def process_batch_job(batch_job_file):
""" Processes a batch job description dictionary
@param batch_job_file: a fullpath to a batch job file
@type batch_job_file: string
@return: 1 if the process was successful, 0 if not
@rtype; int
"""
from invenio.legacy.bibdocfile.cli import cli_fix_marc
def upload_marcxml_file(marcxml):
""" Creates a temporary marcxml file and sends it to bibupload
"""
xml_filename = 'bibencode_'+ str(batch_job['recid']) + '_' + str(uuid.uuid4()) + '.xml'
xml_filename = os.path.join(invenio.config.CFG_TMPSHAREDDIR, xml_filename)
xml_file = file(xml_filename, 'w')
xml_file.write(marcxml)
xml_file.close()
targs = ['-c', xml_filename]
task_low_level_submission('bibupload', 'bibencode', *targs)
#---------#
# GENERAL #
#---------#
_task_write_message("----------- Handling Master -----------")
## Check the validity of the batch file here
batch_job = json_decode_file(batch_job_file)
## Sanitise batch description and raise errrors
batch_job = sanitise_batch_job(batch_job)
## Check if the record exists
# if record_exists(batch_job['recid']) < 1:
# raise Exception("Record not found")
recdoc = BibRecDocs(batch_job['recid'])
#--------------------#
# UPDATE FROM MASTER #
#--------------------#
## We want to add new stuff to the video's record, using the master as input
if getval(batch_job, 'update_from_master'):
found_master = False
bibdocs = recdoc.list_bibdocs()
for bibdoc in bibdocs:
bibdocfiles = bibdoc.list_all_files()
for bibdocfile in bibdocfiles:
comment = bibdocfile.get_comment()
description = bibdocfile.get_description()
subformat = bibdocfile.get_subformat()
m_comment = getval(batch_job, 'bibdoc_master_comment', comment)
m_description = getval(batch_job, 'bibdoc_master_description', description)
m_subformat = getval(batch_job, 'bibdoc_master_subformat', subformat)
if (comment == m_comment and
description == m_description and
subformat == m_subformat):
found_master = True
batch_job['input'] = bibdocfile.get_full_path()
## Get the aspect of the from the record
try:
## Assumes pbcore metadata mapping
batch_job['aspect'] = get_fieldvalues(124, CFG_BIBENCODE_ASPECT_RATIO_MARC_FIELD)[0]
except IndexError:
pass
break
if found_master:
break
if not found_master:
_task_write_message("Video master for record %d not found"
% batch_job['recid'])
task_update_progress("Video master for record %d not found"
% batch_job['recid'])
## Maybe send an email?
return 1
## Clean the job to do no upscaling etc
if getval(batch_job, 'assure_quality'):
batch_job = clean_job_for_quality(batch_job)
global _BATCH_STEPS
_BATCH_STEPS = len(batch_job['jobs'])
## Generate the docname from the input filename's name or given name
bibdoc_video_docname, bibdoc_video_extension = decompose_file(batch_job['input'])[1:]
if not bibdoc_video_extension or getval(batch_job, 'bibdoc_master_extension'):
bibdoc_video_extension = getval(batch_job, 'bibdoc_master_extension')
if getval(batch_job, 'bibdoc_master_docname'):
bibdoc_video_docname = getval(batch_job, 'bibdoc_master_docname')
write_message("Creating BibDoc for %s" % bibdoc_video_docname)
## If the bibdoc exists, receive it
if bibdoc_video_docname in recdoc.get_bibdoc_names():
bibdoc_video = recdoc.get_bibdoc(bibdoc_video_docname)
## Create a new bibdoc if it does not exist
else:
bibdoc_video = recdoc.add_bibdoc(docname=bibdoc_video_docname)
## Get the directory auf the newly created bibdoc to copy stuff there
bibdoc_video_directory = bibdoc_video.get_base_dir()
#--------#
# MASTER #
#--------#
if not getval(batch_job, 'update_from_master'):
if getval(batch_job, 'add_master'):
## Generate the right name for the master
## The master should be hidden first an then renamed
## when it is really available
## !!! FIX !!!
_task_write_message("Adding %s master to the BibDoc"
% bibdoc_video_docname)
master_format = compose_format(
bibdoc_video_extension,
getval(batch_job, 'bibdoc_master_subformat', 'master')
)
