frg: code.binaryoutcast.com/projects/aur…e2a2d3829f7bffda9eb6a33410fa97923f8

;)

Thanks again for that.

frg_Away: code.binaryoutcast.com/projects/aur…e2a2d3829f7bffda9eb6a33410fa97923f8

thanks again for that

NewTobinParadigm np I usually try to fix what I broke :)

I am trying to keep orginal author as much as possible when possible (and I remember to lol)

so it WAS your fault lol

i only got two things to do before tuesday's Interlink's Platform Refresh release code.binaryoutcast.com/projects/aur…ntral/issues?state=open&milestone=1

but after that I can get back on track with my orginal goals

which was updating Interlink bits its self.. Communicator Preferences, maybe even untanging the mess the Thunderbird Anarchy Era introduced

renaming what are obviously mozilla js modules to jsm .. removing bits i half-ass disabled 3 years ago

all kinds of good fun eh frg_Away

6 weeks to de-sabotage the codebase done over 6 months basically

NewTobinParadigm The 2.53 code has working TB code if you need other backports. But not sure how easy they apply to the "older" codebase.

frg_Away: i did take in some backports provided by someone else over the years

might be interesting to take a peak.. ONE in particular is a priority for users tho

that one

it isn't entirely trival and my cpp skills are nearly non-existant.. oh i can apply and test but create new code or adapt newer code back is exceedingly difficult

though it doesn't look AS intemidating now since I did this patch: code.binaryoutcast.com/projects/aur…967f06de28cc8e5a21a63dfa5b61329dd29

this won't be applicable for you at your level since it all works as it is supposed to.. but at mine it is the best solution to balance the intent of the security/privacy bug and what is required to work internally

Mozilla's patch only allowed the file: protocol access to only moz-icon: .. Moonchild unsuccessfully did what was removed and that disallowed moz-icon:file:// from web but not the moz-icon: its self.. and it still left about:feeds unable to use it.. so he didn't fully patch the intent of the mozilla sec bug and broke internal shit which had issues before anyway

Hi

Rainer_Bielefeld: your name is familar

i feel like we have spoken before

anyway i think with the slow death of rss and the incompatibility with current mozilla tech that the surface for this kind of fingerprint data harvesting is very small.. besides, an extension's about page or w/e can access the filesystem anyway and steal a much more complete picture of what is installed or registered as file types vs a website's feed breaking sandbox and querying moz-icon:file:// urls

Currently I'm using installation of unofficial (by wg9s) De SeaMonkey 2.53.12 beta 1 Mozilla/5.0 (NT 6.1; Win64; x64; rv:68.0) Gecko/20100101 Firefox/68.0 Build 20220413141527 (Default Classic Theme, newly created User Profile) on German WIN7 64bit.

as well as not allowing the protocol at ALL to the web its self

But I haven't been asked to update to 2.53.12. Are 2.53.12 beta 1 and 2.53.12 identical?

Rainer_Bielefeld: but my onoficial builds are what is expected to be the next official release.

does WG9s run an aus service?

NewTobinParadigm: no i do not

we have enough issues trying to provide symbols builds for these in the aomunt of space we have to use on the mozilla servers.

WG9s: it isn't very difficult perhaps you should consider adding it so users aren't left hanging.. I could do some php scripting based off stuff i have done before if you tell me your requirements at some point

WG9s: yeah but you could use your own server

WG9s: why do you need to provide symbol builds why not just provide the symbols them selves for winders

as a seperate download

look on my own server you can do as i do to get a new build if there is one. I have a script for this

WG9s: i can't.. well i could now since I moved but you kinda ip banned me at one point lol

Really waht Ip did I ban I can unbann

it was a comcast ip either starting with 69 or 98

what i do is

sorry 65 and 98

mkdir -p ~/builds

cd ~/builds

touch -r *seamonkey253-linux-x86_64* test.$$

wget -N 'wg9s.com/cgi-bin/getbuild?seamonkey253-linux-x86_64' || exit 1

cd ~

if [ builds/*seamonkey253-linux-x86_64* -nt builds/test.$$ ]

then

rm -rf seamonkey wg9s_253

bzcat builds/*seamonkey253-linux-x86_64* | tar xvf -

mv seamonkey wg9s_253

fi

perl

rm -f builds/test.$$

and bash

heh

New I thought I unbanned all of the banned IP addresses on my last birtday in Februrary. Of course you might have pissed me off after that

this year?

well i moved at christmas

but things have changed since then

a lot of things

I'd still love to know why i can't get past nss 3.64 tho

and why random unrelated bits seem to error out for no rasin

still somewhat patched 3.64 is better than a barely patched nss 3.52 and deff vs an unpatched 3.whatever 52 shipped with

i can only gleem right now that 3.65 had some major refactoring or is it major refuckening

:P

oh well I am sure I will figure it out eventually

WG9s: Thx!

I don;t kno says I have benned no one.

hey frg_Away did you manage to import the mapi headers into your tree?

Yes

ah ok

i did too

a lot simpler for me than the changes in the orginal c-c bug

one of these days if i get good enough I will complete the incomplete xul linkage :P

if I can manage it and get more stuff out of libevil i could change how I package and build and redux the xulapp type of thing

that was the dream of xre and toolkit after all

MOZ_INCOMPLETE_EXTERNAL_LINKAGE .. incomplete for too long :(

frg_Away: do you guys have any search plugins where you get revenue from?

Yes ddg and startpage.

ah

another down.. code.binaryoutcast.com/projects/aur…6677036445e571c914e3b94e16747904835 one more change to go before the official blockers are all resolved

feel free to use it if you want lol

frg_Away ewong

This is already gone in latest 2.53. Was only needed for npapi.

well also eme and webrtc

well

gmp was hacky

but i swear the gmp provider relied on npapi plugin provider

frg_Away: the thing is that it is more work than it is worth to remove npapi from the codebase and there are a few edge cases that can be served or hell the slim, admittedly, hope of new plugins

if i killed npapi the possibility of a chrome based ietab type of thing would be dead

because that used a combo of npapi, xpcom binary and js to do the magic

i just didn't know you killed npapi in a 56-based codebase

EME and webrtc are still working in latest 2.53. Tested with a Firefox build. Problem is npapi needed gtk2 and some Linux distributiions already removed it. And it seems the removal has a positive effect on speed an memory usage.

frg_Away: they may have moved some shit over to gmp provider OR transfered some stuff over.. i dunno GMP is evil

also code.binaryoutcast.com/projects/aur…20dfce04f2984b7ca719c659cc3cb1af14b

I am ready to rock!