def step(self):
'''
Move pfd's and/or alto's from ocr to goobi
'''
error = None
try:
self.getVariables()
#===================================================================
# Move BW-pdf file from OCR-server to Goobi
#===================================================================
self.moveFiles(self.ocr_pdf, self.goobi_pdf_bw)
#===================================================================
# Rename pdf from OCR-server to required name.
# E.g. from 1234567890_Antikva.pdf to 1234567890_bw.pdf
#===================================================================
# A: get path for existing bw pdf just moved to Goobi
old_bw_pdf_path = tools.getFirstFileWithExtension(self.goobi_pdf_bw, 'pdf')
old_bw_pdf_path = os.path.join(self.goobi_pdf_bw,old_bw_pdf_path)
# B: Rename to new path created in "getVariables"
shutil.move(old_bw_pdf_path, self.new_bw_pdf_path)
except ValueError as e:
error = str(e)
#return "Could not convert string to int - check config file."
except (TransferError, TransferTimedOut, IOError) as e:
error = str(e)
except Exception as e:
error = str(e)
return error
def getProcessTitle(path, config_file):
'''
We assume that the process title for the given directory
is the same as the name of the toc file minus .toc
'''
config = configparser.RawConfigParser()
config.read(config_file)
toc_dir = os.path.join(path, config.get('process_folder_structure', 'metadata_toc_path'))
toc_name = tools.getFirstFileWithExtension(toc_dir, 'toc')
return toc_name[:-4]
def tocExists(toc_dir):
'''
Ensure a .toc file exists in toc directory
Return filename or None
'''
try:
toc = tools.getFirstFileWithExtension(toc_dir, '.toc')
except IOError:
return False
return toc
def pageCountMatches(pdf_input_dir,input_files_dir,valid_exts):
'''
Compare num pages in pdfinfo with pages in input
picture directory.
return boolean
'''
pdf = tools.getFirstFileWithExtension(pdf_input_dir, '.pdf')
pdfInfo = tools.pdfinfo(os.path.join(pdf_input_dir, pdf))
numPages = int(pdfInfo['Pages'])
numInputFiles = tools.getFileCountWithExtension(input_files_dir,valid_exts)
return numPages == numInputFiles
def getVariables(self):
'''
This method pulls in all the variables
from the command line and the config file
that are necessary for its running.
We need a path to our toc file, our meta.xml
and a link to our DBC data service (eXist API).
Errors in variables will lead to an
Exception being thrown.
'''
self.page_offset = None
self.issnSet = False
process_path = self.command_line.process_path
toc_dir = os.path.join(
process_path,
self.getConfigItem('metadata_toc_path', section='process_folder_structure')
)
toc_name = tools.getFirstFileWithExtension(toc_dir, '.toc')
self.toc_file_path = os.path.join(toc_dir, toc_name)
self.service_url = self.getConfigItem('dbc_service', section='dbc')
self.meta_file = os.path.join(
self.command_line.process_path,
self.getConfigItem('metadata_goobi_file', section='process_files')
)
# Parse initial Goobi METS file to a dictionary tree for processing
self.meta_data,_ = dict_tools.parseXmlToDict(self.meta_file)
# For pdf info
pdf_input = self.getConfigItem('doc_limbpdf_path',
section= 'process_folder_structure')
self.pdf_input_dir = os.path.join(process_path, pdf_input)
# parse for overlapping articles
self.overlapping_articles = self.getSetting('overlapping_articles',
'bool',default=True)
# parse boolean from command line - for overlapping articles
self.default_language = self.getSetting('default_language',
'string',default='da')
def getVariables(self):
'''
Ensure we have the variables necessary to execute the script
Tools will throw an Exception otherwise
'''
process_path = self.command_line.process_path
mets_file_name = self.getConfigItem('metadata_goobi_file', None, 'process_files')
self.mets_file = os.path.join(process_path, mets_file_name)
self.ojs_root = self.getConfigItem('ojs_root')
ojs_metadata_dir = self.getConfigItem('metadata_ojs_path', None, 'process_folder_structure')
self.ojs_metadata_dir = os.path.join(process_path, ojs_metadata_dir)
pdf_path = self.getConfigItem('doc_limbpdf_path', None, 'process_folder_structure')
abs_pdf_path = os.path.join(process_path, pdf_path)
self.pdf_name = tools.getFirstFileWithExtension(abs_pdf_path, '.pdf')
self.pdf_file = os.path.join(abs_pdf_path, self.pdf_name)
# TODO: check files in 'doc_pdf_path' instead of 'doc_limbpdf_path'
# 'doc_limbpdf_path' contains the splitted pdf-files
tools.ensureFilesExist(self.mets_file)
tools.ensureDirsExist(self.ojs_metadata_dir)
# parse boolean from command line
self.overlapping_articles = self.getSetting('overlapping_articles', bool, default=True)
# Get path to generate ojs_dir -> system means "define it from system variables"
self.ojs_journal_path = self.getSetting('ojs_journal_path', default='system')
# we also need the required issue fields
req_fields = self.getConfigItem('issue_required_fields')
self.issue_required_fields = req_fields.split(';')
opt_fields = self.getConfigItem('issue_optional_fields')
self.issue_optional_fields = opt_fields.split(';')
# Set namespaces
self.mets_ns = 'http://www.loc.gov/METS/'
self.goobi_ns = 'http://meta.goobi.org/v1.5.1/'
# Set sections
self.front_matter = []
self.articles = []
self.back_matter = []
def getPdf(self):
self.pdf_name = tools.getFirstFileWithExtension(self.pdf_input_dir, ".pdf")
self.pdf_path = os.path.join(self.pdf_input_dir, self.pdf_name)
self.pdfinfo = tools.pdfinfo(self.pdf_path)