## If a file of the same format is there, something is wrong, remove it!
## it might be caused by a previous corrupted submission etc.
if bibdoc_video.format_already_exists_p(master_format):
bibdoc_video.delete_file(master_format, 1)
bibdoc_video.add_file_new_format(
batch_job['input'],
version=1,
description=getval(batch_job, 'bibdoc_master_description'),
comment=getval(batch_job, 'bibdoc_master_comment'),
docformat=master_format
)
#-----------#
# JOBS LOOP #
#-----------#
return_code = 1
global _BATCH_STEP
for job in batch_job['jobs']:
_task_write_message("----------- Job %s of %s -----------"
% (_BATCH_STEP, _BATCH_STEPS))
## Try to substitute docname with master docname
if getval(job, 'bibdoc_docname'):
job['bibdoc_docname'] = Template(job['bibdoc_docname']).safe_substitute({'bibdoc_master_docname': bibdoc_video_docname})
#-------------#
# TRANSCODING #
#-------------#
if job['mode'] == 'encode':
## Skip the job if assure_quality is not set and marked as fallback
if not getval(batch_job, 'assure_quality') and getval(job, 'fallback'):
continue
if getval(job, 'profile'):
profile = get_encoding_profile(job['profile'])
else:
profile = None
## We need an extension defined fot the video container
bibdoc_video_extension = getval(job, 'extension',
getval(profile, 'extension'))
if not bibdoc_video_extension:
raise Exception("No container/extension defined")
## Get the docname and subformat
bibdoc_video_subformat = getval(job, 'bibdoc_subformat')
bibdoc_slave_video_docname = getval(job, 'bibdoc_docname', bibdoc_video_docname)
## The subformat is incompatible with ffmpegs name convention
## We do the encoding without and rename it afterwards
bibdoc_video_fullpath = compose_file(
bibdoc_video_directory,
bibdoc_video_extension
)
_task_write_message("Transcoding %s to %s;%s" % (bibdoc_slave_video_docname,
bibdoc_video_extension,
bibdoc_video_subformat))
## We encode now directly into the bibdocs directory
encoding_result = encode_video(
input_file=batch_job['input'],
output_file=bibdoc_video_fullpath,
acodec=getval(job, 'audiocodec'),
vcodec=getval(job, 'videocodec'),
abitrate=getval(job, 'videobitrate'),
vbitrate=getval(job, 'audiobitrate'),
resolution=getval(job, 'resolution'),
passes=getval(job, 'passes', 1),
special=getval(job, 'special'),
specialfirst=getval(job, 'specialfirst'),
specialsecond=getval(job, 'specialsecond'),
metadata=getval(job, 'metadata'),
width=getval(job, 'width'),
height=getval(job, 'height'),
aspect=getval(batch_job, 'aspect'), # Aspect for every job
profile=getval(job, 'profile'),
update_fnc=_task_update_overall_status,
message_fnc=_task_write_message
)
return_code &= encoding_result
## only on success
if encoding_result:
## Rename it, adding the subformat
os.rename(bibdoc_video_fullpath,
compose_file(bibdoc_video_directory,
bibdoc_video_extension,
bibdoc_video_subformat,
1,
bibdoc_slave_video_docname)
)
#bibdoc_video._build_file_list()
bibdoc_video.touch()
bibdoc_video._sync_to_db()
bibdoc_video_format = compose_format(bibdoc_video_extension,
bibdoc_video_subformat)
if getval(job, 'bibdoc_comment'):
bibdoc_video.set_comment(getval(job, 'bibdoc_comment'),
bibdoc_video_format)
if getval(job, 'bibdoc_description'):
bibdoc_video.set_description(getval(job, 'bibdoc_description'),
bibdoc_video_format)
#------------#
# EXTRACTION #
#------------#
# if there are multiple extraction jobs, all the produced files
# with the same name will be in the same bibdoc! Make sure that
# you use different subformats or docname templates to avoid
# conflicts.
if job['mode'] == 'extract':
if getval(job, 'profile'):
profile = get_extract_profile(job['profile'])
else:
profile = {}
bibdoc_frame_subformat = getval(job, 'bibdoc_subformat')
_task_write_message("Extracting frames to temporary directory")
tmpdir = invenio.config.CFG_TMPDIR + "/" + str(uuid.uuid4())
os.mkdir(tmpdir)
#Move this to the batch description
bibdoc_frame_docname = getval(job, 'bibdoc_docname', bibdoc_video_docname)
tmpfname = (tmpdir + "/" + bibdoc_frame_docname + '.'
