Text node processor, typo fixes, text2* cli program

mastoposter/text/__init__.py

@@ -14,7 +14,6 @@ GNU General Public License for more details.
 """
 from typing import Callable, Iterable, Literal, Optional
 from bs4.element import Tag, PageElement
-from html import escape
 
 VALID_OUTPUT_TYPES = Literal["plain", "html", "markdown"]
 BULLET = "\u2022"
@@ -39,7 +38,7 @@ node_processors: dict[
     list[
         Callable[
             [
-                Tag,
+                PageElement,
             ],
             Optional[str],
         ]
@@ -49,12 +48,21 @@ node_processors: dict[
 
 def register_converter(tag: str, output_type: VALID_OUTPUT_TYPES = "plain"):
     def decorate(function):
+        node_processors.setdefault((output_type, tag), [])
         node_processors[output_type, tag].append(function)
         return function
 
     return decorate
 
 
+def register_text_node_converter(output_type: VALID_OUTPUT_TYPES = "plain"):
+    def decorate(function):
+        node_processors[output_type, ":text:"] = [function]
+        return function
+
+    return decorate
+
+
 def register_fmt_converter(
     format: str,
     tag: str,
@@ -71,11 +79,15 @@ def register_fmt_converter(
 
 def node_process(el: PageElement, type_: VALID_OUTPUT_TYPES) -> str:
     if isinstance(el, Tag):
-        for func in node_processors[type_, el.name]:
-            result = func(el)  # XXX: could use walrus, but it's py3.8+ only
-            if result:
-                return result
-    return escape(str(el))
+        if (type_, el.name) in node_processors:
+            for func in node_processors[type_, el.name]:
+                result = func(el)
+                if result:
+                    return result
+        return nodes_process(el.children, type_)
+    if (type_, ":text:") in node_processors:
+        return node_processors[type_, ":text:"][0](el) or str(el)
+    return str(el)
 
 
 def nodes_process(

mastoposter/text/__main__.py

@@ -0,0 +1,36 @@
+"""
+mastoposter - configurable reposter from Mastodon-compatible Fediverse servers
+Copyright (C) 2022-2023 hatkidchan <hatkidchan@gmail.com>
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3 of the License, or
+(at your option) any later version.
+
+This program is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+GNU General Public License for more details.
+"""
+
+from mastoposter.text import node_process, VALID_OUTPUT_TYPES
+from argparse import ArgumentParser, FileType
+from typing import get_args as T_get_args
+from bs4 import BeautifulSoup
+import sys
+
+parser = ArgumentParser("mastoposter.text", description="HTML-to-* converter")
+
+parser.add_argument(
+    "--type",
+    "-t",
+    choices=T_get_args(VALID_OUTPUT_TYPES),
+    default=T_get_args(VALID_OUTPUT_TYPES)[0],
+    dest="output_type",
+)
+parser.add_argument("file", default=sys.stdin, type=FileType("r"))
+
+args = parser.parse_args()
+
+soup = BeautifulSoup(args.file.read(), "lxml")
+print(node_process(soup, args.output_type))

mastoposter/text/html.py

@@ -12,10 +12,12 @@ but WITHOUT ANY WARRANTY; without even the implied warranty of
 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 GNU General Public License for more details.
 """
+from bs4 import NavigableString
 from mastoposter.text import (
     nodes_process,
     register_converter,
     register_fmt_converter,
+    register_text_node_converter,
     node_process,
     STRIPE,
     BULLET,
@@ -26,6 +28,11 @@ from bs4.element import Tag
 from html import escape
 
 
+@register_text_node_converter("html")
+def proc_text_node_to_html(txt: NavigableString) -> str:
+    return escape(txt).strip()
+
+
 @register_converter("a", "html")
 def proc_tag_a_to_html(tag: Tag):
     return '<a href="%s">%s</a>' % (
@@ -81,7 +88,7 @@ def proc_tag_ul_to_html(tag: Tag) -> str:
     )
 
 
-@register_converter("li", "html")
+@register_converter("ol", "html")
 def proc_tag_li_to_html(tag: Tag) -> str:
     return "\n" + str.join(
         "\n",

mastoposter/text/markdown.py

@@ -63,7 +63,7 @@ def proc_tag_ul_to_markdown(tag: Tag) -> str:
     )
 
 
-@register_converter("li", "markdown")
+@register_converter("ol", "markdown")
 def proc_tag_li_to_markdown(tag: Tag) -> str:
     return "\n" + str.join(
         "\n",

mastoposter/text/plain.py

@@ -23,14 +23,13 @@ from mastoposter.text import (
 )
 
 from bs4.element import Tag
-from html import escape
 
 
 @register_converter("a", "plain")
 def proc_tag_a_to_plain(tag: Tag):
     return "%s (%s)" % (
         nodes_process(tag.children, "plain"),
-        escape(tag.attrs.get("href", "#")),
+        tag.attrs.get("href", "#"),
     )
 
 
@@ -64,7 +63,7 @@ def proc_tag_ul_to_plain(tag: Tag) -> str:
     )
 
 
-@register_converter("li", "plain")
+@register_converter("ol", "plain")
 def proc_tag_li_to_plain(tag: Tag) -> str:
     return "\n" + str.join(
         "\n",