From add536d1d08f2743393864d456fa9ce567a84fed Mon Sep 17 00:00:00 2001
From: redxef <redxef@redxef.at>
Date: Tue, 11 Mar 2025 18:40:01 +0100
Subject: [PATCH] fix: use better python implementation for import from text
 file.

---
 fromtext.py | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 fromtext.sh | 22 ----------------------
 2 files changed, 50 insertions(+), 22 deletions(-)
 create mode 100755 fromtext.py
 delete mode 100755 fromtext.sh

diff --git a/fromtext.py b/fromtext.py
new file mode 100755
index 0000000..b229358
--- /dev/null
+++ b/fromtext.py
@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+import os
+import sys
+import pathlib
+
+
+def _sanitize_line(l: str) -> str:
+    if '\n' in l and l.startswith('"') and l.endswith('"'):
+        return l[1:-1].replace('""', '"')
+    return l
+
+def process_line(
+        *,
+        columns: list[str],
+        columns_indices: list[int],
+        target_dir: pathlib.Path,
+) -> None:
+    columns = [_sanitize_line(l) for l in columns]
+    columns = [columns[i] for i in columns_indices]
+    slug = ''.join(c for c in columns[0].lower() if c.isalnum() or c == ' ')
+    slug = slug.replace(' ', '-')
+    base_filename = target_dir / slug
+    for i, c in enumerate(columns):
+        with (base_filename.parent / (base_filename.name + f'.{i}.md')).open('w') as fp:
+            fp.write(c)
+
+def main(
+        *,
+        target_column_count: int = 6,
+        columns_indices: list[int] = [3, 4],
+        sep: str = '\t',
+):
+    target_dir = pathlib.Path(next(iter(sys.argv[1:] or ['.'])))
+    buff = ''
+    process_header = True
+    for l in sys.stdin.readlines():
+        if process_header and l.startswith('#'):
+            continue
+        else:
+            process_header = False
+        buff += l
+        split_ = buff.split(sep)
+        if len(split_) == target_column_count:
+            process_line(columns=split_, columns_indices=columns_indices, target_dir=target_dir)
+            buff = ''
+
+if __name__ == '__main__':
+    main()
diff --git a/fromtext.sh b/fromtext.sh
deleted file mode 100755
index b4cb953..0000000
--- a/fromtext.sh
+++ /dev/null
@@ -1,22 +0,0 @@
-#!/bin/sh
-
-SEP='\t'
-COLUMNS='4 5'
-
-main() {
-    while read -r line; do
-        filename=
-        i=0
-        for c in $COLUMNS; do
-            f="$(echo "$line" | awk -F$SEP "{print \$$c}")"
-            [ -z "$f" ] && continue
-            if [ "$filename" = '' ]; then
-                filename="$(echo "$f" | tr '[:upper:]' '[:lower:]' | sed -E -e 's/ /-/g' -e 's/[^a-z0-9-]//g' -n -e 's/^(.{0,128}).*$/\1/p')"
-            fi
-            echo "$f" > "$filename.$i.md"
-            i=$((i+1))
-        done
-    done
-}
-
-main "$@"