+ getval(profile, 'extension',
getval(job, 'extension', 'jpg')))
extraction_result = extract_frames(input_file=batch_job['input'],
output_file=tmpfname,
size=getval(job, 'size'),
positions=getval(job, 'positions'),
numberof=getval(job, 'numberof'),
width=getval(job, 'width'),
height=getval(job, 'height'),
aspect=getval(batch_job, 'aspect'),
profile=getval(job, 'profile'),
update_fnc=_task_update_overall_status,
)
return_code &= extraction_result
## only on success:
if extraction_result:
## for every filename in the directorys, create a bibdoc that contains
## all sizes of the frame from the two directories
files = os.listdir(tmpdir)
for filename in files:
## The docname was altered by BibEncode extract through substitution
## Retrieve it from the filename again
bibdoc_frame_docname, bibdoc_frame_extension = os.path.splitext(filename)
_task_write_message("Creating new bibdoc for %s" % bibdoc_frame_docname)
## If the bibdoc exists, receive it
if bibdoc_frame_docname in recdoc.get_bibdoc_names():
bibdoc_frame = recdoc.get_bibdoc(bibdoc_frame_docname)
## Create a new bibdoc if it does not exist
else:
bibdoc_frame = recdoc.add_bibdoc(docname=bibdoc_frame_docname)
## The filename including path from tmpdir
fname = os.path.join(tmpdir, filename)
bibdoc_frame_format = compose_format(bibdoc_frame_extension, bibdoc_frame_subformat)
## Same as with the master, if the format allready exists,
## override it, because something went wrong before
if bibdoc_frame.format_already_exists_p(bibdoc_frame_format):
bibdoc_frame.delete_file(bibdoc_frame_format, 1)
_task_write_message("Adding %s jpg;%s to BibDoc"
% (bibdoc_frame_docname,
getval(job, 'bibdoc_subformat')))
bibdoc_frame.add_file_new_format(
fname,
version=1,
description=getval(job, 'bibdoc_description'),
comment=getval(job, 'bibdoc_comment'),
docformat=bibdoc_frame_format)
## Remove the temporary folders
_task_write_message("Removing temporary directory")
shutil.rmtree(tmpdir)
_BATCH_STEP = _BATCH_STEP + 1
#-----------------#
# FIX BIBDOC/MARC #
#-----------------#
_task_write_message("----------- Handling MARCXML -----------")
## Fix the BibDoc for all the videos previously created
_task_write_message("Updating BibDoc of %s" % bibdoc_video_docname)
bibdoc_video._build_file_list()
## Fix the MARC
_task_write_message("Fixing MARC")
cli_fix_marc({}, [batch_job['recid']], False)
if getval(batch_job, 'collection'):
## Make the record visible by moving in from the collection
marcxml = ("<record><controlfield tag=\"001\">%d</controlfield>"
"<datafield tag=\"980\" ind1=\" \" ind2=\" \">"
"<subfield code=\"a\">%s</subfield></datafield></record>"
) % (batch_job['recid'], batch_job['collection'])
upload_marcxml_file(marcxml)
#---------------------#
# ADD MASTER METADATA #
#---------------------#
if getval(batch_job, 'add_master_metadata'):
_task_write_message("Adding master metadata")
pbcore = pbcore_metadata(input_file = getval(batch_job, 'input'),
pbcoreIdentifier = batch_job['recid'],
aspect_override = getval(batch_job, 'aspect'))
from invenio_formatter.engines.xslt import format
marcxml = format(pbcore, CFG_BIBENCODE_PBCORE_MARC_XSLT)
upload_marcxml_file(marcxml)
#------------------#
# ADD MARC SNIPPET #
#------------------#
if getval(batch_job, 'marc_snippet'):
marc_snippet = open(getval(batch_job, 'marc_snippet'))
marcxml = marc_snippet.read()
marc_snippet.close()
upload_marcxml_file(marcxml)
#--------------#
# DELETE INPUT #
#--------------#
if getval(batch_job, 'delete_input'):
_task_write_message("Deleting input file")
# only if successfull
if not return_code:
# only if input matches pattern
if getval(batch_job, 'delete_input_pattern', '') in getval(batch_job, 'input'):
try:
os.remove(getval(batch_job, 'input'))
except OSError:
pass
#--------------#
# NOTIFICATION #
#--------------#
## Send Notification emails on errors
if not return_code:
if getval(batch_job, 'notify_user'):
_notify_error_user(getval(batch_job, 'notify_user'),
getval(batch_job, 'submission_filename', batch_job['input']),
getval(batch_job, 'recid'),
getval(batch_job, 'submission_title', ""))
_task_write_message("Notify user because of an error")
if getval(batch_job, 'notify_admin'):
_task_write_message("Notify admin because of an error")
if type(getval(batch_job, 'notify_admin') == type(str()) ):
_notify_error_admin(batch_job,
getval(batch_job, 'notify_admin'))
else:
_notify_error_admin(batch_job)
else:
if getval(batch_job, 'notify_user'):
_task_write_message("Notify user because of success")
_notify_success_user(getval(batch_job, 'notify_user'),
getval(batch_job, 'submission_filename', batch_job['input']),
getval(batch_job, 'recid'),
getval(batch_job, 'submission_title', ""))
return 1
def main():
"""Core loop."""
check_running_process_user()
logfilename = '%s/fulltext_files_migration_kit-%s.log' % (CFG_LOGDIR, datetime.today().strftime('%Y%m%d%H%M%S'))
try:
logfile = open(logfilename, 'w')
except IOError as e:
print(wrap_text_in_a_box('NOTE: it\'s impossible to create the log:\n\n %s\n\nbecause of:\n\n %s\n\nPlease run this migration kit as the same user who runs Invenio (e.g. Apache)' % (logfilename, e), style='conclusion', break_long=False))
sys.exit(1)
bibdoc_bibdoc = retrieve_bibdoc_bibdoc()
print(wrap_text_in_a_box ("""This script migrate the filesystem structure used to store icons files to the new stricter structure.
This script must not be run during normal Invenio operations.
It is safe to run this script. No file will be deleted.
Anyway it is recommended to run a backup of the filesystem structure just in case.
A backup of the database tables involved will be automatically performed.""", style='important'))
if not bibdoc_bibdoc:
print(wrap_text_in_a_box("No need for migration", style='conclusion'))
return
print("%s icons will be migrated/fixed." % len(bibdoc_bibdoc))
wait_for_user()
print("Backing up database tables")
try:
if not backup_tables():
print(wrap_text_in_a_box("""It appears that is not the first time that you run this script.
Backup tables have been already created by a previous run.
In order for the script to go further they need to be removed.""", style='important'))
wait_for_user()
print("Backing up database tables (after dropping previous backup)", end=' ')
backup_tables(drop=True)
print("-> OK")
else:
print("-> OK")
except Exception as e:
print(wrap_text_in_a_box("Unexpected error while backing up tables. Please, do your checks: %s" % e, style='conclusion'))
sys.exit(1)
to_fix_marc = intbitset()
print("Created a complete log file into %s" % logfilename)
try:
try:
for id_bibdoc1, id_bibdoc2 in bibdoc_bibdoc:
try:
record_does_exist = True
recids = get_recid_from_docid(id_bibdoc1)
if not recids:
print("Skipping %s" % id_bibdoc1)
continue
for recid in recids:
if record_exists(recid[0]) > 0:
to_fix_marc.add(recid[0])
else:
record_does_exist = False
if not fix_bibdoc_bibdoc(id_bibdoc1, id_bibdoc2, logfile):
if record_does_exist:
raise StandardError("Error when correcting document ID %s" % id_bibdoc1)
except Exception as err:
print("ERROR: %s" % err, file=logfile)
print(wrap_text_in_a_box("DONE", style='conclusion'))
except:
logfile.close()
register_exception()
print(wrap_text_in_a_box(
title = "INTERRUPTED BECAUSE OF ERROR!",
body = """Please see the log file %s for what was the status prior to the error. Contact %s in case of problems, attaching the log.""" % (logfilename, CFG_SITE_SUPPORT_EMAIL),
style = 'conclusion'))
sys.exit(1)
finally:
print("Scheduling FIX-MARC to synchronize MARCXML for updated records.")
cli_fix_marc(options={}, explicit_recid_set=to_fix_marc)
def process_batch_job(batch_job_file):
""" Processes a batch job description dictionary
@param batch_job_file: a fullpath to a batch job file
@type batch_job_file: string
@return: 1 if the process was successful, 0 if not
@rtype; int
"""
from invenio.legacy.bibdocfile.cli import cli_fix_marc
def upload_marcxml_file(marcxml):
""" Creates a temporary marcxml file and sends it to bibupload
"""
xml_filename = 'bibencode_'+ str(batch_job['recid']) + '_' + str(uuid.uuid4()) + '.xml'
xml_filename = os.path.join(invenio.config.CFG_TMPSHAREDDIR, xml_filename)
xml_file = file(xml_filename, 'w')
xml_file.write(marcxml)
xml_file.close()
targs = ['-c', xml_filename]
task_low_level_submission('bibupload', 'bibencode', *targs)
#---------#
# GENERAL #
#---------#
_task_write_message("----------- Handling Master -----------")
## Check the validity of the batch file here
batch_job = json_decode_file(batch_job_file)
## Sanitise batch description and raise errrors
batch_job = sanitise_batch_job(batch_job)
## Check if the record exists
# if record_exists(batch_job['recid']) < 1:
# raise Exception("Record not found")
recdoc = BibRecDocs(batch_job['recid'])
#--------------------#
# UPDATE FROM MASTER #
#--------------------#
## We want to add new stuff to the video's record, using the master as input
if getval(batch_job, 'update_from_master'):
found_master = False
bibdocs = recdoc.list_bibdocs()
for bibdoc in bibdocs:
bibdocfiles = bibdoc.list_all_files()
for bibdocfile in bibdocfiles:
comment = bibdocfile.get_comment()
description = bibdocfile.get_description()
subformat = bibdocfile.get_subformat()
m_comment = getval(batch_job, 'bibdoc_master_comment', comment)
m_description = getval(batch_job, 'bibdoc_master_description', description)
m_subformat = getval(batch_job, 'bibdoc_master_subformat', subformat)
if (comment == m_comment and
description == m_description and
subformat == m_subformat):
found_master = True
batch_job['input'] = bibdocfile.get_full_path()
## Get the aspect of the from the record
try:
## Assumes pbcore metadata mapping
batch_job['aspect'] = get_fieldvalues(124, CFG_BIBENCODE_ASPECT_RATIO_MARC_FIELD)[0]
except IndexError:
pass
break
if found_master:
break
if not found_master:
_task_write_message("Video master for record %d not found"
% batch_job['recid'])
task_update_progress("Video master for record %d not found"
% batch_job['recid'])
