Preserve HTML formatting when generating readings

obynio · May 19, 2021 · eac2b34 · eac2b34
1 parent f2fb21c
commit eac2b34
Show file tree

Hide file tree

Showing 2 changed files with 37 additions and 15 deletions.
diff --git a/__init__.py b/__init__.py
@@ -23,6 +23,7 @@
 from aqt.utils import showInfo, tooltip
 from aqt.qt import *
 
+from anki.buildinfo import version
 from anki.hooks import addHook
 
 from . import reading
@@ -44,7 +45,6 @@ def generateFurigana(editor, s):
     html = s.selected
     html = re.sub('\[[^\]]*\]', '', html)
     html = mecab.reading(html)
-
     if html == s.selected:
         tooltip(_("Nothing to generate!"))
     else:
@@ -71,14 +71,21 @@ class Selection:
 
     def __init__(self, window, callback):
         self.window = window
-        window.web.page().runJavaScript(self.js_get_html, lambda x: self.setHtml(x, callback))
+        self.setHtml(None, callback)
+
+    def isDeprecated(self):
+        return int(version.replace('.', '')) < 2141
 
     def setHtml(self, elements, callback, allowEmpty=False):
-        self.selected = self.window.web.selectedText()
-        if self.window.web.selectedText() == '':
-            self.window.web.eval("setFormat('selectAll');")
-            self.window.web.page().runJavaScript(self.js_get_html, lambda x: self.setHtml(x, callback, True))
+        self.selected = elements
+        if self.selected == None:
+            if self.isDeprecated():
+                self.window.web.eval("setFormat('selectAll');")
+                self.window.web.page().runJavaScript(self.js_get_html, lambda x: self.setHtml(x, callback, True))
+            else:
+                self.window.web.page().runJavaScript("getCurrentField().fieldHTML", lambda x: self.setHtml(x, callback, True))
             return
+
         self.selected = self.convertMalformedSpaces(self.selected)
         callback(self)
 
@@ -87,6 +94,9 @@ def convertMalformedSpaces(self, text):
 
     def modify(self, html):
         html = self.convertMalformedSpaces(html)
-        self.window.web.eval("setFormat('insertHTML', %s);" % json.dumps(html))
+        if self.isDeprecated():
+            self.window.web.eval("setFormat('insertHTML', %s);" % json.dumps(html))
+        else:
+            self.window.web.page().runJavaScript("getCurrentField().fieldHTML = %s;" % json.dumps(html))
 
 addHook("setupEditorButtons", addButtons)
diff --git a/reading.py b/reading.py
@@ -13,21 +13,31 @@
 import os
 import re
 import subprocess
-from anki.utils import stripHTML, isWin, isMac
+from anki.utils import isWin, isMac
 
 kakasiArgs = ["-isjis", "-osjis", "-u", "-JH", "-KH"]
 mecabArgs = ['--node-format=%m[%f[7]] ', '--eos-format=\n',
              '--unk-format=%m[] ']
 
 mecabDir = os.path.join(os.path.dirname(__file__), "support")
 
+HTML_REPLACER = '▦'
+NEWLINE_REPLACER = '▧'
+
+def htmlReplace(text):
+    pattern = r"(?:<[^<]+?>)"
+    matches = re.findall(pattern, text)
+    text = re.sub(r"<[^<]+?>", HTML_REPLACER, text)
+    return matches, text
+
 def escapeText(text):
     text = text.replace("\n", " ")
     text = text.replace(u'\uff5e', "~")
-    text = re.sub("<br( /)?>", "---newline---", text)
-    text = stripHTML(text)
-    text = text.replace("---newline---", "<br>")
-    return text
+    text = re.sub("<br( /)?>", NEWLINE_REPLACER, text)
+    #showInfo(text)
+    matches, text = htmlReplace(text)
+    text = text.replace(NEWLINE_REPLACER, "<br>")
+    return matches, text
 
 if sys.platform == "win32":
     si = subprocess.STARTUPINFO()
@@ -72,7 +82,7 @@ def ensureOpen(self):
 
     def reading(self, expr):
         self.ensureOpen()
-        expr = escapeText(expr)
+        matches, expr = escapeText(expr)
         self.mecab.stdin.write(expr.encode("utf-8", "ignore") + b'\n')
         self.mecab.stdin.flush()
         expr = self.mecab.stdout.readline().rstrip(b'\r\n').decode('utf-8', "ignore")
@@ -126,7 +136,9 @@ def reading(self, expr):
                         reading[:placeL], kanji[placeL:-placeR],
                         reading[placeL:-placeR], reading[-placeR:]))
         fin = ''.join(out)
-        return fin.strip().replace("< br>", "<br>")
+        for match in matches:
+            fin = fin.replace(HTML_REPLACER, match, 1)
+        return re.sub(r'& ?nbsp ?;', ' ', re.sub(r"< ?br ?>", "<br>", fin.strip()))
 
 # Kakasi
 
@@ -152,7 +164,7 @@ def ensureOpen(self):
 
     def reading(self, expr):
         self.ensureOpen()
-        expr = escapeText(expr)
+        _, expr = escapeText(expr)
         self.kakasi.stdin.write(expr.encode("sjis", "ignore") + b'\n')
         self.kakasi.stdin.flush()
         res = self.kakasi.stdout.readline().rstrip(b'\r\n').decode("sjis")