Process text file of fulltext articles from news publications

I'm relatively new to programming and have started writing some code for work. This script is something I wrote that I'd like to have someone comment on. Am I doing anything silly, stupid or totally unnecessary? Am I making things more complicated than they need to be?
This script takes a plain text file downloaded from a publication database (example file below) and processes it to create a csv of the metadata from all the articles and a separate text file for the full text of each article.

```
from sys import argv
import time
import csv

timestamp = time.strftime('%H%M-%Y%m%d')

class Article(object):
fulltxt = ""
Title = ""
Publication = ""
Date = ""
Year = ""
Author = ""
PQID = ""

csvName = 'Metadata_'+timestamp+'.csv'
fields = ['ID', 'Title', 'Author', 'Publication', 'Date', 'Year']
csvFile = open(csvName, 'wb')
csvwriter = csv.DictWriter(csvFile, delimiter=',', fieldnames=fields)
#write csv headers
csvwriter.writerow(dict((fn, fn) for fn in fields))

def abbreviate(pub):
#a publication abbreviation is included in the filename of the textfile
pubs = {
'USA TODAY (pre-1997 Fulltext)' : 'USAT',
'USA TODAY' : 'USAT',
'New York Times' : 'NYT',
'Wall Street Journal' : 'WSJ'
}
return pubs.get(pub, "X")

def writeMetadata(doc):
#write the field values of the current doc to the csv file
if doc.fulltxt != "":
writefields = dict((field, getattr(doc, field)) for field in vars(doc) if not field.startswith('__') and field != 'fulltxt')
csvwriter.writerow(writefields)

def writeArticle(doc):
name = abbreviate(doc.Publication)+'_'+doc.Year+'_'+doc.PQID+'.txt'
txtFile = open(name, 'a')
txtFile.write(doc.Title)
txtFile.write(doc.fulltxt)
txtFile.close()

script, filename = argv
docs = open(filename)
readingtxt = False

for line in docs:
line = line.strip()
if line == '____________________________________________________________':
#write ou

Constants are CAPITALas a convention, so:

TIMESTAMP = time.strftime('%H%M-%Y%m%d')
CSV_NAME = 'Metadata_' + timestamp + '.csv' # <- added some spaces
FIELDS = ['ID', 'Title', 'Author', 'Publication', 'Date', 'Year']

Passing to more serious comments:

elif line.startswith("Title:"):
     doc.Title = line[6:]
elif line.startswith("Publication title:"):
     doc.Publication = line[19:]
elif line.startswith("Author:"):
     doc.Author = line[8:]
elif line.startswith("Publication date:"):
     doc.Date = line[18:]
elif line.startswith("ProQuest document ID:"):
     doc.PQID = line[22:]
elif line.startswith("Publication year:"):
     doc.Year = line[18:]

^^ That was a real puzzler, I know understood that the slicing is there to avoid saving things like "ProQuest document ID:" or "Publication year:", I suggest just using split and [1], like this:

elif line.startswith("Publication year:"):
     doc.Year = line.split("Publication year:")[1]

The same result of what you did but more intuitive

In fact you should use a loop:

for salient_part in ("Title", "Publication Title:" ...):
    if line.startswith(salient_part):
        doc[salient_part] = line.split(salient_part)[1]

^ The loop above may need other modifications to your code before working properly (not tested).

Possibly, you may stop using the

class Article(object):
    fulltxt = ""
    Title = ""
    Publication = ""
    Date = ""
    Year = ""
    Author = ""
    ID = ""

And just go for a plain dict, that sure would make the loop simple to implement.

Use with

txtFile = open(name, 'a')
txtFile.write(Article.Title)
txtFile.write(Article.fulltxt)
txtFile.close()

Should become

with open(name, 'a') as f:
     f.write(Article.Title)
     f.write(Article.fulltxt)

As it automatically closes the file.

Boring == possibly wrong

if line == '____________________________________________________________':

^^ Counting that the above has the correct number of underscores is boring so I will not do it (which is the correct number anyway?) you should do

if line == '_' * NUMBER_OF_UNDERSCORES_NEEDED:

StackExchange Code Review Q#105478, answer score: 5