ivsort.py

a simple script to sort Hebrew words with vowels and accents according to their lexical order (which differs from the unicode point order or even the locale’s collation string)

From the docstring:

Module for properly sorting Hebrew words with niqqud – according to consonants first and vowels later. If run from the command line, takes a file with a list of words as input or reads lines from stdin.

disclaimer: there is no universal method for the ordering of Hebrew vowels. The order used by this module is:

Sheva, Hataf Segol, Hataf Patah, Hataf Qamets, Hiriq, Tsere, Segol, Patah, Qamets, Holem, Holem Vav, Qibbuts, Shureq

This order may seem foreign to those accustomed to Latin-based languages, but it was recommended to me by the Academy for the Hebrew Language in Israel, and it is very similar to orders use by Unicode and standard Israeli keyboard layouts. It is also more “scientific” in the sense that it follows a phonological pattern (horizontal to vertical vowels), rather than a traditional, Latin-based pattern (aeiou).

This does differ from the one convention common in Israeli works, in that it treats Sin as a separate letter from Shin. This is not especially common in Israel, but it is the convention in all of the widely used Biblical lexicons and other Western Biblical scholarship. I do this both because it would be pedagogical malpractice to give my students glossaries that differ in this regard from the works they will be using later, and because it is more accurate from the perspective of historical linguistics.

I would probably revise this approach if this sort engine were targeted at modern Hebrew.

$ cat wordlist.txt | ivsort
...
$ ivsort wordlist.txt

from ivsort import ivsort
sorted_list = ivsort(unsorted_iterable)