />
  <style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
span.underline{text-decoration: underline;}
div.column{display: inline-block; vertical-align: top; width: 50%;}
</style>
</head>
<body>
<div id=3D"jscontent">
<p>no scripts</p>
</div>
<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>
</body>
</html>

--=-TUjG03Ah9OcLtWE56Ucm--
```
jorsn added a commit to jorsn/astroid that referenced this issue Jun 4, 2024
- fixes astroidmail#720
- alternative to removing iframes as in astroidmail#740 or astroidmail#732
  or the semi-working astroidmail#743

This does not allow the document (message content) to execute
javascript, as can be verified with the following email,
which displays "no scripts" by default, but "scripts active" when js is
active.

```
Date: Tue, 1 January 1970 00:00:00 +0000
From:
Subject:
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="=-TUjG03Ah9OcLtWE56Ucm"

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<div id=3D"jscontent">
  no scripts
</div>

<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>

--=-TUjG03Ah9OcLtWE56Ucm
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html>
<html xmlns=3D"http://www.w3.org/1999/xhtml" lang xml:lang>
<head>
  <meta charset=3D"utf-8" />
  <meta name=3D"generator" content=3D"pandoc" />
  <meta name=3D"viewport" content=3D"width=3Ddevice-width, initial-scale=3D=
1.0, user-scalable=3Dyes" />
  <style>
code{white-space: pre-wrap;}
span.smallcaps{font-variant: small-caps;}
span.underline{text-decoration: underline;}
div.column{display: inline-block; vertical-align: top; width: 50%;}
</style>
</head>
<body>
<div id=3D"jscontent">
<p>no scripts</p>
</div>
<script>
  d =3D document.getElementById('jscontent');
  d.innerHTML =3D "scripts active";
</script>
</body>
</html>

--=-TUjG03Ah9OcLtWE56Ucm--
```
@jorsn jorsn closed this as completed in #747 Jun 4, 2024
Kleinjohann added a commit to Kleinjohann/dotfiles that referenced this issue Jul 17, 2024
need to figure out how to get the proper upstream fix to work
yzhs pushed a commit to yzhs/astroid that referenced this issue Sep 22, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment