<br>For some strange reason it looks like the following message ended-up in the bit bucket.<br>So here it is again in case you missed it.<br><br>Cheers,<br><br>Jean-Claude Beaudoin<br><br><br><div class="gmail_quote">---------- Forwarded message ----------<br>
From: <b class="gmail_sendername">Jean-Claude Beaudoin</b> <span dir="ltr"><<a href="mailto:jean.claude.beaudoin@gmail.com">jean.claude.beaudoin@gmail.com</a>></span><br>Date: Tue, Mar 10, 2009 at 5:30 AM<br>Subject: Bug in user-homedir-pathname under MS-Windows.<br>
To: ecls-list <<a href="mailto:ecls-list@lists.sourceforge.net">ecls-list@lists.sourceforge.net</a>><br><br><br><br>Hello,<br><br>Here is attached a patch against 8.12 to fix a small bug in function "user-homedir-pathname" when<br>
ran under MS-Windows. Basically, drive and path were inverted resulting in a nonsensical<br>
pathname.<br><br>In the patch I also set the conditional compilation such that the same behavior is produced<br>on MS-Windows that ECL be compiled under MS Visual C or under MingW32/gcc. I would<br>like to advocate that a typical MS-Windows user does not want to know by what tool his instance<br>
of ECL was produced and should not be surprised by what would be to him inexplicable differences<br>of behavior. On the top of it, if you wish to maintain "a la Unix" behavior you can always set<br>the HOME environment variable and it will have precedence over HOMEDRIVE HOMEPATH.<br>
<br>Cheers,<br><font color="#888888"><br>Jean-Claude Beaudoin<br><br><br>
</font></div><br>