## Maybe send an email?
return 1
## Clean the job to do no upscaling etc
if getval(batch_job, 'assure_quality'):
batch_job = clean_job_for_quality(batch_job)
global _BATCH_STEPS
_BATCH_STEPS = len(batch_job['jobs'])
## Generate the docname from the input filename's name or given name
bibdoc_video_docname, bibdoc_video_extension = decompose_file(batch_job['input'])[1:]
if not bibdoc_video_extension or getval(batch_job, 'bibdoc_master_extension'):
bibdoc_video_extension = getval(batch_job, 'bibdoc_master_extension')
if getval(batch_job, 'bibdoc_master_docname'):
bibdoc_video_docname = getval(batch_job, 'bibdoc_master_docname')
write_message("Creating BibDoc for %s" % bibdoc_video_docname)
## If the bibdoc exists, receive it
if bibdoc_video_docname in recdoc.get_bibdoc_names():
bibdoc_video = recdoc.get_bibdoc(bibdoc_video_docname)
## Create a new bibdoc if it does not exist
else:
bibdoc_video = recdoc.add_bibdoc(docname=bibdoc_video_docname)
## Get the directory auf the newly created bibdoc to copy stuff there
bibdoc_video_directory = bibdoc_video.get_base_dir()
#--------#
# MASTER #
#--------#
if not getval(batch_job, 'update_from_master'):
if getval(batch_job, 'add_master'):
## Generate the right name for the master
## The master should be hidden first an then renamed
## when it is really available
## !!! FIX !!!
_task_write_message("Adding %s master to the BibDoc"
% bibdoc_video_docname)
master_format = compose_format(
bibdoc_video_extension,
getval(batch_job, 'bibdoc_master_subformat', 'master')
)
## If a file of the same format is there, something is wrong, remove it!
## it might be caused by a previous corrupted submission etc.
if bibdoc_video.format_already_exists_p(master_format):
bibdoc_video.delete_file(master_format, 1)
bibdoc_video.add_file_new_format(
batch_job['input'],
version=1,
description=getval(batch_job, 'bibdoc_master_description'),
comment=getval(batch_job, 'bibdoc_master_comment'),
docformat=master_format
)
#-----------#
# JOBS LOOP #
#-----------#
return_code = 1
global _BATCH_STEP
for job in batch_job['jobs']:
_task_write_message("----------- Job %s of %s -----------"
% (_BATCH_STEP, _BATCH_STEPS))
## Try to substitute docname with master docname
if getval(job, 'bibdoc_docname'):
job['bibdoc_docname'] = Template(job['bibdoc_docname']).safe_substitute({'bibdoc_master_docname': bibdoc_video_docname})
#-------------#
# TRANSCODING #
#-------------#
if job['mode'] == 'encode':
## Skip the job if assure_quality is not set and marked as fallback
if not getval(batch_job, 'assure_quality') and getval(job, 'fallback'):
continue
if getval(job, 'profile'):
profile = get_encoding_profile(job['profile'])
else:
profile = None
## We need an extension defined fot the video container
bibdoc_video_extension = getval(job, 'extension',
getval(profile, 'extension'))
if not bibdoc_video_extension:
raise Exception("No container/extension defined")
## Get the docname and subformat
bibdoc_video_subformat = getval(job, 'bibdoc_subformat')
bibdoc_slave_video_docname = getval(job, 'bibdoc_docname', bibdoc_video_docname)
## The subformat is incompatible with ffmpegs name convention
## We do the encoding without and rename it afterwards
bibdoc_video_fullpath = compose_file(
bibdoc_video_directory,
bibdoc_video_extension
)
_task_write_message("Transcoding %s to %s;%s" % (bibdoc_slave_video_docname,
bibdoc_video_extension,
bibdoc_video_subformat))
## We encode now directly into the bibdocs directory
encoding_result = encode_video(
input_file=batch_job['input'],
output_file=bibdoc_video_fullpath,
acodec=getval(job, 'audiocodec'),
vcodec=getval(job, 'videocodec'),
abitrate=getval(job, 'videobitrate'),
vbitrate=getval(job, 'audiobitrate'),
resolution=getval(job, 'resolution'),
passes=getval(job, 'passes', 1),
special=getval(job, 'special'),
specialfirst=getval(job, 'specialfirst'),
specialsecond=getval(job, 'specialsecond'),
metadata=getval(job, 'metadata'),
width=getval(job, 'width'),
height=getval(job, 'height'),
aspect=getval(batch_job, 'aspect'), # Aspect for every job
profile=getval(job, 'profile'),
update_fnc=_task_update_overall_status,
message_fnc=_task_write_message
)
return_code &= encoding_result
## only on success
if encoding_result:
