Revert "stop (un)escaping media filenames"

This was flawed - while non-Latin text is now acceptable
in an IRI, we still need to be concerned with reserved characters
such as spaces, and Anki unfortunately has been storing the filenames
in unencoded form in the DB, meaning we must encode them at display
time. We won't be able to move away from this until existing notes
are rewritten, and it will probably require breaking compatibility with
older clients.

https://en.wikipedia.org/wiki/Internationalized_Resource_Identifier

This reverts commit 14110add55.

pylib/anki/media.py

@@ -8,10 +8,12 @@ import pprint
 import re
 import sys
 import time
-from typing import Any, Callable, List, Optional, Tuple
+import urllib.error
+import urllib.parse
+import urllib.request
+from typing import Any, Callable, List, Match, Optional, Tuple
 
 from anki import media_pb2
-from anki._legacy import deprecated
 from anki.consts import *
 from anki.latex import render_latex, render_latex_returning_errors
 from anki.models import NotetypeId
@@ -183,12 +185,27 @@ class MediaManager:
             txt = re.sub(reg, "", txt)
         return txt
 
-    @deprecated(info="no longer required")
     def escapeImages(self, string: str, unescape: bool = False) -> str:
-        return string
+        "escape_media_filenames alias for compatibility with add-ons."
+        return self.escape_media_filenames(string, unescape)
 
-    @deprecated(info="no longer required")
     def escape_media_filenames(self, string: str, unescape: bool = False) -> str:
+        "Apply or remove percent encoding to filenames in html tags (audio, image, object)."
+        fn: Callable
+        if unescape:
+            fn = urllib.parse.unquote
+        else:
+            fn = urllib.parse.quote
+
+        def repl(match: Match) -> str:
+            tag = match.group(0)
+            fname = match.group("fname")
+            if re.match("(https?|ftp)://", fname):
+                return tag
+            return tag.replace(fname, fn(fname))
+
+        for reg in self.html_media_regexps:
+            string = re.sub(reg, repl, string)
         return string
 
     # Checking media

pylib/tests/test_media.py

@@ -49,6 +49,10 @@ def test_strings():
     assert sp("aoeu") == "aoeu"
     assert sp("aoeu[sound:foo.mp3]aoeu") == "aoeuaoeu"
     assert sp("a<img src=yo>oeu") == "aoeu"
+    es = col.media.escape_media_filenames
+    assert es("aoeu") == "aoeu"
+    assert es("<img src='http://foo.com'>") == "<img src='http://foo.com'>"
+    assert es('<img src="foo bar.jpg">') == '<img src="foo%20bar.jpg">'
 
 
 def test_deckIntegration():

qt/aqt/editor.py

@@ -434,7 +434,10 @@ $editorToolbar.then(({{ toolbar }}) => toolbar.appendGroup({{
         if not self.note:
             return
 
-        data = self.note.items()
+        data = [
+            (fld, self.mw.col.media.escape_media_filenames(val))
+            for fld, val in self.note.items()
+        ]
         self.widget.show()
         self.updateTags()
 
@@ -584,9 +587,13 @@ $editorToolbar.then(({{ toolbar }}) => toolbar.appendGroup({{
         if html.find(">") > -1:
             # filter html through beautifulsoup so we can strip out things like a
             # leading </div>
+            html_escaped = self.mw.col.media.escape_media_filenames(html)
             with warnings.catch_warnings():
                 warnings.simplefilter("ignore", UserWarning)
-                html = str(BeautifulSoup(html, "html.parser"))
+                html_escaped = str(BeautifulSoup(html_escaped, "html.parser"))
+                html = self.mw.col.media.escape_media_filenames(
+                    html_escaped, unescape=True
+                )
         self.note.fields[field] = html
         if not self.addMode:
             self._save_current_note()
@@ -1260,9 +1267,15 @@ def remove_null_bytes(txt: str, editor: Editor) -> str:
     return txt.replace("\x00", "")
 
 
+def reverse_url_quoting(txt: str, editor: Editor) -> str:
+    # reverse the url quoting we added to get images to display
+    return editor.mw.col.media.escape_media_filenames(txt, unescape=True)
+
+
 gui_hooks.editor_will_use_font_for_field.append(fontMungeHack)
 gui_hooks.editor_will_munge_html.append(munge_html)
 gui_hooks.editor_will_munge_html.append(remove_null_bytes)
+gui_hooks.editor_will_munge_html.append(reverse_url_quoting)
 
 
 def set_cloze_button(editor: Editor) -> None:

qt/aqt/main.py

@@ -484,6 +484,7 @@ class AnkiQt(QMainWindow):
             return aqt.sound.av_refs_to_play_icons(text)
 
     def prepare_card_text_for_display(self, text: str) -> str:
+        text = self.col.media.escape_media_filenames(text)
         text = self._add_play_buttons(text)
         return text
 

qt/aqt/utils.py

@@ -34,7 +34,6 @@ from PyQt5.QtWidgets import (
 )
 
 import aqt
-from anki._legacy import deprecated
 from anki.collection import Collection
 from anki.lang import TR, tr_legacyglobal  # pylint: disable=unused-import
 from anki.utils import invalidFilename, isMac, isWin, noBundledLibs, versionWithBuild
@@ -680,8 +679,9 @@ def restore_combo_history(comboBox: QComboBox, name: str) -> List[str]:
     return history
 
 
-@deprecated(info="use mw.prepare_card_text_for_display()")
 def mungeQA(col: Collection, txt: str) -> str:
+    print("mungeQA() deprecated; use mw.prepare_card_text_for_display()")
+    txt = col.media.escape_media_filenames(txt)
     return txt