run_vecalign

import os
import subprocess
import sys

def run_vecalign(name,volume):
    j_file = f"./all_volumes/{name}_jp_{volume}.txt"
    k_file = f"./all_volumes/{name}_ko_{volume}.txt"
    j_overlap = f"./all_volumes/{name}_jp_{volume}.overlap.txt"
    k_overlap = f"./all_volumes/{name}_ko_{volume}.overlap.txt"
    j_embed = f"./all_volumes/{name}_jp_{volume}.embed"
    k_embed = f"./all_volumes/{name}_ko_{volume}.embed"
    alignment = f"./alignments/{name}_{volume}.txt"
    alignment_v = f"./alignments/{name}_{volume}.verbose.txt"

    subprocess.run(f"./vecalign/overlap.py -i {j_file} -o {j_overlap} -n 3",shell=True)
    subprocess.run(f"./vecalign/overlap.py -i {k_file} -o {k_overlap} -n 3",shell=True)
    subprocess.run(f"./LASER/tasks/embed/embed.sh {j_overlap} ja {j_embed}",shell=True)
    subprocess.run(f"./LASER/tasks/embed/embed.sh {k_overlap} ko {k_embed}",shell=True)
    subprocess.run(f"./vecalign/vecalign.py --alignment_max_size 3 --src {j_file} --tgt {k_file} --src_embed {j_overlap} {j_embed} --tgt_embed {k_overlap} {k_embed} > {alignment}",shell=True)
    # with open(alignment,'r',encoding='utf-8') as file:
    #     lines = file.readlines()
    # with open(j_file,'r',encoding='utf-8') as f:
    #     jp = f.read().split('\n')
    # with open(k_file,'r',encoding='utf-8') as f:
    #     ko = f.read().split('\n')
    # with open(alignment_v,'w',encoding='utf-8') as destfile:
    #     for line in lines:
    #         ret = ""
    #         t=line.split(':')
    #         tgt,dst,score = t[0],t[1],t[2]
    #         tgt = tgt.replace('[','').replace(']','').split(',')
    #         dst = dst.replace('[','').replace(']','').split(',')
    #         if len(tgt[0])==0:
    #             continue
    #         if len(dst[0])==0:
    #             continue
    #         for no in tgt:
    #             ret += jp[int(no)]
    #         for no in dst:
    #             ret += ko[int(no)]
    #         ret += score
    #         destfile.write(ret)
    os.remove(j_overlap)
    os.remove(k_overlap)
    os.remove(j_embed)
    os.remove(k_embed)

def parse_infofile(infofile):
    ret={}
    with open(infofile,'r') as file:
        t = file.read()
    l = t.split('\n')
    current_key = ""
    for content in l:
        if content.strip().isalpha() and len(content)>1:
            current_key = content.strip()
            ret[current_key] = []
        else:
            for num in content.split():
                ret[current_key].append(num)
    return ret

if __name__ == "__main__":
    info = parse_infofile("# volumes.txt")
    print("start.")
    for name in info.keys():
        for vol in info[name]:
            try:
                run_vecalign(name,vol)
            except KeyboardInterrupt:
                raise KeyboardInterrupt
            except:
                print(f"error with {name} {vol}")