## Rename it, adding the subformat
os.rename(bibdoc_video_fullpath,
compose_file(bibdoc_video_directory,
bibdoc_video_extension,
bibdoc_video_subformat,
1,
bibdoc_slave_video_docname)
)
#bibdoc_video._build_file_list()
bibdoc_video.touch()
bibdoc_video._sync_to_db()
bibdoc_video_format = compose_format(bibdoc_video_extension,
bibdoc_video_subformat)
if getval(job, 'bibdoc_comment'):
bibdoc_video.set_comment(getval(job, 'bibdoc_comment'),
bibdoc_video_format)
if getval(job, 'bibdoc_description'):
bibdoc_video.set_description(getval(job, 'bibdoc_description'),
bibdoc_video_format)
#------------#
# EXTRACTION #
#------------#
# if there are multiple extraction jobs, all the produced files
# with the same name will be in the same bibdoc! Make sure that
# you use different subformats or docname templates to avoid
# conflicts.
if job['mode'] == 'extract':
if getval(job, 'profile'):
profile = get_extract_profile(job['profile'])
else:
profile = {}
bibdoc_frame_subformat = getval(job, 'bibdoc_subformat')
_task_write_message("Extracting frames to temporary directory")
tmpdir = invenio.config.CFG_TMPDIR + "/" + str(uuid.uuid4())
os.mkdir(tmpdir)
#Move this to the batch description
bibdoc_frame_docname = getval(job, 'bibdoc_docname', bibdoc_video_docname)
tmpfname = (tmpdir + "/" + bibdoc_frame_docname + '.'
+ getval(profile, 'extension',
getval(job, 'extension', 'jpg')))
extraction_result = extract_frames(input_file=batch_job['input'],
output_file=tmpfname,
size=getval(job, 'size'),
positions=getval(job, 'positions'),
numberof=getval(job, 'numberof'),
width=getval(job, 'width'),
height=getval(job, 'height'),
aspect=getval(batch_job, 'aspect'),
profile=getval(job, 'profile'),
update_fnc=_task_update_overall_status,
)
return_code &= extraction_result
## only on success:
if extraction_result:
## for every filename in the directorys, create a bibdoc that contains
## all sizes of the frame from the two directories
files = os.listdir(tmpdir)
for filename in files:
## The docname was altered by BibEncode extract through substitution
## Retrieve it from the filename again
bibdoc_frame_docname, bibdoc_frame_extension = os.path.splitext(filename)
_task_write_message("Creating new bibdoc for %s" % bibdoc_frame_docname)
## If the bibdoc exists, receive it
if bibdoc_frame_docname in recdoc.get_bibdoc_names():
bibdoc_frame = recdoc.get_bibdoc(bibdoc_frame_docname)
## Create a new bibdoc if it does not exist
else:
bibdoc_frame = recdoc.add_bibdoc(docname=bibdoc_frame_docname)
## The filename including path from tmpdir
fname = os.path.join(tmpdir, filename)
bibdoc_frame_format = compose_format(bibdoc_frame_extension, bibdoc_frame_subformat)
## Same as with the master, if the format allready exists,
## override it, because something went wrong before
if bibdoc_frame.format_already_exists_p(bibdoc_frame_format):
bibdoc_frame.delete_file(bibdoc_frame_format, 1)
_task_write_message("Adding %s jpg;%s to BibDoc"
% (bibdoc_frame_docname,
getval(job, 'bibdoc_subformat')))
bibdoc_frame.add_file_new_format(
fname,
version=1,
description=getval(job, 'bibdoc_description'),
comment=getval(job, 'bibdoc_comment'),
docformat=bibdoc_frame_format)
## Remove the temporary folders
_task_write_message("Removing temporary directory")
shutil.rmtree(tmpdir)
_BATCH_STEP = _BATCH_STEP + 1
#-----------------#
# FIX BIBDOC/MARC #
#-----------------#
_task_write_message("----------- Handling MARCXML -----------")
## Fix the BibDoc for all the videos previously created
_task_write_message("Updating BibDoc of %s" % bibdoc_video_docname)
bibdoc_video._build_file_list()
## Fix the MARC
_task_write_message("Fixing MARC")
cli_fix_marc({}, [batch_job['recid']], False)
if getval(batch_job, 'collection'):
## Make the record visible by moving in from the collection
marcxml = ("<record><controlfield tag=\"001\">%d</controlfield>"
"<datafield tag=\"980\" ind1=\" \" ind2=\" \">"
"<subfield code=\"a\">%s</subfield></datafield></record>"
) % (batch_job['recid'], batch_job['collection'])
upload_marcxml_file(marcxml)
#---------------------#
# ADD MASTER METADATA #
#---------------------#
if getval(batch_job, 'add_master_metadata'):
_task_write_message("Adding master metadata")
