adding separate database class

src/database.py

@@ -0,0 +1,23 @@
+import sqlite3
+import os
+
+class Database:
+    def __init__(self, args):
+        filename = ":memory:" if args.db is None else args.db
+
+        if not args.keep_db and os.path.exists(filename):
+            os.remove(filename)
+
+        self.new = not os.path.exists(filename)
+        self.db = sqlite3.connect(filename)
+    
+    def execute(self, *args, **kwargs):
+        return self.db.execute(*args, **kwargs)
+
+    def init(self, *args, **kwargs):
+        # same as execute, only skipped if not a new database file 
+        if self.new:
+            return self.execute(*args, **kwargs)
+    
+    def commit(self):
+        self.db.commit()

src/wani.py

@@ -16,6 +16,7 @@ from match_store import MatchStore
 from word_stats import WordStats
 from writer import Writer
 from loader import load_files
+from database import Database
 
 
 def match_file(words, structures):
@@ -37,8 +38,9 @@ def match_file(words, structures):
 def main(args):
     structures, lemma_msds, max_num_components = build_structures(args)
 
+    database = Database(args)
     match_store = MatchStore(args)
-    word_stats = WordStats(lemma_msds)
+    word_stats = WordStats(lemma_msds, database)
 
     if args.parallel:
         num_parallel = int(args.parallel)
@@ -139,6 +141,11 @@ if __name__ == '__main__':
     parser.add_argument('--sort-reversed',
                         help="Sort in reversed ored", action='store_true')
 
+    parser.add_argument('--db',
+                        help="Database file to use (instead of memory)", default=None)
+    parser.add_argument('--keep-db',
+                        help="Does not recreate new database file", action='store_true')
+
     parser.add_argument('--pc-tag',
                         help='Tag for separators, usually pc or c', default="pc")
     parser.add_argument('--parallel',

src/word_stats.py

@@ -1,34 +1,30 @@
 from collections import defaultdict, Counter
 
 from progress_bar import progress
-import sqlite3
 
 
 class WordStats:
-    def __init__(self, lemma_features):
+    def __init__(self, lemma_features, db):
         self.lemma_features = lemma_features
+        self.db = db
 
         self.all_words = 0
         self.memoized_msd_merges = {}
 
-        with open("sqlite.db", 'w') as fp:
-            fp.write("")
-
-        self.db = sqlite3.connect('sqlite.db')
-        self.db.execute("""CREATE TABLE UniqWords (
+        self.db.init("""CREATE TABLE UniqWords (
             uw_id INTEGER PRIMARY KEY, 
             lemma varchar(64), 
             msd varchar(16), 
             text varchar(64), 
             frequency int
             )""")
-        self.db.execute("CREATE TABLE CommonMsd (lemma varchar(64), msd varchar(16))")
-        self.db.execute("CREATE TABLE WordCount (lemma varchar(64), msd0 char, frequency int)")
+        self.db.init("CREATE TABLE CommonMsd (lemma varchar(64), msd varchar(16))")
+        self.db.init("CREATE TABLE WordCount (lemma varchar(64), msd0 char, frequency int)")
 
-        self.db.execute("CREATE INDEX lemma_msd_text_on_uw ON UniqWords (lemma, msd, text)")
-        self.db.execute("CREATE INDEX lemma_on_uw ON UniqWords (lemma)")
-        self.db.execute("CREATE INDEX lemma_on_cm ON CommonMsd (lemma)")
-        self.db.execute("CREATE INDEX lemma_msd0_on_wc ON WordCount (lemma, msd0)")
+        self.db.init("CREATE INDEX lemma_msd_text_on_uw ON UniqWords (lemma, msd, text)")
+        self.db.init("CREATE INDEX lemma_on_uw ON UniqWords (lemma)")
+        self.db.init("CREATE INDEX lemma_on_cm ON CommonMsd (lemma)")
+        self.db.init("CREATE INDEX lemma_msd0_on_wc ON WordCount (lemma, msd0)")
 
     def add_words(self, words):
         for w in progress(words, "adding-words"):