feat: add shell version

.gitignore

@@ -0,0 +1,3 @@
+.direnv
+.envrc
+.history

keinverlag

@@ -0,0 +1,58 @@
+#!/bin/sh
+
+kv_GET () {
+  route=$1
+  shift
+  curl -s "https://www.keinverlag.de$route" "$@" | iconv -f latin1 -t utf8
+}
+
+kv_author_id () {
+  if [ $# -ne 1 ]; then
+    echo Please call kv_author_id with an author name. >/dev/stderr
+    exit 1
+  fi
+
+  author_name=$1
+
+  kv_GET "/$author_name.kv" \
+    | sed -n 's/.*autor=\([0-9]\+\).*/\1/p' \
+    | head -1
+}
+
+kv_text () {
+  if [ $# -ne 1 ]; then
+    echo Please call kv_text with a text ID. >/dev/stderr
+    exit 1
+  fi
+
+  text_id=$1
+
+  kv_GET "/$text_id.text" \
+    | sed -n '/<h1>/,/<!-- Kommentarbox -->/p' \
+    | pandoc -f html -t plain
+}
+
+kv_author_texts () {
+  if [ $# -ne 1 ]; then
+    echo Please call kv_author_texts with an author ID. >/dev/stderr
+    exit 1
+  fi
+
+  author_id=$1
+
+  kv_GET '/autorentexte.php' -d sortby=datum -d start=0 -d limit=10000 -d autor="$author_id" \
+    | sed -n 's/.*<a href="\([0-9]\+\)\.text">.*/\1/p'
+}
+
+case $1 in
+  text)
+    shift
+    kv_text "$@";;
+  author)
+    shift
+    for text_id in $(kv_author_texts "$(kv_author_id "$@")"); do
+      kv_text "$text_id"
+    done ;;
+  *)
+    echo >/dev/stderr "Usage: $0 text|author ID"
+esac

kevin.py

@@ -6,16 +6,19 @@ import re
 import requests
 
 
-def soup_from(url):
-    return BeautifulSoup(requests.get(url).text, "lxml")
+def soup_from(response):
+    return BeautifulSoup(response.text, "lxml")
 
 
 class Author:
     def __init__(self, author_id: int) -> None:
-        author_texts_url = "https://www.keinverlag.de/autorentexte.php?start=0&limit=1000000&sortby=tnr&autor={}".format(author_id)
-        soup = soup_from(author_texts_url)
+        response = requests.get(
+            "https://www.keinverlag.de/autorentexte.php",
+            params={"start": 0, "limit": 10000, "sortby": "tnr", "author": author_id},
+        )
+        soup = soup_from(response)
         self.texts = []  # type: List[Text]
-        for text in soup.select("ul.textliste > li > a[href$=\".text\"]"):
+        for text in soup.select('ul.textliste > li > a[href$=".text"]'):
             # strip off the last five characters (".text")
             text_id = int(text["href"][:-5])
             try:
@@ -37,16 +40,26 @@ class Author:
 
 class Text:
     def __init__(self, text_id: int) -> None:
-        normalization = {132: "\"", 147: "\"", 0x96: "--", 0x91: "'", 0x92: "'", 0x97: "---"}
+        normalization = {
+            132: '"',
+            147: '"',
+            0x96: "--",
+            0x91: "'",
+            0x92: "'",
+            0x97: "---",
+        }
         text_url = "https://www.keinverlag.de/{}.text".format(text_id)
         soup = soup_from(text_url)
         try:
             self.title = soup.select("h1 > span")[0].text.translate(normalization)
-            self.content = BeautifulSoup(re.sub(
-                r'<span style="font-style: italic;">(([\n\r]|.)*?)</span>',
-                r"_\1_",
-                str(soup.select(".fliesstext > span")[0])
-                ), "lxml").text.translate(normalization)
+            self.content = BeautifulSoup(
+                re.sub(
+                    r'<span style="font-style: italic;">(([\n\r]|.)*?)</span>',
+                    r"_\1_",
+                    str(soup.select(".fliesstext > span")[0]),
+                ),
+                "lxml",
+            ).text.translate(normalization)
             self.author = soup.select("h3 > a")[2].text
             self.type = soup.select("h1 ~ h3")[0].text
         except IndexError:
@@ -56,8 +69,11 @@ class Text:
         return "#### {maybe_author}{title}{maybe_type}\n\n{content}".format(
             title=self.title,
             maybe_author=self.author + ": " if with_author else "",
-            maybe_type=" ("+self.type+")" if with_type else "",
-            content="\n".join(line + "\\" if line else "" for line in self.content.splitlines()))
+            maybe_type=" (" + self.type + ")" if with_type else "",
+            content="\n".join(
+                line + "\\" if line else "" for line in self.content.splitlines()
+            ),
+        )
 
 
 if __name__ == "__main__":
@@ -67,11 +83,17 @@ if __name__ == "__main__":
 
     handle_text = subparsers.add_parser("text", help="Handle one text")
     handle_text.add_argument("tid", help="KeinVerlag text id", type=int)
-    handle_text.set_defaults(func=lambda a: print(Text(a.tid).markdown(with_type=a.type)))
+    handle_text.set_defaults(
+        func=lambda a: print(Text(a.tid).markdown(with_type=a.type))
+    )
 
-    handle_author = subparsers.add_parser("author", help="Handle all texts by an author")
+    handle_author = subparsers.add_parser(
+        "author", help="Handle all texts by an author"
+    )
     handle_author.add_argument("aid", help="KeinVerlag author id", type=str)
-    handle_author.set_defaults(func=lambda a: print(Author(a.aid).markdown(with_type=a.type)))
+    handle_author.set_defaults(
+        func=lambda a: print(Author(a.aid).markdown(with_type=a.type))
+    )
 
     args = parser.parse_args()
     args.func(args)

shell.nix

@@ -0,0 +1,10 @@
+{ pkgs ? import <nixpkgs> {} }:
+pkgs.mkShell {
+  buildInputs = with pkgs; [
+    pandoc
+    python3Packages.beautifulsoup4
+    python3Packages.requests
+    python3Packages.lxml
+  ];
+  shellHook = "export HISTFILE=${toString ./.history}";
+}