pbcore = pbcore_metadata(input_file = getval(batch_job, 'input'),
pbcoreIdentifier = batch_job['recid'],
aspect_override = getval(batch_job, 'aspect'))
from invenio.modules.formatter.engines.xslt import format
marcxml = format(pbcore, CFG_BIBENCODE_PBCORE_MARC_XSLT)
upload_marcxml_file(marcxml)
#------------------#
# ADD MARC SNIPPET #
#------------------#
if getval(batch_job, 'marc_snippet'):
marc_snippet = open(getval(batch_job, 'marc_snippet'))
marcxml = marc_snippet.read()
marc_snippet.close()
upload_marcxml_file(marcxml)
#--------------#
# DELETE INPUT #
#--------------#
if getval(batch_job, 'delete_input'):
_task_write_message("Deleting input file")
# only if successfull
if not return_code:
# only if input matches pattern
if getval(batch_job, 'delete_input_pattern', '') in getval(batch_job, 'input'):
try:
os.remove(getval(batch_job, 'input'))
except OSError:
pass
#--------------#
# NOTIFICATION #
#--------------#
## Send Notification emails on errors
if not return_code:
if getval(batch_job, 'notify_user'):
_notify_error_user(getval(batch_job, 'notify_user'),
getval(batch_job, 'submission_filename', batch_job['input']),
getval(batch_job, 'recid'),
getval(batch_job, 'submission_title', ""))
_task_write_message("Notify user because of an error")
if getval(batch_job, 'notify_admin'):
_task_write_message("Notify admin because of an error")
if type(getval(batch_job, 'notify_admin') == type(str()) ):
_notify_error_admin(batch_job,
getval(batch_job, 'notify_admin'))
else:
_notify_error_admin(batch_job)
else:
if getval(batch_job, 'notify_user'):
_task_write_message("Notify user because of success")
_notify_success_user(getval(batch_job, 'notify_user'),
getval(batch_job, 'submission_filename', batch_job['input']),
getval(batch_job, 'recid'),
getval(batch_job, 'submission_title', ""))
return 1
done, exceptions = create_icons_for_record(recid, icon_sizes,
icon_format_mappings,
docnames, add_default_icon,
inherit_moreinfo)
updated += done
if exceptions:
for ex in exceptions:
write_message(ex)
else:
write_message("Recid %s DONE." % recid)
task_update_progress("Done %d out of %d." % (i, len(recids)))
task_sleep_now_if_required(can_stop_too=True)
if updated:
cli_fix_marc(None, explicit_recid_set=recids, interactive=False)
return 1
def _create_icon(file_path, icon_size, docname, icon_format='gif', verbosity=9):
"""
Creates icon of given file.
Returns path to the icon. If creation fails, return None, and
register exception (send email to admin).
@param file_path: full path to icon
@type file_path: string
@param icon_size: the scaling information to be used for the
(prev_desc, prev_comment) = \
get_description_and_comment(bibarchive.get_bibdoc(docname).list_latest_files())
# List all files that are not icons or subformats
current_files = [bibdocfile.get_path() for bibdocfile in bibdoc.list_latest_files() if \
not bibdocfile.get_subformat() and not bibdocfile.is_icon()]
## current_files = []
## if not force:
## current_files = [bibdocfile.get_path() for bibdocfile bibdoc.list_latest_files()]
for current_filepath in current_files:
# Convert
new_files = createRelatedFormats(fullpath=current_filepath,
overwrite=force,
consider_version=True)
# Append
for new_file in new_files:
try:
bibdoc = bibarchive.add_new_format(new_file, docname,
prev_desc, prev_comment)
except Exception, e:
write_message(
"Could not add format to BibDoc with docname %s: %s" %
(docname, e))
continue
write_message("All files successfully processed and attached")
cli_fix_marc(None, [recid], interactive=False)
return 1
def main():
"""Core loop."""
check_running_process_user()
logfilename = '%s/fulltext_files_migration_kit-%s.log' % (
CFG_LOGDIR, datetime.today().strftime('%Y%m%d%H%M%S'))
try:
logfile = open(logfilename, 'w')
except IOError as e:
print(
wrap_text_in_a_box(
'NOTE: it\'s impossible to create the log:\n\n %s\n\nbecause of:\n\n %s\n\nPlease run this migration kit as the same user who runs Invenio (e.g. Apache)'
% (logfilename, e),
style='conclusion',
break_long=False))
sys.exit(1)
bibdoc_bibdoc = retrieve_bibdoc_bibdoc()
print(
wrap_text_in_a_box(
"""This script migrate the filesystem structure used to store icons files to the new stricter structure.
