From 9a0d70a45fde2d2a5d7af84f39f91e3c2ea91dff Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jos=C3=A9=20Matos?= <jamatos@lyx.org>
Date: Wed, 11 Mar 2015 12:04:46 +0000
Subject: [PATCH] First step to make lyx2lyx code support both python 2 and
 python 3 (3.3+)

Most of the changes are related with dictionaries returning views instead of
lists.

xrange -> range (since xrange is gone in python 3)

The code that is special to support both python 2 and 3 is enclosed in a comment
 \# Provide support for both python 2 and 3
and
 \# End of code to support for both python 2 and 3

And so later it can be removed safely when python 2 is no longer supported.
---
 lib/lyx2lyx/LyX.py                    | 12 +++++------
 lib/lyx2lyx/generate_encoding_info.py |  9 ++++----
 lib/lyx2lyx/lyx2lyx_tools.py          |  2 +-
 lib/lyx2lyx/lyx_1_2.py                |  6 +++---
 lib/lyx2lyx/lyx_1_4.py                |  4 ++--
 lib/lyx2lyx/lyx_1_5.py                | 31 +++++++++++++++++----------
 lib/lyx2lyx/lyx_1_6.py                |  4 ++--
 lib/lyx2lyx/lyx_2_1.py                | 24 ++++++++++-----------
 lib/lyx2lyx/lyx_2_2.py                |  4 ++--
 lib/lyx2lyx/parser_tools.py           | 20 ++++++++---------
 lib/lyx2lyx/unicode_symbols.py        |  6 ++++++
 11 files changed, 69 insertions(+), 53 deletions(-)

diff --git a/lib/lyx2lyx/LyX.py b/lib/lyx2lyx/LyX.py
index fdfa34d0a1..89b08c8d28 100644
--- a/lib/lyx2lyx/LyX.py
+++ b/lib/lyx2lyx/LyX.py
@@ -80,12 +80,12 @@ format_relation = [("0_06",    [200], minor_versions("0.6" , 4)),
                    ("1_1_6_3", [218], ["1.1", "1.1.6.3","1.1.6.4"]),
                    ("1_2",     [220], minor_versions("1.2" , 4)),
                    ("1_3",     [221], minor_versions("1.3" , 7)),
-                   ("1_4", range(222,246), minor_versions("1.4" , 5)),
-                   ("1_5", range(246,277), minor_versions("1.5" , 7)),
-                   ("1_6", range(277,346), minor_versions("1.6" , 10)),
-                   ("2_0", range(346,414), minor_versions("2.0", 8)),
-                   ("2_1", range(414,475), minor_versions("2.1", 0)),
-                   ("2_2", range(475,483), minor_versions("2.2", 0))
+                   ("1_4", list(range(222,246)), minor_versions("1.4" , 5)),
+                   ("1_5", list(range(246,277)), minor_versions("1.5" , 7)),
+                   ("1_6", list(range(277,346)), minor_versions("1.6" , 10)),
+                   ("2_0", list(range(346,414)), minor_versions("2.0", 8)),
+                   ("2_1", list(range(414,475)), minor_versions("2.1", 0)),
+                   ("2_2", list(range(475,483)), minor_versions("2.2", 0))
                   ]
 
 ####################################################################
diff --git a/lib/lyx2lyx/generate_encoding_info.py b/lib/lyx2lyx/generate_encoding_info.py
index 6766ef0520..50725b7662 100644
--- a/lib/lyx2lyx/generate_encoding_info.py
+++ b/lib/lyx2lyx/generate_encoding_info.py
@@ -19,6 +19,7 @@
 """ This module parses lib/languages and prints it as a python
 dictionary, ready to use by other python modules"""
 
+from __future__ import print_function
 import pprint
 
 def parse_line(line):
@@ -55,8 +56,8 @@ if __name__ == '__main__':
             lang[tmp[0]] = tmp[1:]
 
