bug#42844: 27.1; Emacs 27.1: SVG file not rendered as image

["Pierre Téchoueyres" <pierre.techoueyres@free.fr>]

Le mercredi 19 août 2020 à 19:33, phillip.lord@russet.org.uk a écrit :

> On 2020-08-19 18:39, Pierre Téchoueyres wrote:
>> Hello Eli,
>> I've also tested and followed your advice with the dependency
>> walker.
>> I've found the following missing ddl:
>> - libffi-7.dll
>> - libgio-2.0-0.dll
>> - libglib-2.0-0.dll
>> - libgmodule-2.0-0.dll
>> - libgobject-2.0-0.dll
>> - libpcre-1.dll
>> don't know why the libffi-7 is needed as there is an libffi-6 in the
>> directory. But with thoses files I can display SVG images inside emacs.
>> Pierre
>
>
> Pretty sure all of this stems from 61c2183a440. I dropped using
> pactree and used pacman directly to determine dependencies. I also
> exclude glib2 as a dependency hence the problem. If I include this, we
> go back to getting python and half the of msys2 included in the
> download bundle.
>
> I guess for Emacs-27 I should just include glib2 and be exclude so of
> its dependencies (gettext and python probably). The question is
> whether to do this for Emacs-27.1 or 27.2.
>

From my user side point of view: just add the missing dll in the archive
for 27.1 and let the dust lay on.

I suggest it  because, if like me, your company has and antivirus
software which block some actions on programs without good reputation
you're lock with officials distributions. Lukly Emacs is on the list
...

see https://whitelist.kaspersky.com/advisor

For example, at my work, writing to
c:/windows/system32/drivers/etc/hosts is forbiden for not known
applications like the ones you're compiling yourself (and trust me this
bother me quite a lot).

>
> For Emacs-28, I think I need a better dependency mechanism than
> looking at msys2, perhaps based around the shell script that was sent
> earlier, or an automated (or horror manual) use of dependency walker.

If could help you in testing, ask for it.