This script must not be run during normal Invenio operations.
It is safe to run this script. No file will be deleted.
Anyway it is recommended to run a backup of the filesystem structure just in case.
A backup of the database tables involved will be automatically performed.""",
style='important'))
if not bibdoc_bibdoc:
print(wrap_text_in_a_box("No need for migration", style='conclusion'))
return
print("%s icons will be migrated/fixed." % len(bibdoc_bibdoc))
wait_for_user()
print("Backing up database tables")
try:
if not backup_tables():
print(
wrap_text_in_a_box(
"""It appears that is not the first time that you run this script.
Backup tables have been already created by a previous run.
In order for the script to go further they need to be removed.""",
style='important'))
wait_for_user()
print(
"Backing up database tables (after dropping previous backup)",
end=' ')
backup_tables(drop=True)
print("-> OK")
else:
print("-> OK")
except Exception as e:
print(
wrap_text_in_a_box(
"Unexpected error while backing up tables. Please, do your checks: %s"
% e,
style='conclusion'))
sys.exit(1)
to_fix_marc = intbitset()
print("Created a complete log file into %s" % logfilename)
try:
try:
for id_bibdoc1, id_bibdoc2 in bibdoc_bibdoc:
try:
record_does_exist = True
recids = get_recid_from_docid(id_bibdoc1)
if not recids:
print("Skipping %s" % id_bibdoc1)
continue
for recid in recids:
if record_exists(recid[0]) > 0:
to_fix_marc.add(recid[0])
else:
record_does_exist = False
if not fix_bibdoc_bibdoc(id_bibdoc1, id_bibdoc2, logfile):
if record_does_exist:
raise StandardError(
"Error when correcting document ID %s" %
id_bibdoc1)
except Exception as err:
print("ERROR: %s" % err, file=logfile)
print(wrap_text_in_a_box("DONE", style='conclusion'))
except:
logfile.close()
register_exception()
print(
wrap_text_in_a_box(
title="INTERRUPTED BECAUSE OF ERROR!",
body=
"""Please see the log file %s for what was the status prior to the error. Contact %s in case of problems, attaching the log."""
% (logfilename, CFG_SITE_SUPPORT_EMAIL),
style='conclusion'))
sys.exit(1)
finally:
print(
"Scheduling FIX-MARC to synchronize MARCXML for updated records.")
cli_fix_marc(options={}, explicit_recid_set=to_fix_marc)
done, exceptions = create_icons_for_record(recid, icon_sizes,
icon_format_mappings,
docnames, add_default_icon,
inherit_moreinfo)
updated += done
if exceptions:
for ex in exceptions:
write_message(ex)
else:
write_message("Recid %s DONE." % recid)
task_update_progress("Done %d out of %d." % (i, len(recids)))
task_sleep_now_if_required(can_stop_too=True)
if updated:
cli_fix_marc(None, explicit_recid_set=recids, interactive=False)
return 1
def _create_icon(file_path,
icon_size,
docname,
icon_format='gif',
verbosity=9):
"""
Creates icon of given file.
Returns path to the icon. If creation fails, return None, and
register exception (send email to admin).
(prev_desc, prev_comment) = \
get_description_and_comment(bibarchive.get_bibdoc(docname).list_latest_files())
# List all files that are not icons or subformats
current_files = [bibdocfile.get_path() for bibdocfile in bibdoc.list_latest_files() if \
not bibdocfile.get_subformat() and not bibdocfile.is_icon()]
## current_files = []
## if not force:
## current_files = [bibdocfile.get_path() for bibdocfile bibdoc.list_latest_files()]
for current_filepath in current_files:
# Convert
new_files = createRelatedFormats(fullpath=current_filepath,
overwrite=force,
consider_version=True)
# Append
for new_file in new_files:
try:
bibdoc = bibarchive.add_new_format(new_file,
docname,
prev_desc,
prev_comment)
except Exception, e:
write_message("Could not add format to BibDoc with docname %s: %s" % (docname, e))
continue
write_message("All files successfully processed and attached")
cli_fix_marc(None, [recid], interactive=False)
return 1