[hanzi-flash] hsk stuff

hanzi-flash/README.md

@@ -2,4 +2,9 @@
 
 Generate flashcards for a range of frequent hanzi characters.
 
-You need the [hanziDB CSV](https://github.com/ruddfawcett/hanziDB.csv) file for this to work.
+There are multiple sources for words:
+
+- [hanziDB CSV](https://github.com/ruddfawcett/hanziDB.csv) (purely frequency based)
+- [hsk CSV](https://github.com/plaktos/hsk_csv) (common usage words, graded by difficulty)
+
+Be careful to keep the filenames as `hanzi_db.csv` and `hsk*.csv` because the script tries to recognize these.

hanzi-flash/hanzi_flash.py

@@ -8,6 +8,7 @@ Based on https://github.com/ruddfawcett/hanziDB.csv
 import csv
 import itertools
 import argparse
+import re
 from pathlib import Path
 
 parser = argparse.ArgumentParser()
@@ -17,9 +18,15 @@ parser.add_argument("-O", "--output", default="hanzi_flash.csv", type=Path)
 parser.add_argument("-i", "--input", default="hanzi_db.csv", type=Path)
 args = parser.parse_args()
 
+offset = 1
+
+fname: str = args.input.stem
+if fname.startswith("hsk"):
+    offset = 0
+
 with open(args.input) as csv_file:
     reader = csv.reader(csv_file)
     with open(args.output, "w") as outp_file:
         writer = csv.writer(outp_file)
         for row in itertools.islice(reader, args.start, args.end + 1):
-            writer.writerow([row[1], f"{row[2]} ({row[3]})"])
+            writer.writerow([row[offset], f"{row[offset+1]} ({row[offset+2]})"])