Leaderboard

I've never understood why anyone would use getit over the main source repo for open source projects

An alternate package manager needs to be controlled and run by an organisation outside of Embarcadero which would give the community a little bit more leverage.

This is the perfect time and place to bring it up and get people behind it! I've started reading a little about it and it looks well thought out.

I wasn't going to bring it up in this thread.. but since you opened the door... I've been working on one for a while now https://delphi.dev - but it's a big project for one person and I have a business to run and a family. Docs - https://docs.delphi.dev/ Source https://github.com/DelphiPackageManager Installer - https://github.com/DelphiPackageManager/DPM/releases (codesigned by VSoft Technologies Pty Ltd). It is working well, but I have yet to implement loading design time packages - the reason for that is a lack of spare time. I did redo the IDE user interface a while back and that part is much easier to use than before (modelled on Jetbrain's Rider nuget support) - I use it daily. DPM supports Delphi XE2 - 12.0 - which should make upgrading to newer delphi versions easy - if the packages you use are available for the newer version of delphi, just open your project in that new version (with dpm installed) and dpm will automatically download and install the packages required for the project (and update search paths accordingly). Packages can be hosted by a package server, or a directory (ie network file share). The website is also a package server - that part is working well, however the ui needs a lot more work so that users can create api keys for pubishing packages. Right now the only packages there are ones I forked and published. The website is an asp.net 8.0 application (because I have a lot of experience with it) with a react front end, using a postgresql db - running on ubuntu server on a vm on one of my servers in Sydney - not ideal for latency for the rest of the world - it is behind cloudfare but api routes are excluded (cloudfare does help as a cdn for the images and javascript). If interest in/usage of the server actually took off I would move it to a cloud host in the US to improve latency (that would require some funding). The actual package files are on a cdn (bunny.net) and are regionally cached, so once someone near you has downloaded a package it is cached on their edge and its fast to download. Right now that's not costing me much but if usage ramps up it would. I went with bunny.net purely because of pricing - but apart from the stupid name it's actually a pretty good service. Recently someone (thanks @Geoffrey Smith ) contributed a tool for creating dpm package definition files (json) and we have been working on that together (it's installed by default now). Geoff created this tool to ease creating package definitions for some commercial libraires he is using. I would love to see more interest in the project and welcome contributions.

Convenience, laziness, curiosity. I've found myself browsing the GetIt packages (when it worked) and coming across something that looks interesting but I had not seen before. To try it out, it's really easy to just click Install and start using it to see if it's something I want to keep around. After it's installed, it might get "stuck" there out of laziness or because I only used it once or twice. To switch to the original repo means uninstalling the GetIt package, cloning the repo, learning how it's installed or how to access it... not difficult, just steps to take when there's time and it's going to become a steady part of my library.

Looks like a bit much. It depends on the number of design time packages. Anyway, it should not have changed unless you installed a bunch of new ones. There already exists a feature request to move that into a thread.

Is this file in a location that is mirrored with a cloud service (OneDrive, DropBox, etc)? If so, remove that location from the cloud service.

That field is not read by MemoView? Why convert it to String then again to TDateTime? In script you can use <MyTable."MyDateTimeField"> to read it as Variant that will keep TDateTime as it is (actually a Double).

Thanks, but interfaces are not even involved in the code in question.

You want to change the text value from a frxMemoView from date-time to display hour and minutes? Because mm is month... and HH is not documented. See documentation: https://docwiki.embarcadero.com/Libraries/Alexandria/en/System.SysUtils.FormatDateTime EDIT: Now I see that mm can show minuts if after hh.

Hi, Yes, it is half of the problem in your case, here you didn't mention who is the owner of the EXE and the owner of the parent folder, both these values plays role, The other half is either 1) System Protection has kicked in and sees that file (folder) as valuable and important, in this case there is copies and copies in the hidden/strict restore folder (System Volume Information) on the drive, you need to disable the protection and clear that folder then enable if you want it. 2) Windows Defender has kicked in and most likely sent the exe sample to Microsoft labs, i never ran this cloud protection (my main developing machine is still Windows 10), so don't know exactly what it does or how it behave, https://learn.microsoft.com/en-us/microsoft-365/security/defender-endpoint/cloud-protection-microsoft-antivirus-sample-submission?view=o365-worldwide Also you can run Process Monitor https://learn.microsoft.com/en-us/sysinternals/downloads/procmon remove/exclude all other processes and leave only BDS.exe in this case (to catch dcc32 operation), from there you should have the real and internal failure on API level this example of that same compiler error when the target exe is running and when was not

Let's not dive to deeply into the mudhole that is corporate practices. All we know, is that Something went south Cleanup takes time Practices can only improve It is outside our control We are short on patience

When this happens again try following what files are sought via ProcMon.

I don't buy it, I don't think this is just hardware failure. My conspiracy theory is that it started with hardware failure and then they realized that the backup didn't backup for 20 years. No matter what hardware failure you have, it doesn't take two weeks to get it back up and running.

And how could spikes in download even destroy a server (hardware)...

Actually, it hasn't. What was missing has happened before, i.e. it does not automatically download iOS support (weird). From the Xcode menu select Settings, then the Platforms tab. For the iOS 17.2 item, click Get: (which I have already done in this picture) Even weirder: while downloading it says "simulator". Once that completes, deployment should work.