Opened 10 years ago
Closed 10 years ago
#46545 closed defect (fixed)
py(27|34)-scipy @0.15.0 +atlas +gcc48 fails to build
| Reported by: | josephsacco | Owned by: | seanfarley (Sean Farley) |
|---|---|---|---|
| Priority: | Normal | Milestone: | |
| Component: | ports | Version: | 2.3.3 |
| Keywords: | Cc: | ||
| Port: | py-scipy |
Description
I have run into dependency problem. I have installed:
py27-numpy @1.9.1_2+atlas+gcc48 (active)
py34-numpy @1.9.1_2+atlas+gcc48 (active)
atlas @3.10.2_1+gcc48 (active)
Below is the last section from the log file.
-Joseph
...
DEBUG: Executing variant gcc48 provides gcc48 Error: py-numpy: Variant gcc48 conflicts with gfortran DEBUG: Error evaluating variants
while executing
"error "Error evaluating variants""
(procedure "mportopen" line 50) invoked from within
"mportopen $porturl [array get interp_options] [array get variations]" Error: Unable to open port: Error evaluating variants DEBUG: upgrade py27-numpy failed
while executing
"macports::_upgrade_mport_deps $mport $target"
(procedure "mportexec" line 39) invoked from within
"mportexec $workername $target"
Attachments (1)
Change History (4)
Changed 10 years ago by josephsacco
comment:1 Changed 10 years ago by josephsacco
Poking around in port1.0/group/compilers-1.0.tcl, it appears that gfortran-mp-4.9 is being requested, even though the build is using gcc48. Why is that?
-Joseph
comment:2 Changed 10 years ago by seanfarley (Sean Farley)
| Owner: | changed from macports-tickets@… to sean@… |
|---|---|
| Status: | new → assigned |
Yep, that's my fault
comment:3 Changed 10 years ago by seanfarley (Sean Farley)
| Resolution: | → fixed |
|---|---|
| Status: | assigned → closed |
Should be fixed in r131526.
Log file