Question

This is the script that has very kindly been given to me as a starter:

#!/usr/bin/python
# -*- coding: utf-8 -*-
from __future__ import with_statement    # needed for Python 2.5
from itertools import chain

def chunk(s):
    """Split a string on whitespace or hyphens"""
    return chain(*(c.split("-") for c in s.split()))

def process(latin, gloss, trans):
    chunks = zip(chunk(latin), chunk(gloss))
    # now you have to DO SOMETHING with the chunks!

def main():
    with open("examples.txt") as inf:
        try:
            while True:
                latin = inf.next().strip()
                gloss = inf.next().strip()
                trans = inf.next().strip()
                process(latin, gloss, trans)
                inf.next()    # skip blank line
        except StopIteration:
            # reached end of file
            pass

if __name__=="__main__":
    main()

I’m not sure if I’m missing anything, but the output is simply blank, taking me back to $.

I’m trying to do the following:
I have a text with a language other than English, broken up into morphemes (parts of each word) using hyphens, with the English gloss (linguistic translation of each morpheme) and a direct translation below.

eg.

Itali-am fat-o profug-us Lavini-a-que ven-it

Italy-Fem:Sg:Acc fate-Neut:Sg:Abl fleeing-Masc:Sg:Nom Lavinian-Neut:Pl:Acc come:Perf-3-Sg:Indic:Act

‘in flight [driven] by fate came to Italy and the Lavinian [shores]’

I’ll have several texts such as the above in one file – i.e.

blank line

a line of latin broken up with hyphens

a line of gloss broken up with corresponding hyphens, using colons to join elements

a line of translation

blank line

latin

gloss

translation

ad infinitum.

What I need to do is write a file that gives me the following output:

Itali:    1    Italy
am:    1    Fem:Sg:Acc
fat:    1    fate
o:    1    Neut:Sg:Abl
profug:   1    fleeing
us:    1    Masc:Sg:Nom
Lavini:    1    Lavinian
a:    1    Neug:Pl:Acc
que:    1    come:Perf
ven:    1   3
it:     1   Sg:Indic:Act

where the first column represents the first line of text without hyphens; the second column indicates the number of occurrences (it’s only 1 each in this example), and the third column is the English translation of the first column, as written in the text.

If there’s a latin morpheme with no corresponding English gloss/translation, the Latin column will be as normal but the English column will print [unknown], like:

a:  1   [unknown]

And if the opposite, i.e. an English morpheme with no corresponding Latin, it should print

[unknown]:  1   kitten

Finally, the prog needs to be able to deal with homophonous morphemes (i.e. two identically spelled latin morphemes with different meanings).
e.g.

a:  16  Neuter:Plural
a:  28  Feminine:Singular

Again, it’s homework, and any pointers would be wonderful. Working on putting together some script now to upload here for critique!

Answer 1

Processing the file is a bit tricky because of the multiline structure; rather than the usual line-by-line iteration, I suggest something like this (I presume the file does not actually begin with a blank line, per your example, but that it uses blank lines as separators):

with open("input.txt") as inf:
    try:
        while True:
            latin = inf.next().strip()
            gloss = inf.next().strip()
            trans = inf.next().strip()
            process(latin, gloss, trans)
            inf.next()    # skip blank line
    except StopIteration:
        # reached end of file
        pass

process must then split latin and gloss into chunks and pair them appropriately:

from itertools import chain

def chunk(s):
    """Split a string on whitespace or hyphens"""
    return chain(*(c.split("-") for c in s.split()))

def process(latin, gloss, trans):
    chunks = zip(chunk(latin), chunk(gloss))

Calling this like

process(
    "Itali-am fat-o profug-us Lavini-a-que ven-it",
    "Italy-Fem:Sg:Acc fate-Neut:Sg:Abl fleeing-Masc:Sg:Nom Lavinian-Neut:Pl:Acc come:Perf-3-Sg:Indic:Act",
    "in flight [driven] by fate came to Italy and the Lavinian [shores]")

leaves chunks containing

[('Itali', 'Italy'),
 ('am', 'Fem:Sg:Acc'),
 ('fat', 'fate'),
 ('o', 'Neut:Sg:Abl'),
 ('profug', 'fleeing'),
 ('us', 'Masc:Sg:Nom'),
 ('Lavini', 'Lavinian'),
 ('a', 'Neut:Pl:Acc'),
 ('que', 'come:Perf'),
 ('ven', '3'),
 ('it', 'Sg:Indic:Act')]

The rest is an exercise for the student – keeping a running count of the chunks, then sorting and displaying it appropriately. Hope that helps!