Hi, On Fri, Aug 21, 2020 at 3:53 PM <tutilu...@tutanota.com> wrote:
> pgAdmin v4.25 has yet again broken something. After being harassed to > upgrade to the latest version, pgAdmin has stopped opening itself properly. > > It no longer opens in the correct Pale Moon profile. It now opens in the > *default* profile, which completely breaks everything. I've previously > tried to explain how important it is for pgAdmin to have its own dedicated > GUI/"webview", so I'm not going to go into details about why this is so > important here. It seems to fall on deaf ears either way. > > My `Browser Command` is: > > "C:\Program Files\Pale Moon\palemoon.exe" -p "pgAdmin" --no-remote "%URL%" > > In other words, identical to v4.24 and every version before that. Yet it > now suddenly opens in the default profile, apparently ignoring my command > entirely. I also tried restarting the entire machine to make sure it wasn't > some weird temporary glitch. > > If I run this in Windows, it opens the right profile, so it's not my > browser at fault: > > "C:\Program Files\Pale Moon\palemoon.exe" -p "pgAdmin" --no-remote > "about:config" > > I have tested pgAdmin 4 v4.25 on MacOS with Firefox browser and it is working fine. My command: "/Applications/Firefox.app/Contents/MacOS/firefox" "%URL%" -p testprofile Is there no quality assurance for this program? I have never encountered > any software in my life which breaks as frequently as this... which says a > lot. > > If I go to the notification icon of pgAdmin, right click it and click "New > pgAdmin window", it opens in the right profile, as it's supposed to. So > clearly it's *able* to do it, somehow, but only in this weird way. It would > mean many clicks each time I want to open pgAdmin and doesn't address the > actual problem. > > In desperation, I even tried to make a completely new Pale Moon shortcut: > > "C:\Program Files\Pale Moon\palemoon.exe" -p "pgAdmin" --no-remote " > http://127.0.0.1:49707/browser/"; > > But then it just says: > > > Unauthorized > > The server could not verify that you are authorized to access the URL > requested. You either supplied the wrong credentials (e.g. a bad password), > or your browser doesn't understand how to supply the credentials required. > > I have no idea how this is even technically possible, but v4.25 clearly > has broken *something*. > > If this sounds "rude", it's because I feel horribly insulted as a user > when this mission-critical software (because it *is* mission-critical) > keeps breaking in weird ways. I've plowed down countless hours of my life > just working around bugs and "features" added to pgAdmin 4. I feel like I'm > *fighting against* it rather than using a great tool. I'm almost at the > point where I'm going to make my own database manager, since pgAdmin 4 is > not just very slow and glitchy, but downright *broken* now. > > And how do I turn off that nag screen which constantly keeps pestering me > to "upgrade" to the new version, which is very likely to break everything? > I want to disable that message so I never have to see it and never have to > get another broken pgAdmin version on my system ever again. I would go > through this Hell at the most once a year or something. > > To disable the upgrade notification set UPGRADE_CHECK_ENABLED = False. Please refer https://www.pgadmin.org/docs/pgadmin4/4.25/config_py.html to overide the default settings. Also, you can log the bug/feature request @ https://redmine.postgresql.org/projects/pgadmin4 Thanks, Khushboo > It feels like I'm one some kind of "bleeding edge" branch rather than the > stable/production one. Please consider testing your software before > releasing it. > > I truly don't understand how you can keep doing this to your users. Do you > hate us? There are some really good ideas and neat features inside this > program, but it's hidden behind all these layers of madness which ruins the > entire experience and cripples the user when they need it the most. > > I could very well never even have found the work-around. Most users > probably would not have tried half of what I've tried to get this to work. >
