Excel checker

SeriesVONA_translationcheck.py

import pandas as pd
import re

indf = pd.read_excel("inputfile.xlsx")

# Build a data frame with source and target segments
df = pd.concat([indf.Japanese, indf.German], axis = 1)
df.columns = ["Source", "Target"]

# Check if model numbers in source are also in target and vice versa

pattern = re.compile(r"[A-Z0-9][A-Z0-9]+")

fout = open('checkeroutput1.txt', 'w')
fout.write("The following lines may not have the same model numbers in source and target.\n")

for i in df.index:
    left = re.findall(pattern, df.Source[i])
    right = re.findall(pattern, df.Target[i])
    triggered = False
    for code in left:
        if code not in right:
#            print("ALERT. Model numbers may not be the same in source and target.")
#            print("Row ", i, "   |    ", df.Target[i], "   |    ", df.Source[i])
            triggered = True
            break
    if not triggered:
        for code in right:
            if code not in left:
                triggered = True
                break
    if triggered:
        outstring = "Row " + str(i) + "\t" + df.Target[i] + "\t" + df.Source[i] + "\n"
        fout.write(outstring)
        
fout.close()


# Check if target contains repetitions of more than 3 words
fout = open('checkeroutput2.txt', 'w')
fout.write("The following lines have repetitions of 3+ words in the target field.\n")

for i in df.index:
    a = re.search(r"(\w+\W+\w+\W+\w+).*\1", df.Target[i])
    if a:
        outstring = "Row " + str(i) + "\t" + df.Target[i] + "\t" + df.Source[i] + "\n"
        fout.write(outstring)
        
fout.close()