Skip to content
This repository has been archived by the owner on Nov 22, 2024. It is now read-only.

Add the ability to store multipart/alternative messages #36

Closed

Conversation

settermjd
Copy link

@settermjd settermjd commented Nov 20, 2023

Q A
Documentation yes
Bugfix yes
BC Break no
New Feature no
RFC no
QA no

Description

This change adds the ability to properly parse messages where the parent message's content-type is set to multipart/alternative into multiple Part objects. Previously, this content-type header would be ignored, leaving alternative parts being rolled into one single Part object. See 7.2.3 The Multipart/alternative subtype for more information.

At the moment, I don't know if this will cause an effective BC break.

This change adds the ability to properly parse messages where the
content type is "multipart/alternative" into multiple Part objects.
Previously, the "multipart/alternative" content-type header would be
ignored, leaving alternative parts being rolled into one larger Part
object.

Signed-off-by: Matthew Setter <[email protected]>
This is a small change to correct some minor points in the related
documentation.

Signed-off-by: Matthew Setter <[email protected]>
Copy link
Member

@weierophinney weierophinney left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A couple minor change suggestions, but overall looks solid (though I might prefer we typehint the $parts property as Part[]).

Also... Run phpcbf and phpcs before your next push.

src/Part.php Outdated Show resolved Hide resolved
test/MessageTest.php Outdated Show resolved Hide resolved
settermjd and others added 6 commits November 24, 2023 09:12
I wasn't aware at the time that I created the previous commit with the
constant, of the existence of a constant with the same value. So this
commit removes my duplicate constant, replacing it with the existing
one.

Signed-off-by: Matthew Setter <[email protected]>
Co-authored-by: Matthew Weier O'Phinney <[email protected]>
Signed-off-by: Matthew Setter <[email protected]>
As per @MWOP's suggestion, this change properly typehints the protected
$parts property in Part to return an array of Part objects.

Signed-off-by: Matthew Setter <[email protected]>
Signed-off-by: Matthew Setter <[email protected]>
This change adds a series of assertions for attachments on
multipart/alternative messages, to help verify that they contain what
they're expected to contain. I felt that it was helpful to have
assertions for attachments (or non text/HTML content) as well, so that
more of the code could be properly tested.

Signed-off-by: Matthew Setter <[email protected]>
These were originally used because it was a fairly generic email that I
had at the time, while writing the tests. However, as per @MWOP's
suggestion, it should be more generic (and not preference a given,
commercial, email provider).

Signed-off-by: Matthew Setter <[email protected]>
@settermjd settermjd force-pushed the add-multipart-alternative-support branch from 9372edc to f67b0c0 Compare November 23, 2023 23:12
@settermjd
Copy link
Author

All changes made and checks pass, except for Autocloser.

This change simplistically sets a Part's filename property if the
content disposition header is set, it's an attachment, and the filename
property is set. I've deliberately avoided any considerations about
sanitising the filename, deciding, at this stage, to leave that to the
calling code up the line.

Signed-off-by: Matthew Setter <[email protected]>
@settermjd settermjd force-pushed the add-multipart-alternative-support branch from 4224580 to 93c2cd6 Compare November 29, 2023 06:47
Previously, the code only set a filename for a message part if the part
contained a content-disposition header where the disposition type was
set to "attachment" and the filename parameter was set. After re-reading
https://www.w3.org/Protocols/HTTP/Issues/content-disposition.txt, I see
that this was wrong, as the content-disposition header can have the
filename parameter set if the type is attachment or inline.

This commit corrects that earlier mistake.

Signed-off-by: Matthew Setter <[email protected]>
@gsteel
Copy link
Member

gsteel commented Nov 22, 2024

@gsteel gsteel closed this Nov 22, 2024
@gsteel gsteel added the Won't Fix This will not be worked on label Nov 22, 2024
@gsteel gsteel assigned gsteel and unassigned settermjd Nov 22, 2024
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Enhancement Won't Fix This will not be worked on
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants