davisd / python-scriptures Goto Github PK

('Obadiah', 'Obad', 'Obad', [21]),
should be
('Obadiah', 'Obad', 'Obad(?:iah)?', [21]),

As it stands this does not return anything
scriptures.extract('Obadiah 3')
but this does
scriptures.extract('Obad 3')

Let me know if you want a pull request.

EDIT Fix a big in my fix. :-)

I'm using python-scriptures to parse sermon passages out of ID3 tags from the sermon mp3 files. I came across this passage: I Sam 28-28:2 which fails with the following trace:

Traceback (most recent call last):
  File "./mp3grabber.py", line 211, in <module>
    postSermon(u, tags)
  File "./mp3grabber.py", line 173, in postSermon
    sermon = sermon_tagger.tag_sermon(sermon, tags)     
  File "/home/micah/dev/fairwaypark/orgsermons.py", line 144, in tag_sermon
    raise e
  File "/home/micah/dev/fairwaypark/orgsermons.py", line 118, in tag_sermon
    book = extract_book(scriptures.extract(tags_text))
  File "/home/micah/dev/fairwaypark/scriptures/references.py", line 28, in extract
    references.append(normalize_reference(*r.groups()))
  File "/home/micah/dev/fairwaypark/scriptures/references.py", line 120, in normalize_reference
    or (chapter == end_chapter and end_verse < verse))):
TypeError: unorderable types: int() < NoneType()

Hi, this is a great script, nice feature set. I was hoping this script could return how the references would be spoken literally. Example:
"1 PETER 5:5 NKJ"
Result: First Peter Chapter 5 Verse 5
"1 JOHN 3:8b CEB"
Result: First John Chapter 3 Verse 8
"2 CORINTHIANS 10:4-5 CEB "
Result: Second Corinthians Chapter 10 Verses 4 and 5
"Genesis 3:1-5"
Result: Genesis Chapter 3 Verses 1 through 5

Thanks for your work.

One thing that would be really useful here is the ability to chain related readings. For example, if I did:

kjv.extract("Matt 23:37-39,24:1-14")
I'd like it to return:
[(u'Matthew', 23, 37, 23, 39), (u'Matthew', 24, 1, 24, 14)]

(Actually, since Matt 23 ends at 39, better would be
[(u'Matthew', 23, 37, 24, 14)]

Right now, it ignores the 24:1-14 as it doesn't begin with a book name. I'd like it to remember the previous book name and if a book name (or chapter) is missing, assume it from the previous one.

Best would be:
kjv.extract("Matt 22,23:1-14,18,35-39,24:1-14,18")
returning
[(u'Matthew', 22, 1, 23, 14), (u'Matthew', 23, 18, 23, 18), (u'Matthew', 23, 35, 24, 14), (u'Matthew', 24, 18, 24, 18)]

I'd be happy if it required them to be consecutive. I wouldn't expect:
kjv.extract("If you look at Matt 23:10-14, you'll see one of the 15 occurrence of the word 'hypocrite'")
to give ('Matthew', '23, 10, 23, 15)

Maybe this could be a different method than extract?

I just started using this; nicely done!

FYI, blocher/python-scriptures introduced new features, such as "support for non-contiguous passages" commit ref.

At least some of these features would be nice to back-port into your repo and update the release on pypi.

If you'd like help with a PR for this, please feel free to ping me

Would you be willing to accept a merge request that adds the Greek Old Testament (Septuagint/LXX) books?

This would require a special case for the Psalms since the numbering is different in the LXX, Daniel because there is additional text, the addition of several books (e.g. 1-4 Maccabees), among other changes.

Would be great to be able to extract abbreviations that have a period.

For example, we currently have this:

>>> extract("Num. 1:2 and 1 John 1")
[('I John', 1, 1, 1, 10)]

would be nice to have

>>> extract("Num. 1:2 and 1 John 1")
[('Numbers', 1, 2, 1, 2), ('I John', 1, 1, 1, 10)]