chore: import kevin

kevin/.gitignore

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

kevin/epub.css

@@ -0,0 +1,3 @@
+body{margin:40px auto;max-width:650px;line-height:1.6;font-size:18px;color:#444;padding:0}
+a{color:inherit;text-decoration:none}
+a:hover{text-decoration:underline}

kevin/keinverlag

@@ -0,0 +1,61 @@
+#!/bin/sh
+
+alias to_utf8='iconv -f latin1 -t utf8'
+alias curl_GET='curl -X GET -s -G'
+
+BASE_URL=https://www.keinverlag.de
+
+kv_author_id () {
+  if [ $# -ne 1 ]; then
+    echo Please call kv_author_id with an author name. >/dev/stderr
+    exit 1
+  fi
+
+  author_name=$1
+
+  curl_GET "$BASE_URL/$author_name.kv" \
+    | to_utf8 \
+    | 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=$1
+
+  curl_GET "$BASE_URL/$text_id.text" \
+    | to_utf8 \
+    | sed -n '/<h1>/,/<!-- Kommentarbox -->/p' \
+    | sed 's/<h3>.\+<\/h3>//g' \
+    | 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=$1
+
+  curl_GET "$BASE_URL/autorentexte.php" -d sortby=tnr -d start=0 -d limit=10000 -d autor="$author_id" \
+    | to_utf8 \
+    | sed -n 's/.*<li><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 "$@")" | uniq); do
+      kv_text "$text_id"
+    done ;;
+  *)
+    echo >/dev/stderr "Usage: $0 text|author ID"
+esac

kevin/kevin.py

@@ -0,0 +1,99 @@
+#!/usr/bin/env python3
+from argparse import ArgumentParser
+from bs4 import BeautifulSoup
+from typing import List
+import re
+import requests
+
+
+def soup_from(response):
+    return BeautifulSoup(response.text, "lxml")
+
+
+class Author:
+    def __init__(self, author_id: int) -> None:
+        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"]'):
+            # strip off the last five characters (".text")
+            text_id = int(text["href"][:-5])
+            try:
+                self.texts.append(Text(text_id))
+            except ValueError:
+                continue
+
+    def markdown(self, *, with_type: bool = False) -> str:
+        name = self.texts[0].author
+
+        def __gen():
+            yield "% {}".format(name)
+            for text in self.texts:
+                yield "\n\n* * *\n\n"
+                yield text.markdown(with_author=False, with_type=with_type)
+
+        return "\n".join(__gen())
+
+
+class Text:
+    def __init__(self, text_id: int) -> None:
+        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.author = soup.select("h3 > a")[2].text
+            self.type = soup.select("h1 ~ h3")[0].text
+        except IndexError:
+            raise ValueError("Text {} not available.".format(text_id))
+
+    def markdown(self, *, with_author: bool = True, with_type: bool = False) -> str:
+        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()
+            ),
+        )
+
+
+if __name__ == "__main__":
+    parser = ArgumentParser()
+    parser.add_argument("--type", help="Include text type", action="store_true")
+    subparsers = parser.add_subparsers()
+
+    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_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))
+    )
+
+    args = parser.parse_args()
+    args.func(args)

kevin/kevin.sh

@@ -0,0 +1,2 @@
+#!/bin/sh
+python3 kevin.py author "$1" | pandoc -f markdown+smart --table-of-contents --toc-depth=6 --standalone --css=epub.css -o "$2"

kevin/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}";
+}