Claudine/fix normalization and house cleaning (#3843)

* remove  normalization in  glyph filter

remove glyph normalization gui/lettering/main_panel, as it  breaks the glyph filtering

remove some spelling mistake in the comments, correct some  comment

if a glyph from the  svg font file is a non chasing mark (like combining accent, or cedilla) the inkscape label of the  corresponding glyph  group created by lettering_font_sample  was empty or unreadable, make it  visible . Same in the line label.

and although it is not related
correct one glyph in malika font

* correct warning in  lettering_font_sample

do  the warnings correctly when there is somethng wrong  with the glyph list in the font.json gile

* Update lettering_font_sample.py

fix a typo

5 files changed, 52 insertions, 34 deletions

diff --git a/lib/extensions/lettering_edit_json.py b/lib/extensions/lettering_edit_json.py
index b86c4fcb..3c5aa3f4 100644
--- a/lib/extensions/lettering_edit_json.py
+++ b/lib/extensions/lettering_edit_json.py
@@ -16,7 +16,7 @@ from .base import InkstitchExtension
 
 class LetteringEditJson(InkstitchExtension):
     '''
-    This extension helps font creators to generate an output of every glyph from a selected font
+    This extension helps font creators modify the JSON file of a lettering font.
     '''
     def effect(self):
         layer = Layer()
diff --git a/lib/gui/lettering/main_panel.py b/lib/gui/lettering/main_panel.py
index f0532560..fc6d16ad 100644
--- a/lib/gui/lettering/main_panel.py
+++ b/lib/gui/lettering/main_panel.py
@@ -5,7 +5,6 @@
 
 import json
 from base64 import b64decode
-import unicodedata
 
 import inkex
 import wx
@@ -123,12 +122,16 @@ class LetteringPanel(wx.Panel):
         self.fonts_by_id = {}
 
         # font size filter value
+
         filter_size = self.options_panel.font_size_filter.GetValue()
         filter_glyph = self.options_panel.font_glyph_filter.GetValue()
         filter_category = self.options_panel.font_category_filter.GetSelection() - 1
 
         # Set of all glyphs in input string (except whitespace characters), normalized in the same way that we normalize font glyphs
-        glyphs = set(*unicodedata.normalize("NFC", self.options_panel.text_editor.GetValue().replace(r"\s", "")))
+        # do not normalize the glyphs yet, available_glyphs are not normalized in the font json file
+        # glyphs = set(l for l  in  unicodedata.normalize("NFC", self.options_panel.text_editor.GetValue().replace(r"\s", "")))
+
+        glyphs = set(letter for letter in self.options_panel.text_editor.GetValue().replace(r"\s", ""))
 
         for font in self.font_list:
             if filter_glyph and glyphs and not glyphs.issubset(font.available_glyphs):
@@ -319,7 +322,7 @@ class LetteringPanel(wx.Panel):
             else:
                 pass
 
-        # the text scaling group label is dependend on the user language, so it would break in international file exchange if we used it
+        # the text scaling group label is dependent on the user language, so it would break in international file exchange if we used it
         # scaling (correction transform) on the parent group is already applied, so let's use that for recognition
         if destination_group.get('transform', None) is None:
             destination_group.attrib['transform'] = 'scale(%s)' % (self.settings.scale / 100.0)
diff --git a/lib/gui/lettering_font_sample.py b/lib/gui/lettering_font_sample.py
index e19544dc..9f42a755 100644
--- a/lib/gui/lettering_font_sample.py
+++ b/lib/gui/lettering_font_sample.py
@@ -8,6 +8,7 @@ from copy import deepcopy
 import wx
 import wx.adv
 from inkex import Group, errormsg
+import unicodedata
 
 from ..commands import ensure_command_symbols
 from ..i18n import _
@@ -178,29 +179,14 @@ class FontSampleFrame(wx.Frame):
         text = ''
         width = 0
         last_glyph = None
-        printed_warning = False
-        update_glyphlist_warning = _(
-            "The glyphlist for this font seems to be outdated.\n\n"
-            "Please update the glyph list for {font_name}:\n"
-            "* Open Extensions > Ink/Stitch > Font Management > Edit JSON\n"
-            "* Select this font and apply."
-        ).format(font_name=self.font.marked_custom_font_name)
-
-        self.duplicate_warning()
-
-        # font variant glyph list length falls short if a single quote sign is available
-        # let's add it in the length comparison
-        if len(set(self.font.available_glyphs)) != len(self.font_variant.glyphs):
-            errormsg(update_glyphlist_warning)
-            printed_warning = True
+        outdated = False
 
         for glyph in self.font.available_glyphs:
             glyph_obj = self.font_variant[glyph]
             if glyph_obj is None:
-                if not printed_warning:
-                    errormsg(update_glyphlist_warning)
-                printed_warning = True
+                outdated = True
                 continue
+
             if last_glyph is not None:
                 width_to_add = (glyph_obj.min_x - self.font.kerning_pairs.get(f'{last_glyph} {glyph}', 0)) * scale
                 width += width_to_add
@@ -219,8 +205,8 @@ class FontSampleFrame(wx.Frame):
             text += glyph
             width += width_to_add
 