 
-    print "# This file is generated by generate_incoding_info.py from lib/languages file."
-    print "# Do not change this file directly."
-    print
-    print "lang = ",
+    print ("# This file is generated by generate_incoding_info.py from lib/languages file.")
+    print ("# Do not change this file directly.")
+    print ()
+    print ("lang = ", end = " ")
     pprint.pprint(lang)
diff --git a/lib/lyx2lyx/lyx2lyx_tools.py b/lib/lyx2lyx/lyx2lyx_tools.py
index fefc53f377..9c6943909a 100644
--- a/lib/lyx2lyx/lyx2lyx_tools.py
+++ b/lib/lyx2lyx/lyx2lyx_tools.py
@@ -283,7 +283,7 @@ def latex_length(slen):
     units = {"text%":"\\textwidth", "col%":"\\columnwidth",
              "page%":"\\paperwidth", "line%":"\\linewidth",
              "theight%":"\\textheight", "pheight%":"\\paperheight"}
-    for unit in units.keys():
+    for unit in list(units.keys()):
         i = slen.find(unit)
         if i == -1:
             continue
diff --git a/lib/lyx2lyx/lyx_1_2.py b/lib/lyx2lyx/lyx_1_2.py
index a47fcaf1a0..19b6647201 100644
--- a/lib/lyx2lyx/lyx_1_2.py
+++ b/lib/lyx2lyx/lyx_1_2.py
@@ -154,7 +154,7 @@ def remove_oldfloat(document):
         j = find_token(lines, "\\end_float", i+1)
 
         floattype = lines[i].split()[1]
-        if not floats.has_key(floattype):
+        if floattype not in floats:
             document.warning("Error! Unknown float type " + floattype)
             floattype = "fig"
 
@@ -284,7 +284,7 @@ def remove_pextra(document):
 
 def is_empty(lines):
     " Are all the lines empty?"
-    return filter(is_nonempty_line, lines) == []
+    return list(filter(is_nonempty_line, lines)) == []
 
 
 move_rexp =  re.compile(r"\\(family|series|shape|size|emph|numeric|bar|noun|end_deeper)")
@@ -358,7 +358,7 @@ def remove_oldert(document):
                     tmp.append(line)
 
             if is_empty(tmp):
-                if filter(lambda x:x != "", tmp) != []:
+                if [x for x in tmp if x != ""] != []:
                     if new == []:
                         # This is not necessary, but we want the output to be
                         # as similar as posible to the lyx format
diff --git a/lib/lyx2lyx/lyx_1_4.py b/lib/lyx2lyx/lyx_1_4.py
index 9fb583f071..c1a4591b54 100644
--- a/lib/lyx2lyx/lyx_1_4.py
+++ b/lib/lyx2lyx/lyx_1_4.py
@@ -1495,7 +1495,7 @@ def convert_len(len, special):
         len = '%f\\' % len2value(len) + special
 
     # Convert LyX units to LaTeX units
-    for unit in units.keys():
+    for unit in list(units.keys()):
         if len.find(unit) != -1:
             len = '%f' % (len2value(len) / 100) + units[unit]
             break
@@ -1571,7 +1571,7 @@ def convert_frameless_box(document):
                   'inner_pos':1, 'use_parbox':'0', 'width':'100col%',
                   'special':'none', 'height':'1in',
                   'height_special':'totalheight', 'collapsed':'false'}
-        for key in params.keys():
+        for key in list(params.keys()):
             value = get_value(document.body, key, i, j).replace('"', '')
             if value != "":
                 if key == 'position':
diff --git a/lib/lyx2lyx/lyx_1_5.py b/lib/lyx2lyx/lyx_1_5.py
index 96d9589a45..8856d0b7c6 100644
--- a/lib/lyx2lyx/lyx_1_5.py
+++ b/lib/lyx2lyx/lyx_1_5.py
@@ -26,6 +26,15 @@ import sys, os
 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
 
+# Provide support for both python 2 and 3
+PY2 = sys.version_info[0] == 2
+if not PY2:
+    text_type = str
+    unichr = chr
+else:
+    text_type = unicode
+    unichr = unichr
+# End of code to support for both python 2 and 3
 
