~bzr-pqm/bzr/bzr.dev

Viewing changes to bzrlib/trace.py

Committer: Canonical.com Patch Queue Manager
Date: 2010-01-07 08:57:14 UTC
mfrom: (4634.118.1 2.0.4-trace-failure)
mto: (4634.128.1 2.0) (4949.1.1 integration) (4953.1.1 jam-integration) (4957.1.1 jam-integration) (4964.1.1 integration) (4967.1.1 jam-integration) (4981.1.2 2.1.0rc1-dev) (4999.3.1 apport) (4797.17.1 alias-loc-docs-337834-2.1) (5051.4.1 376388-dot-bazaar-ownership) (5045.1.1 369501-version) (5346.4.1 cleanup) (5174.1.1 jam-integration) (5074.5.1 fix-523746-dev)
mto: This revision was merged to the branch mainline in revision 4951.
Revision ID: pqm@pqm.ubuntu.com-20100107085714-bfuh2lmk855sjexc

(jam) Give a clear error when we fail to open ~/.bzr.log,
instead of 'no handlers for "bzr"' (#503886).

added added

removed removed

237

bzr_log_file.write("bug reports to https://bugs.launchpad.net/bzr/+filebug\n\n")

238

return bzr_log_file

239

except IOError, e:

240

warning("failed to open trace file: %s" % (e))

240

# If we are failing to open the log, then most likely logging has not

241

# been set up yet. So we just write to stderr rather than using

242

# 'warning()'. If we using warning(), users get the unhelpful 'no

243

# handlers registered for "bzr"' when something goes wrong on the

244

# server. (bug #503886)

245

sys.stderr.write("failed to open trace file: %s\n" % (e,))

241

246

# TODO: What should happen if we fail to open the trace file? Maybe the

242

247

# objects should be pointed at /dev/null or the equivalent? Currently

243

248

# returns None which will cause failures later.