Re[2]: Redirectors.pm vs DecodeShortURLs.pm

"Kent Oyer" <[email protected]> Fri, 22 May 2026 20:06:50 +0000
Newsgroups gmane.mail.spam.spamassassin.devel
Message-ID <[email protected]>
------=_Part_125_473435295.1779480410
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

If we merge these plugins, we might lose the ability to detect whether a =
link is "shortener" or a "redirector". Specifically, __URL_SHORTENER will=
 not work like it does today. It will either fire for both shorteners AND=
 redirectors (preferred), or it won't fire at all. In reality, we'd proba=
bly have a new __URL_REDIRECTOR rule and __URL_SHORTENER would remain as =
an alias for backwards compatibility. =0APersonally I'm fine with this. I=
 think the line between shortener and redirector is blurry and mostly irr=
elevant. They both have legitimate uses and they're both abusable by spam=
mers. =0AMy current plan is to make the Redirectors plugin handle url_sho=
rtener/url_shortener_get directives instead of changing those directives =
to url_redirector/url_redirector_get. That way people can still use the D=
ecodeShortURLs plugin while it's being deprecated and we don't have to ma=
intain two lists. We could aim to deprecate it in 4.0.3 and remove it in =
4.1.x.=0AIf we really REALLY need to track shorteners separately, I still=
 suggest merging the plugins but setting a flag on URLs that we consider =
"shorteners" so eval rules can differentiate. =0AThanks=0AKent

------=_Part_125_473435295.1779480410
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE html><html><head><meta http-equiv=3D"Content-Type" content=3D"t=
ext/html; charset=3Dutf-8" /></head><body><div data-crea=3D"font-wrapper"=
 style=3D"font-family: Tahoma, sans-serif; font-size: 16px; direction: lt=
r"><div>If we merge these plugins, we might lose the ability to detect wh=
ether a link is "shortener" or a "redirector". Specifically,=C2=A0__URL_S=
HORTENER will not work like it does today. It will either fire for both s=
horteners AND redirectors (preferred), or it won't fire at all. In realit=
y, we'd probably have a new __URL_REDIRECTOR rule and __URL_SHORTENER wou=
ld remain as an alias for backwards compatibility.=C2=A0</div><div><br></=
div><div>Personally I'm fine with this. I think the line between shortene=
r and redirector is blurry and mostly irrelevant. They both have legitima=
te uses and they're both abusable by spammers.=C2=A0</div><div><br></div>=
<div>My current plan is to make the Redirectors plugin handle url_shorten=
er/url_shortener_get directives instead of changing those directives to u=
rl_redirector/url_redirector_get. That way people can still use the Decod=
eShortURLs plugin while it's being deprecated and we don't have to mainta=
in two lists. We could aim to deprecate it in 4.0.3 and remove it in 4.1.=
x.</div><div><br></div><div>If we really REALLY need to track shorteners =
separately, I still suggest merging the plugins but setting a flag on URL=
s that we consider "shorteners" so eval rules can differentiate.=C2=A0</d=
iv><div><br></div><div>Thanks</div><div>Kent</div><div><br></div><div><br=
></div></div></body></html>

------=_Part_125_473435295.1779480410--