 ####################################################################
 # Private helper functions
@@ -93,7 +102,7 @@ def convert_font_settings(document):
     if font_scheme == '':
         document.warning("Malformed LyX document: Empty `\\fontscheme'.")
         font_scheme = 'default'
-    if not font_scheme in roman_fonts.keys():
+    if not font_scheme in list(roman_fonts.keys()):
         document.warning("Malformed LyX document: Unknown `\\fontscheme' `%s'." % font_scheme)
         font_scheme = 'default'
     document.header[i:i+1] = ['\\font_roman %s' % roman_fonts[font_scheme],
@@ -163,7 +172,7 @@ def revert_font_settings(document):
         del document.header[i]
     if font_tt_scale != '100':
         document.warning("Conversion of '\\font_tt_scale' not yet implemented.")
-    for font_scheme in roman_fonts.keys():
+    for font_scheme in list(roman_fonts.keys()):
         if (roman_fonts[font_scheme] == fonts['roman'] and
             sans_fonts[font_scheme] == fonts['sans'] and
             typewriter_fonts[font_scheme] == fonts['typewriter']):
@@ -376,7 +385,7 @@ def revert_unicode_line(document, i, insets, spec_chars, replacement_character =
             last_char = character
         except:
             # Try to replace with ERT/math inset
-            if spec_chars.has_key(character):
+            if character in spec_chars:
                 command = spec_chars[character][0] # the command to replace unicode
                 flag1 = spec_chars[character][1]
                 flag2 = spec_chars[character][2]
@@ -1211,7 +1220,7 @@ def revert_accent(document):
         try:
             document.body[i] = normalize("NFD", document.body[i])
         except TypeError:
-            document.body[i] = normalize("NFD", unicode(document.body[i], 'utf-8'))
+            document.body[i] = normalize("NFD", text_type(document.body[i], 'utf-8'))
 
     # Replace accented characters with InsetLaTeXAccent
     # Do not convert characters that can be represented in the chosen
@@ -1346,15 +1355,15 @@ def normalize_font_whitespace(document, char_properties):
             # 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():
+            if "\\lang" in list(char_properties.keys()):
                 char_properties["\\lang"] = \
                     get_paragraph_language(document, i + 1)
 
-        elif len(words) > 1 and words[0] in char_properties.keys():
+        elif len(words) > 1 and words[0] in list(char_properties.keys()):
             # we have a font change
             if char_properties[words[0]] == words[1]:
                 # property gets reset
-                if words[0] in changes.keys():
+                if words[0] in list(changes.keys()):
                     del changes[words[0]]
                 defaultproperty = True
             else:
@@ -1372,11 +1381,11 @@ def normalize_font_whitespace(document, char_properties):
                 lines[i-1] = lines[i-1][:-1]
                 # a space before the font change
                 added_lines = [" "]
-                for k in changes.keys():
+                for k in list(changes.keys()):
                     # exclude property k because that is already in lines[i]
                     if k != words[0]:
                         added_lines[1:1] = ["%s %s" % (k, changes[k])]
-                for k in changes.keys():
+                for k in list(changes.keys()):
                     # exclude property k because that must be added below anyway
                     if k != words[0]:
                         added_lines[0:0] = ["%s %s" % (k, char_properties[k])]
@@ -1400,11 +1409,11 @@ def normalize_font_whitespace(document, char_properties):
                         continue
                 lines[i+1] = lines[i+1][1:]
                 added_lines = [" "]
-                for k in changes.keys():
+                for k in list(changes.keys()):
                     # exclude property k because that is already in lines[i]
                     if k != words[0]:
                         added_lines[1:1] = ["%s %s" % (k, changes[k])]
-                for k in changes.keys():
+                for k in list(changes.keys()):
                     # exclude property k because that must be added below anyway
                     if k != words[0]:
                         added_lines[0:0] = ["%s %s" % (k, char_properties[k])]
diff --git a/lib/lyx2lyx/lyx_1_6.py b/lib/lyx2lyx/lyx_1_6.py
index 2d0964c3af..a30e50e8fb 100644
--- a/lib/lyx2lyx/lyx_1_6.py
+++ b/lib/lyx2lyx/lyx_1_6.py
@@ -94,7 +94,7 @@ def convert_len(len):
              "theight%":"\\backslash\ntextheight", "pheight%":"\\backslash\npageheight"}
 
