##// END OF EJS Templates
logging: include more info in action logging...
logging: include more info in action logging These messages are very frequent and thus shows what is going on on the server - and it is nice to have some extra information to give some context for other messages that might be logged.

File last commit:

r2451:402a96fc beta
r3135:e7ba6928 beta
Show More
utils.py
97 lines | 2.9 KiB | text/x-python | PythonLexer
Added vcs testsuite for better integration tests + added fetching...
r2451 """
Utilities for tests only. These are not or should not be used normally -
functions here are crafted as we don't want to use ``vcs`` to verify tests.
"""
import os
import re
import sys
from subprocess import Popen
class VCSTestError(Exception):
pass
def run_command(cmd, args):
"""
Runs command on the system with given ``args``.
"""
command = ' '.join((cmd, args))
p = Popen(command, shell=True)
status = os.waitpid(p.pid, 0)[1]
return status
def eprint(msg):
"""
Prints given ``msg`` into sys.stderr as nose test runner hides all output
from sys.stdout by default and if we want to pipe stream somewhere we don't
need those verbose messages anyway.
Appends line break.
"""
sys.stderr.write(msg)
sys.stderr.write('\n')
class SCMFetcher(object):
def __init__(self, alias, test_repo_path, remote_repo, clone_cmd):
"""
:param clone_cmd: command which would clone remote repository; pass
only first bits - remote path and destination would be appended
using ``remote_repo`` and ``test_repo_path``
"""
self.alias = alias
self.test_repo_path = test_repo_path
self.remote_repo = remote_repo
self.clone_cmd = clone_cmd
def setup(self):
if not os.path.isdir(self.test_repo_path):
self.fetch_repo()
def fetch_repo(self):
"""
Tries to fetch repository from remote path.
"""
remote = self.remote_repo
eprint("Fetching repository %s into %s" % (remote, self.test_repo_path))
run_command(self.clone_cmd, '%s %s' % (remote, self.test_repo_path))
def get_normalized_path(path):
"""
If given path exists, new path would be generated and returned. Otherwise
same whats given is returned. Assumes that there would be no more than
10000 same named files.
"""
if os.path.exists(path):
dir, basename = os.path.split(path)
splitted_name = basename.split('.')
if len(splitted_name) > 1:
ext = splitted_name[-1]
else:
ext = None
name = '.'.join(splitted_name[:-1])
matcher = re.compile(r'^.*-(\d{5})$')
start = 0
m = matcher.match(name)
if not m:
# Haven't append number yet so return first
newname = '%s-00000' % name
newpath = os.path.join(dir, newname)
if ext:
newpath = '.'.join((newpath, ext))
return get_normalized_path(newpath)
else:
start = int(m.group(1)[-5:]) + 1
for x in xrange(start, 10000):
newname = name[:-5] + str(x).rjust(5, '0')
newpath = os.path.join(dir, newname)
if ext:
newpath = '.'.join((newpath, ext))
if not os.path.exists(newpath):
return newpath
raise VCSTestError("Couldn't compute new path for %s" % path)
return path