Convert uniprot_trembl to massive csv

unisplat.py

from tqdm import tqdm
import csv

order = [
    "",
    "ID",
    "AC",
    "DT",
    "DE",
    "GN",
    "OS",
    "OG",
    "OC",
    "OX",
    "OH",
    "RN",
    "RP",
    "RC",
    "RX",
    "RG",
    "RA",
    "RT",
    "RL",
    "CC",
    "DR",
    "PE",
    "KW",
    "FT",
    "SQ",
    "  ",
    "//",    
]

columns = [
    "identification",
    "accession",
    "date",
    "description",
    "gene_names",
    "organism_species",
    "organelle",
    "organism_classification",
    "taxonomy_cross_reference",
    "organism_host",
    "reference_number",
    "reference_position",
    "reference_comments",
    "reference_cross_references",
    "reference_group",
    "reference_authors",
    "reference_title",
    "reference_location",
    "comments_or_notes",
    "database_cross_references",
    "protein_existence",
    "keywords",
    "feature_table_data",
    "sequence_header",
    "sequence",
]

current_code = ""
column_string = []
seq_buffer = []

def handle_line(line, writer):
    global current_code, column_string, seq_buffer
    code = line[:2]
    rest = line[4:]
    if code == current_code:
        column_string.append(rest.strip())
    else:
        index = order.index(current_code)
        next_token = order[(index+1) % len(order)]
        while code != next_token:
            seq_buffer.append("")
            current_code = next_token
            index = order.index(current_code)
            next_token = order[(index+1) % len(order)]
        current_code = code
        seq_buffer.append("\n".join(column_string))
        column_string = []
        if code == "//":
            writer.writerow(seq_buffer)
            seq_buffer = []
    
with open("/home/paul/data/uniprot_trembl.csv", 'w') as csvfile: 
    csvwriter = csv.writer(csvfile) 
    csvwriter.writerow(columns)
    with open("/home/paul/data/uniprot_trembl.dat", 'r') as infile:
        for line in tqdm(infile):
            handle_line(line, csvwriter)