jdupes 1.31.2 released, and I hate websites
If it’s not already painfully obvious (two years of no update posts plus a glitched SSL cert didn’t tip you off?) I’m very bad about not updating websites. I am seriously considering dumping WordPress for all of my websites and scratching out plain old boring static HTML. I’m also considering rolling jdupes.com up and redirecting it to jodybruchon.com/software at some point. Anyway…
After version 1.28.0, I managed to do some hash database code cleanups that contained an inverted logic mistake which basically halted all hash database speedups. The database would still be maintained by the program but the bit that goes “is the file path we’re examining already in the database?” had reverse logic which meant it would always return “nope!” so the whole point of the feature has been dead for over two years…and perhaps because it’s only a performance enhancement rather than a behavioral change, it took that long for anyone to actually notice! The release of 1.31.2 fixes the problem. There may be some performance improvements beyond the hashdb feature as well.
Bonus: I went to the trouble of adding Windows 7+ Explorer taskbar progress indication to jdupes. If you run jdupes, your command prompt taskbar icon will reflect the duplicate scanning progress. It’s a small thing, but it’ll probably make some nerd somewhere that runs long dedupe runs on Windows very happy. A development note from this: if you don’t want to unnecessarily inflate your Windows binary by 80-90 KiB by linking against libuuid just to get the few GUIDs required for progress indication, define INITGUID and #include <initguid.h> on MSYS2 MinGW-w64 and enjoy your slightly leaner final binary.
I still hate websites. Ironically, I would probably update more often if I didn’t have to log into WordPress and wait forever for my shared hosting server to slowly process all that PHP. Stay tuned.
One final breadcrumb…a pure C Win32 mail client project I’ve only just barely begun: Spite Mail.