def list_outfiles_for_cleanup(self, dump_dir, dump_names=None):
files = Dump.list_outfiles_for_cleanup(self, dump_dir, dump_names)
files_to_return = []
if self.page_id_range:
# this file is for one page range only
if ',' in self.page_id_range:
(first_page_id, last_page_id) = self.page_id_range.split(',', 2)
first_page_id = int(first_page_id)
last_page_id = int(last_page_id)
else:
first_page_id = int(self.page_id_range)
last_page_id = None
# checkpoint files cover specific page ranges. for those,
# list only files within the given page range for cleanup
for fname in files:
if fname.is_checkpoint_file:
if (not first_page_id or
(fname.first_page_id and
(int(fname.first_page_id) >= first_page_id))):
if (not last_page_id or
(fname.last_page_id and
(int(fname.last_page_id) <= last_page_id))):
files_to_return.append(fname)
else:
files_to_return.append(fname)
else:
files_to_return = files
return files_to_return
def list_outfiles_for_cleanup(self, dump_dir):
"""
returns: list of DumpFilename
"""
dump_names = self.list_dumpnames()
dfnames = []
dfnames.extend(Dump.list_outfiles_for_cleanup(self, dump_dir, dump_names))
return dfnames
def get_tmp_files(self, dump_dir, dump_names=None):
"""
list temporary output files currently existing
returns:
list of DumpFilename
"""
dfnames = Dump.list_outfiles_for_cleanup(self, dump_dir, dump_names)
return [dfname for dfname in dfnames if dfname.is_temp_file]
def list_outfiles_for_cleanup(self, dump_dir, dump_names=None):
"""
list output files including checkpoint files currently existing
(from the dump run for the current wiki and date), in case
we have been requested to clean up before a retry
args:
DumpDir, list of dump names ("stub-meta-history", ...)
returns:
list of DumpFilename
"""
dfnames = Dump.list_outfiles_for_cleanup(self, dump_dir, dump_names)
dfnames_to_return = []
if self.jobinfo['pageid_range']:
# this file is for one page range only
if ',' in self.jobinfo['pageid_range']:
(first_page_id, last_page_id) = self.jobinfo['pageid_range'].split(',', 2)
first_page_id = int(first_page_id)
last_page_id = int(last_page_id)
else:
first_page_id = int(self.jobinfo['pageid_range'])
last_page_id = None
# checkpoint files cover specific page ranges. for those,
# list only files within the given page range for cleanup
for dfname in dfnames:
if dfname.is_checkpoint_file:
if (not first_page_id or
(dfname.first_page_id and
(int(dfname.first_page_id) >= first_page_id))):
if (not last_page_id or
(dfname.last_page_id and
(int(dfname.last_page_id) <= last_page_id))):
dfnames_to_return.append(dfname)
else:
dfnames_to_return.append(dfname)
else:
dfnames_to_return = dfnames
return dfnames_to_return
def list_outfiles_for_cleanup(self, dump_dir):
dump_names = self.list_dumpnames()
files = []
files.extend(Dump.list_outfiles_for_cleanup(self, dump_dir, dump_names))
return files
