IssueID # 1538: incorpated script for converting tei to dictionary

This commit is contained in:
Cyprian Laskowski 2020-09-30 23:08:14 +02:00
parent 4e7d6d50c1
commit 5e7ac5d832
4 changed files with 20 additions and 9 deletions

7
README
View File

@ -1,5 +1,5 @@
Pipeline for assigning (first creating, if necessary) superbaza
structure_ids to a file of arbitrary Slovene strings, line by line.
Pipeline for parsing a file of arbitrary Slovene string and assigning
(first creating, if necessary) structure_ids for each string.
Example usage:
@ -8,5 +8,6 @@ $ ./setup.sh
$ echo "velika miza" > ../tmp/strings.txt
$ echo "kdo ne more mimo česa" >> ../tmp/strings.txt
$ echo "pazi, avto!" >> ../tmp/strings.txt
$ echo "počitnice" >> ../tmp/strings.txt
$ source ../venv/bin/activate
$ python pipeline.py ../tmp/strings.txt ../tmp/output.xml
$ python pipeline.py ../tmp/strings.txt ../tmp/dictionary.xml

1
scripts/.gitignore vendored
View File

@ -2,3 +2,4 @@
/conllu_to_xml.py
/wani.py
/create_structures.py
/tei_to_dictionary.py

View File

@ -17,6 +17,8 @@ CONLLU_TEI_SCRIPT_NAME = 'conllu_to_xml.py'
MWE_EXTRACTION_SCRIPT_NAME = 'wani.py'
STRUCTURE_ASSIGNMENT_SCRIPT_NAME = 'assign_structures.py'
STRUCTURE_CREATION_SCRIPT_NAME = 'create_structures.py'
STRUCTURE_CREATION_SCRIPT_NAME = 'create_structures.py'
TEI_DICTIONARY_SCRIPT_NAME = 'tei_to_dictionary.py'
OBELIKS_JAR_FILE_NAME = '../resources/obeliks.jar'
TRANSLATION_FILE_NAME = '../resources/dict.xml'
@ -33,6 +35,7 @@ TEI_STRUCTURE_2_FILE_NAME = TMP_DIRECTORY + '/tei_with_structure_ids2.xml'
MWE_CSV_1_FILE_NAME = TMP_DIRECTORY + '/mwes1.csv'
MWE_CSV_2_FILE_NAME = TMP_DIRECTORY + '/mwes2.csv'
STRUCTURE_NEW_FILE_NAME = TMP_DIRECTORY + '/structures_new.xml'
DICTIONARY_XML_FILE_NAME = TMP_DIRECTORY + '/dictionary.xml'
NLP_CONFIG_MAP = {
'treebank': 'sl_ssj_jos',
@ -50,14 +53,14 @@ def run_pipeline(input_file_name, output_file_name):
run_obeliks4J(STRING_LIST_FILE_NAME, OBELIKS_RAW_FILE_NAME)
fix_xml_ids(OBELIKS_RAW_FILE_NAME, OBELIKS_TWEAKED_FILE_NAME)
run_classla(OBELIKS_TWEAKED_FILE_NAME, CLASSLA_FILE_NAME)
convert_to_tei(CLASSLA_FILE_NAME, TEI_INIT_FILE_NAME)
run_tei_conversion(CLASSLA_FILE_NAME, TEI_INIT_FILE_NAME)
run_mwe_extraction(STRUCTURE_CURRENT_FILE_NAME, TEI_INIT_FILE_NAME, MWE_CSV_1_FILE_NAME)
run_structure_assignment(STRING_LIST_FILE_NAME, TEI_INIT_FILE_NAME, MWE_CSV_1_FILE_NAME, TEI_STRUCTURE_1_FILE_NAME)
run_structure_creation(STRUCTURE_CURRENT_FILE_NAME, TEI_STRUCTURE_1_FILE_NAME, STRUCTURE_NEW_FILE_NAME)
run_mwe_extraction(STRUCTURE_NEW_FILE_NAME, TEI_INIT_FILE_NAME, MWE_CSV_2_FILE_NAME)
run_structure_assignment(STRING_LIST_FILE_NAME, TEI_INIT_FILE_NAME, MWE_CSV_2_FILE_NAME, TEI_STRUCTURE_2_FILE_NAME)
# run_format_lexicon() # TODO: implement
shutil.copyfile(TEI_STRUCTURE_2_FILE_NAME, output_file_name)
run_dictionary_conversion(TEI_STRUCTURE_2_FILE_NAME, DICTIONARY_XML_FILE_NAME)
shutil.copyfile(DICTIONARY_XML_FILE_NAME, output_file_name)
def run_obeliks4J(obeliks_file_name, classla_file_name):
print('Running obeliks ...')
@ -86,7 +89,7 @@ def run_classla(obeliks_file_name, classla_file_name):
result = nlp(doc)
result.conll_file.write_conll(classla_file_name)
def convert_to_tei(classla_file_name, tei_file_name):
def run_tei_conversion(classla_file_name, tei_file_name):
print('Converting to tei ...')
convert_command = ' '.join(['python', CONLLU_TEI_SCRIPT_NAME, classla_file_name, tei_file_name, '--translate', TRANSLATION_FILE_NAME])
os.system(convert_command)
@ -104,7 +107,12 @@ def run_structure_assignment(input_file_name, tei_file_name, mwe_csv_file_name,
def run_structure_creation(input_file_name, tei_file_name, output_file_name):
print('Creating missing structures ...')
assignment_command = ' '.join(['python', STRUCTURE_CREATION_SCRIPT_NAME, '-infile', input_file_name, '-tei', tei_file_name, '-outfile', output_file_name])
os.system(assignment_command)
creation_command = ' '.join(['python', STRUCTURE_CREATION_SCRIPT_NAME, '-infile', input_file_name, '-tei', tei_file_name, '-outfile', output_file_name])
os.system(creation_command)
def run_dictionary_conversion(tei_file_name, xml_file_name):
print('Converting to dictionary xml format ...')
convert_command = ' '.join(['python', TEI_DICTIONARY_SCRIPT_NAME, '-infile', tei_file_name, '-outfile', xml_file_name])
os.system(convert_command)
run_pipeline(input_file_name, output_file_name)

View File

@ -28,6 +28,7 @@ cd scripts
ln -s ../nova_slovnica/python/scripts/conllu_to_xml.py .
ln -s ../nova_slovnica/python/scripts/assign_structures.py .
ln -s ../nova_slovnica/python/scripts/create_structures.py .
ln -s ../nova_slovnica/python/scripts/tei_to_dictionary.py .
ln -s ../luscenje_struktur/luscenje_struktur/wani.py .
cd ..