     # Convert LyX units to LaTeX units
-    for unit in units.keys():
+    for unit in list(units.keys()):
         if len.find(unit) != -1:
             len = '%f' % (len2value(len) / 100)
             len = len.strip('0') + units[unit]
@@ -1765,7 +1765,7 @@ def convert_module_names(document):
     return
   newmodlist = []
   for mod in modlist:
-    if modulemap.has_key(mod):
+    if mod in modulemap:
       newmodlist.append(modulemap[mod])
     else:
       document.warning("Can't find module %s in the module map!" % mod)
diff --git a/lib/lyx2lyx/lyx_2_1.py b/lib/lyx2lyx/lyx_2_1.py
index 798a36b9b1..f4d7398f04 100644
--- a/lib/lyx2lyx/lyx_2_1.py
+++ b/lib/lyx2lyx/lyx_2_1.py
@@ -398,7 +398,7 @@ def convert_japanese_encodings(document):
     if i == -1:
         return
     val = get_value(document.header, "\\inputencoding", i)
-    if val in jap_enc_dict.keys():
+    if val in list(jap_enc_dict.keys()):
         document.header[i] = "\\inputencoding %s" % jap_enc_dict[val]
 
 
@@ -413,7 +413,7 @@ def revert_japanese_encodings(document):
     if i == -1:
         return
     val = get_value(document.header, "\\inputencoding", i)
-    if val in jap_enc_dict.keys():
+    if val in list(jap_enc_dict.keys()):
         document.header[i] = "\\inputencoding %s" % jap_enc_dict[val]
 
 
@@ -1227,7 +1227,7 @@ def revert_mathdesign(document):
         if i == -1:
             return
         val = get_value(document.header, "\\font_roman", i)
-        if val in mathdesign_dict.keys():
+        if val in list(mathdesign_dict.keys()):
             preamble = "\\usepackage[%s" % mathdesign_dict[val]
             expert = False
             j = find_token(document.header, "\\font_osf true", 0)
@@ -1391,7 +1391,7 @@ def revert_mathfonts(document):
                 k = find_token(document.header, "\\font_osf true", 0)
                 if k != -1:
                     rm += "-osf"
-                if rm in mathfont_dict.keys():
+                if rm in list(mathfont_dict.keys()):
                     add_to_preamble(document, mathfont_dict[rm])
                     document.header[j] = "\\font_roman default"
                     if k != -1:
@@ -1412,7 +1412,7 @@ def revert_mdnomath(document):
         if i == -1:
             return
         val = get_value(document.header, "\\font_roman", i)
-        if val in mathdesign_dict.keys():
+        if val in list(mathdesign_dict.keys()):
             j = find_token(document.header, "\\font_math", 0)
             if j == -1:
                 document.header[i] = "\\font_roman %s" % mathdesign_dict[val]
@@ -1437,7 +1437,7 @@ def convert_mdnomath(document):
         if i == -1:
             return
         val = get_value(document.header, "\\font_roman", i)
-        if val in mathdesign_dict.keys():
+        if val in list(mathdesign_dict.keys()):
              document.header[i] = "\\font_roman %s" % mathdesign_dict[val]
 
 
@@ -1454,7 +1454,7 @@ def revert_newtxmath(document):
         "minion-ntxm":  "\\usepackage[minion]{newtxmath}",
         "newtxmath":  "\\usepackage{newtxmath}",
         }
-        if val in mathfont_dict.keys():
+        if val in list(mathfont_dict.keys()):
             add_to_preamble(document, mathfont_dict[val])
             document.header[i] = "\\font_math auto"
 
