script for building DictLat from William Whitaker's Words. Contains 35288 lemmas.

This commit is contained in:
aarne
2017-01-19 14:56:43 +00:00
parent 7a8b1caa32
commit 892f50d57d
2 changed files with 93 additions and 0 deletions
+33
View File
@@ -0,0 +1,33 @@
Aarne Ranta 2017-01-19
A script for building a Latin dictionary from William Whitaker's Words
Source: http://archives.nd.edu/whitaker/dictpage.zip
See also: https://en.wikipedia.org/wiki/William_Whitaker's_Words
License: BSD for the script. Unknown for the source, therefore not redistributed here.
Usage:
runghc MkLatin.hs
Output:
DictLatAbs.gf
DictLat.gf
Uses Herbert Lange's RGL implementation of Latin (darcs version of 2017-01-19 with ParadigmsLat.mkAdv added by AR)
Coverage of first version: 90%; 3938 missing out of 39227 entries. Marked with leading "--"
Reasons for uncovered entries
- uncommon category (e.g. structural words)
- missing case in ParadigmsLat.gf, e.g. deponent verbs
Matching for adjectives (A) uncertain, because we only use ParadigmsLat.gf.mkA : Str -> A
Verb subcat information is uncertain, too. It is based on the src's TRANS and INTRANS flags.