mynameisvinn/emailparser

remove signature blocks from emails

emailparser's Issues

the module is not installing...ERROR: Could not find a version that satisfies the requirement Parser (from versions: none) ERROR: No matching distribution found for Parser

ZeroDivisionError

---

ZeroDivisionError Traceback (most recent call last)
in
1 file = r"C:\Users\Sashankh\axaxaxax.txt"
----> 2 convert(file)

in convert(fname, threshold)
11 fn = fname.split(".")
12 new_fname = fn[0] + "_clean." + fn[1]
---> 13 _generate_text(sentences, new_fname)
14
15 def _read_email(fname):

in _generate_text(sentences, fname, threshold)
43 with open(fname, "w") as new_file:
44 for sentence in sentences:
---> 45 if _prob_block(sentence, tagger) < threshold:
46 new_file.write(sentence)
47

in prob_block(sentence, pos_tagger)
61 doc = pos_tagger(sentence)
62 verb_count = np.sum([token.pos != "VERB" for token in doc])
---> 63 return float(verb_count) / len(doc)

ZeroDivisionError: float division by zero

module not working even for the sample email

Hi I was trying using your code in Python 3.6. and it gives the following error:

Sentence boundaries unset. You can add the 'sentencizer' component to the pipeline with: nlp.add_pipe(nlp.create_pipe('sentencizer')) Alternatively, add the dependency parser, or set sentence boundaries by setting doc[i].sent_start

Not sure how to rectify this. Can you please help

Misinterpreting headers

This may be useful for some emails but the idea of taking the ratio of non-verbs to total words in a given sentence will not always work. If you have a html email with headers like

Tent Setup Guide Newsletter

This strategy will fault in this example as the given heading does not have verb structures and has 100% non verbs. This strategy will replace useful features such as headers in the given mail.

TypeError: a bytes-like object is required, not 'str'

Hi vin,

Nice code! Very useful!!

I'm trying to re-run your code using your example, but I got the following error message. Sincerely hope you could help me with. Thanks a lot!

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-6-5e07bfb80a1e> in <module>()
      1 pos_tagger = English()
      2 msg_raw = read_email('emails/test0.txt')
----> 3 sentences = corpus2sentences(msg_raw)

~\Desktop\EmailParser-master\Parser.py in corpus2sentences(corpus)
     12     """split corpus into a list of sentences.
     13     """
---> 14     return corpus.strip().split('\n')
     15 
     16 def generate_text(sentences, pos_parser, fname, threshold=0.9):

TypeError: a bytes-like object is required, not 'str'

Thanks,
S

OSError: [E050] Can't find model 'en_core_web_sm'. It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory.

Getting this error while running the example file also -

`---------------------------------------------------------------------------
OSError Traceback (most recent call last)
in
----> 1 convert(fname) # a copy of test0, without signature block

~\EmailParser-master\Parser.py in convert(fname, threshold)
11 fn = fname.split(".")
12 new_fname = fn[0] + "_clean." + fn[1]
---> 13 _generate_text(sentences, new_fname)
14
15 def _read_email(fname):

~\EmailParser-master\Parser.py in _generate_text(sentences, fname, threshold)
41 Lower thresholds will result in more false positives.
42 """
---> 43 tagger = spacy.load('en_core_web_sm')
44
45 with open(fname, "w") as new_file:

~\Anaconda3\lib\site-packages\spacy_init_.py in load(name, **overrides)
28 if depr_path not in (True, False, None):
29 warnings.warn(Warnings.W001.format(path=depr_path), DeprecationWarning)
---> 30 return util.load_model(name, **overrides)
31
32

~\Anaconda3\lib\site-packages\spacy\util.py in load_model(name, **overrides)
173 elif hasattr(name, "exists"): # Path or Path-like to model data
174 return load_model_from_path(name, **overrides)
--> 175 raise IOError(Errors.E050.format(name=name))
176
177

OSError: [E050] Can't find model 'en_core_web_sm'. It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory.`