def add_request(lang, filenames):
job_req = {
'jobname':
'build_djvu',
'run_cmd':
'python',
'force':
True,
'args': [
os.path.expanduser('build_djvu_from_ocr_rate.py'),
'-lang:' + lang,
'-only_text',
],
'max_vmem':
768,
}
for f in filenames:
job_req['args'].append(f)
db_obj = sge_jobs.DbJob()
print job_req
db_obj.add_request(**job_req)
def handle_status(params, start_response):
default_limit = 50
max_limit = 1000
state_filter = params.get('filter', '')
cmd_filter = params.get('cmd_filter', None)
limit = get_int_param(params, 'limit', default_limit, max_limit)
offset = get_int_param(params, 'offset', 0, None)
#log(params)
db_obj = sge_jobs.DbJob()
text = common_html.get_head(
'hocr', css='shared.css').encode('utf-8') + '\n <body>\n'
html, jobs = job_table(db_obj, state_filter, limit, offset, default_limit,
max_limit, cmd_filter)
text += html
text += accounting_table(db_obj, jobs, state_filter, limit, offset,
default_limit, max_limit)
text += ' </body>\n</html>'
start_response('200 OK', [('Content-Type', 'text/html; charset=UTF-8'),
('Content-Length', str(len(text))),
('Access-Control-Allow-Origin', '*')])
return [text]
def gen_hocr_request(params):
job_req = {
'jobname':
'hocr',
'run_cmd':
'python',
'args': [
os.path.expanduser('~/phe/hocr/hocr.py'),
'-lang:' + params['lang'], '-book:' + params['book']
],
'max_vmem':
1024,
}
db_obj = sge_jobs.DbJob()
db_obj.add_request(**job_req)
def add_hocr_request(lang, book, force = False):
job_req = {
'jobname' : 'hocr',
'run_cmd' : 'python',
'args' : [
os.path.expanduser('~/phe/hocr/hocr.py'),
'-lang:' + lang,
'-book:' + book
],
'max_vmem' : 2048,
}
if force:
job_req['force'] = True
db_obj = sge_jobs.DbJob()
db_obj.add_request(**job_req)
def add_tidy_ocr_request(lang, filename):
job_req = {
'jobname' : 'tidy_ocr',
'run_cmd' : 'python',
'force' : False,
'args' : [
os.path.expanduser('~/botpywi/tidy_ocr.py'),
'-lang:' + lang,
'-monochrome',
'' + filename
],
'max_vmem' : 2048,
}
db_obj = sge_jobs.DbJob()
print job_req
db_obj.add_request(**job_req)
def queue_pdf_to_djvu(ia_id):
job_req = {
'jobname' : 'pdf_to_djvu',
'run_cmd' : 'python',
'force' : True,
'args' : [
os.path.expanduser('~/phe/ocr/pdf_to_djvu.py'),
# FIXME: later use command line switch to provide a more general
# service
ia_id,
],
'max_vmem' : 2048,
}
db_obj = sge_jobs.DbJob()
print job_req
db_obj.add_request(**job_req)
def add_ocr_request(lang, filename):
job_req = {
'jobname':
'ocr',
'run_cmd':
'python',
'force':
True,
'args': [
os.path.expanduser('~/phe/ocr/ocr_djvu.py'), '-lang:' + lang,
'' + filename
],
'max_vmem':
2048,
}
db_obj = sge_jobs.DbJob()
print job_req
db_obj.add_request(**job_req)
def add_hocr_request(lang, sublang, filename):
job_req = {
'jobname':
'correct_ocr',
'run_cmd':
'python',
'force':
True,
'args': [
os.path.expanduser('~/botpywi/correct_ocr.py'), '-lang:' + lang,
'-auto', '' + filename
],
'max_vmem':
2048,
}
if sublang:
job_req['args'].append('-sublang:' + sublang),
db_obj = sge_jobs.DbJob()
print job_req
db_obj.add_request(**job_req)