"sudo port -v selfupdate" fails with error code 25

A selfupdate seemed to be successful, but failed with the following error code: Congratulations, you have successfully installed the MacPorts system. To get the Portfiles and update the system, add /opt/local/bin to your PATH and run:

How to handle Case-sensitive files?

Related to tickets #22119 (supposedly fixed) and #22046 (still open), it seems as though qt4 4.6.3 and 4.7.0b1 both still have 2 sets of files that are different only in caps (content of all 4 is the same): {{{ qt4-mac/include/phonon/Phonon qt4-mac/include/phonon/phonon }}} and (not surprisin

Re: How to handle Case-sensitive files?

On 2010-6-24 03:22 , Michael Dickens wrote: > Related to tickets #22119 (supposedly fixed) and #22046 (still open), it > seems as though qt4 4.6.3 and 4.7.0b1 both still have 2 sets of files > that are different only in caps (content of all 4 is the same): > > {{{ > qt4-mac/include/phonon/Phonon

Re: How to handle Case-sensitive files?

> This is a known problem with the flat registry, > . Try switching to sqlite. Ah; in my search for "case sensitive" I didn't go back that far (just to around ticket 2 or so, which I thought would be far enough) :) Thank you for pointing that out. I -ho

Re: [68957] trunk/base/ChangeLog

On Jun 18, 2010, at 15:04, Jeremy Lavergne wrote: >> The sha256 checksum was broken? > > Yes, the function didn't exist (second half of the patch)! > > At first I figured it was just that we hadn't applied it yet to debug > message. When i looked further, running `port -d checksum` without the

Re: [68957] trunk/base/ChangeLog

On Wed, Jun 23, 2010 at 07:15:47PM -0500, Ryan Schmidt wrote: [snip] > For another, I'm unsure we really need sha256 checksums in there. It's > already complete overkill that we're putting three different checksums; using > four verges on crazy. The only reason we put more than one checksum at a

Re: [68957] trunk/base/ChangeLog

On 06/23/2010 05:21 PM, Eric Hall wrote: On Wed, Jun 23, 2010 at 07:15:47PM -0500, Ryan Schmidt wrote: [snip] For another, I'm unsure we really need sha256 checksums in there. It's already complete overkill that we're putting three different checksums; using four verges on crazy. The only rea

Re: [68957] trunk/base/ChangeLog

> My point was there was nothing broken about sha256 in MacPorts, and your > patch did not fix anything. Rather, it added a new functionality: it added a > sha256 line to the debug output from the checksum phase when the checksums > don't match. When I ran port checksum on something with sha256

Re: [68957] trunk/base/ChangeLog

On 2010-6-24 10:15 , Ryan Schmidt wrote: > > On Jun 18, 2010, at 15:04, Jeremy Lavergne wrote: > >>> The sha256 checksum was broken? >> >> Yes, the function didn't exist (second half of the patch)! >> >> At first I figured it was just that we hadn't applied it yet to debug >> message. When i loo

Re: [68957] trunk/base/ChangeLog

On Jun 23, 2010, at 19:26, Jeremy Lavergne wrote: >> My point was there was nothing broken about sha256 in MacPorts, and your >> patch did not fix anything. Rather, it added a new functionality: it added a >> sha256 line to the debug output from the checksum phase when the checksums >> don't m

Re: How to handle Case-sensitive files?

On Jun 23, 2010, at 17:55, Michael Dickens wrote: >> This is a known problem with the flat registry, >> . Try switching to sqlite. > > Ah; in my search for "case sensitive" I didn't go back that far (just to > around ticket 2 or so, which I thought woul

Re: [68957] trunk/base/ChangeLog

On Jun 23, 2010, at 19:21, Eric Hall wrote: > On Wed, Jun 23, 2010 at 07:15:47PM -0500, Ryan Schmidt wrote: > >> For another, I'm unsure we really need sha256 checksums in there. It's >> already complete overkill that we're putting three different checksums; >> using four verges on crazy. The o

Re: How to handle Case-sensitive files?

On Wed, 23 Jun 2010 19:47 -0500, "Ryan Schmidt" wrote: > > Is it possible to add a check into the code: if the file system is flat > > & then if a file of the same name has already been added (for this port > > being activated, or some other port), and if for this port, then print > > out a warnin