+        self.out_dated_warning(outdated)
         self._render_text(text)
-
         self.GetTopLevelParent().Close()
 
     def sortable(self):
@@ -229,20 +215,40 @@ class FontSampleFrame(wx.Frame):
             color_sort = False
         return color_sort
 
-    def duplicate_warning(self):
-        # warn about duplicated glyphs
+    def out_dated_warning(self, outdated=False):
+        # called with outdated == True when some glyphs present in the font.json glyph list are not present in the svg font file
+
+        update_glyphlist_warning = _(
+            "The glyphlist for this font seems to be outdated.\n\n"
+            "Please update the glyph list for {font_name}:\n"
+            "* Open Extensions > Ink/Stitch > Font Management > Edit JSON\n"
+            "* Select this font and apply."
+        ).format(font_name=self.font.marked_custom_font_name)
+
+        # warning in case of duplicates in the glyph list of the font.json file
         if len(set(self.font.available_glyphs)) != len(self.font.available_glyphs):
-            duplicated_glyphs = " ".join(
-                [glyph for glyph in set(self.font.available_glyphs) if self.font.available_glyphs.count(glyph) > 1]
-            )
-            errormsg(_("Found duplicated glyphs in font file: {duplicated_glyphs}").format(duplicated_glyphs=duplicated_glyphs))
+            outdated = True
+
+        # this will cause a warning if some glyphs of the svg font are not present in the font.json glyph list
+        if len(set(self.font.available_glyphs)) != len(self.font_variant.glyphs):
+            outdated = True
+
+        if outdated:
+            errormsg(update_glyphlist_warning)
 
     def _render_text(self, text):
         lines = text.splitlines()
         position = {'x': 0, 'y': 0}
         for line in lines:
             group = Group()
-            group.label = line
+            label = ""
+            # make the label of the group line clearly show the non spacing marks
+            for character in line:
+                if unicodedata.category(character) != 'Mn':
+                    label += character
+                else:
+                    label += ' ' + character
+            group.label = label
             group.set("inkstitch:letter-group", "line")
             glyphs = []
             skip = []
@@ -297,7 +303,12 @@ class FontSampleFrame(wx.Frame):
         # because this is not unique it will be overwritten by inkscape when inserted into the document
         node.set("id", "glyph")
         node.set("inkstitch:letter-group", "glyph")
+        # force inkscape to show a label when the glyph is only a non spacing mark
+        if len(node.label) == 1 and unicodedata.category(node.label) == 'Mn':
+            node.label = ' ' + node.label
+
         group.add(node)
+
         return position
 
     def cancel(self, event):
diff --git a/lib/lettering/font.py b/lib/lettering/font.py
index 5ae6743a..1d9f8b40 100644
--- a/lib/lettering/font.py
+++ b/lib/lettering/font.py
@@ -452,7 +452,10 @@ class Font(object):
         # because this is not unique it will be overwritten by inkscape when inserted into the document
         node.set("id", "glyph")
         node.set("inkstitch:letter-group", "glyph")
-
+        # force inkscape to show a label when the glyph is only a non-spacing mark
+        if len(node.label) == 1 and unicodedata.category(node.label) == 'Mn':
+            # force inkscape to show a label when the glyph is only a non-spacing mark
+            node.label = ' ' + node.label
         return node
 
     def _update_commands(self, node, glyph, id_extension=""):
diff --git a/lib/lettering/font_variant.py b/lib/lettering/font_variant.py
index 80d9d3e6..0e4836e8 100644
--- a/lib/lettering/font_variant.py
+++ b/lib/lettering/font_variant.py
@@ -198,7 +198,7 @@ class FontVariant(object):
                 shape = 'fina'
             else:
                 shape = 'isol'
-        # in the middle of the actual word, the shape of a glyph is medi if previous glyph is bendinng,  init otherwise
+        # in the middle of the actual word, the shape of a glyph is medi if previous glyph is binding,  init otherwise
         elif previous_is_binding:
             shape = 'medi'
         else:
@@ -207,9 +207,10 @@ class FontVariant(object):
         return shape
 
     def get_next_glyph(self, word, i, previous_is_binding):
-        # search for the glyph of word that starts at i,taking  into acount the previous glyph binding status
+        # word[:i] has been processed, this function returns the glyph starting at word[i]
+        # taking into acount the previous glyph binding status
 
-        # find  all the glyphs in tthe font that start with first letter of the glyph
+        # find  all the glyphs in the font that start with first letter of the glyph
         glyph_selection = self.glyphs_start_with(word[i])
 
         # find the longest glyph that match