lyx2lyx to convert the language of spaces surrounding text in a foriegn

language to be in the paragraph's language. This is necessary due to format
change 259, but the language was forgotten in lyx2lyx back then. It's important
for Bidi documents.

(fixes bug http://bugzilla.lyx.org/show_bug.cgi?id=3789)


git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@18766 a592a061-630c-0410-9148-cb99ea01b6c8

development/FORMAT

@@ -1,5 +1,12 @@
 LyX file-format changes
 -----------------------
+
+2007-06-13 Dov Feldstern <dov@lyx.org>
+	* format incremented to 274: applying the conversion done in format 259
+		to the \lang property, which was forgotten back then... This is
+		slightly more complicated, because the default language has to be
+		determined on a per-paragraph basis.
+
 2007-06-13 Bo Peng <ben.bob@gmail.com>
 	* format incremented to 273: add --Separator-- environment to 
 		separate consecutive environments

lib/lyx2lyx/LyX.py

@@ -77,7 +77,7 @@ format_relation = [("0_06",    [200], generate_minor_versions("0.6" , 4)),
                    ("1_2",     [220], generate_minor_versions("1.2" , 4)),
                    ("1_3",     [221], generate_minor_versions("1.3" , 7)),
                    ("1_4", range(222,246), generate_minor_versions("1.4" , 4)),
-                   ("1_5", range(246,274), generate_minor_versions("1.5" , 0))]
+                   ("1_5", range(246,275), generate_minor_versions("1.5" , 0))]
 
 
 def formats_list():

lib/lyx2lyx/lyx_1_5.py

@@ -22,7 +22,7 @@
 import re
 import unicodedata
 
-from parser_tools import find_re, find_token, find_token_backwards, find_token_exact, find_tokens, find_end_of, get_value
+from parser_tools import find_re, find_token, find_token_backwards, find_token_exact, find_tokens, find_end_of, get_value, find_beginning_of, find_nonempty_line
 from LyX import get_encoding
 
 
@@ -37,6 +37,10 @@ def find_end_of_layout(lines, i):
     " Find end of layout, where lines[i] is included."
     return find_end_of(lines, i, "\\begin_layout", "\\end_layout")
 
+def find_beginning_of_layout(lines, i):
+    "Find beginning of layout, where lines[i] is included."
+    return find_beginning_of(lines, i, "\\begin_layout", "\\end_layout")
+
 # End of helper functions
 ####################################################################
 
@@ -1099,22 +1103,59 @@ def revert_accent(document):
         document.body[i] = unicodedata.normalize("NFKC", document.body[i])
 
 
-def normalize_font_whitespace(document):
+def normalize_font_whitespace_259(document):
     """ Before format 259 the font changes were ignored if a
     whitespace was the first or last character in the sequence, this function
     transfers the whitespace outside."""
-
-    if document.backend != "latex":
-        return
-
-    lines = document.body
-
+	
     char_properties = {"\\series": "default",
                        "\\emph": "default",
                        "\\color": "none",
                        "\\shape": "default",
                        "\\bar": "default",
                        "\\family": "default"}
+    return normalize_font_whitespace(document, char_properties)
+
+def normalize_font_whitespace_274(document):
+    """ Before format 259 (sic) the font changes were ignored if a
+    whitespace was the first or last character in the sequence. This was 
+    corrected for most font properties in format 259, but the language 
+    was forgotten then. This function applies the same conversion done
+    there (namely, transfers the whitespace outside) for font language
+    changes, as well."""
+
+    char_properties = {"\\lang": "default"}
+    return normalize_font_whitespace(document, char_properties)
+
+def get_paragraph_language(document, i):
+    """ Return the language of the paragraph in which line i of the document
+    body is. If the first thing in the paragraph is a \\lang command, that
+    is the paragraph's langauge; otherwise, the paragraph's language is the 
+    document's language."""
+
+    lines = document.body
+	
+    first_nonempty_line = \
+        find_nonempty_line(lines, find_beginning_of_layout(lines, i) + 1)
+
+    words = lines[first_nonempty_line].split()
+
+    if len(words) > 1 and words[0] == "\\lang":
+        return words[1]
+    else:
+        return document.language
+	
+def normalize_font_whitespace(document, char_properties):
+    """ Before format 259 the font changes were ignored if a
+    whitespace was the first or last character in the sequence, this function
+    transfers the whitespace outside. Only a change in one of the properties
+    in the provided	char_properties is handled by this function."""
+
+    if document.backend != "latex":
+        return
+
+    lines = document.body
+
     changes = {}
 
     i = 0
@@ -1124,6 +1165,10 @@ def normalize_font_whitespace(document):
         if len(words) > 0 and words[0] == "\\begin_layout":
             # a new paragraph resets all font changes
             changes.clear()
+            # also reset the default language to be the paragraph's language
+            if "\\lang" in char_properties.keys():
+                char_properties["\\lang"] = \
+                    get_paragraph_language(document, i + 1)
 
         elif len(words) > 1 and words[0] in char_properties.keys():
             # we have a font change
@@ -1722,7 +1767,7 @@ convert = [[246, []],
            [256, []],
            [257, [convert_caption]],
            [258, [convert_lyxline]],
-           [259, [convert_accent, normalize_font_whitespace]],
+           [259, [convert_accent, normalize_font_whitespace_259]],
            [260, []],
            [261, [convert_changes]],
            [262, []],
@@ -1737,9 +1782,11 @@ convert = [[246, []],
            [271, [convert_ext_font_sizes]],
            [272, []],
            [273, []],
+           [274, [normalize_font_whitespace_274]]
           ]
 
 revert =  [
+           [273, []],
            [272, [revert_separator_layout]],
            [271, [revert_preamble_listings_params, revert_listings_inset, revert_include_listings]],
            [270, [revert_ext_font_sizes]],

src/Buffer.cpp

@@ -142,7 +142,7 @@ using std::string;
 
 namespace {
 
-int const LYX_FORMAT = 273;
+int const LYX_FORMAT = 274;
 
 } // namespace anon