@@ -3630,7 +3630,7 @@ def convert_captionlayouts(document):
         if i == -1:
             return
         val = get_value(document.body, "\\begin_layout", i)
-        if val in caption_dict.keys():
+        if val in list(caption_dict.keys()):
             j = find_end_of_layout(document.body, i)
             if j == -1:
                 document.warning("Malformed LyX document: Missing `\\end_layout'.")
@@ -3666,7 +3666,7 @@ def revert_captionlayouts(document):
         val = ""
         if m:
             val = m.group(1)
-        if val not in caption_dict.keys():
+        if val not in list(caption_dict.keys()):
             i += 1
             continue
         
@@ -3847,7 +3847,7 @@ def revert_newframes(document):
         val = ""
         if m:
             val = m.group(1)
-        if val not in frame_dict.keys():
+        if val not in list(frame_dict.keys()):
             i += 1
             continue
         # Find end of sequence
@@ -3963,7 +3963,7 @@ def convert_encodings(document):
     if i == -1:
         return
     val = get_value(document.header, "\\inputencoding", i)
-    if val in LaTeX2LyX_enc_dict.keys():
+    if val in list(LaTeX2LyX_enc_dict.keys()):
         document.header[i] = "\\inputencoding %s" % LaTeX2LyX_enc_dict[val]
     elif val not in known_enc_tuple:
         document.warning("Ignoring unknown input encoding: `%s'" % val)
@@ -4004,7 +4004,7 @@ def revert_encodings(document):
     if i == -1:
         return
     val = get_value(document.header, "\\inputencoding", i)
-    if val in LyX2LaTeX_enc_dict.keys():
+    if val in list(LyX2LaTeX_enc_dict.keys()):
         document.header[i] = "\\inputencoding %s" % LyX2LaTeX_enc_dict[val]
     elif val not in known_enc_tuple:
         document.warning("Ignoring unknown input encoding: `%s'" % val)
diff --git a/lib/lyx2lyx/lyx_2_2.py b/lib/lyx2lyx/lyx_2_2.py
index 82f12e26cb..b4ce71e535 100644
--- a/lib/lyx2lyx/lyx_2_2.py
+++ b/lib/lyx2lyx/lyx_2_2.py
@@ -74,7 +74,7 @@ def convert_separator(document):
             lay = get_containing_layout(document.body, j-1)
             if lay != False:
                 content = "\n".join(document.body[lay[1]:lay[2]])
-                for val in sty_dict.keys():
+                for val in list(sty_dict.keys()):
                     if content.find("\\%s" % val) != -1:
                         document.body[j:j] = ["\\%s %s" % (val, sty_dict[val])]
                         i = i + 1
@@ -103,7 +103,7 @@ def convert_separator(document):
                and find_token(document.body, "\\begin_inset VSpace", lay[1], lay[2]) == -1:
                 # reset any text style before inserting the inset
                 content = "\n".join(document.body[lay[1]:lay[2]])
-                for val in sty_dict.keys():
+                for val in list(sty_dict.keys()):
                     if content.find("\\%s" % val) != -1:
                         document.body[j:j] = ["\\%s %s" % (val, sty_dict[val])]
                         i = i + 1
diff --git a/lib/lyx2lyx/parser_tools.py b/lib/lyx2lyx/parser_tools.py
index 4e30f63bf8..85e2b6ae8c 100644
--- a/lib/lyx2lyx/parser_tools.py
+++ b/lib/lyx2lyx/parser_tools.py
@@ -56,9 +56,9 @@ find_re(lines, rexp, start[, end]):
 get_value(lines, token, start[, end[, default]):
   Similar to find_token, but it returns what follows the 
   token on the found line. Example:
-    get_value(document.header, "\use_xetex", 0)
+    get_value(document.header, "\\use_xetex", 0)
   will find a line like:
-    \use_xetex true
+    \\use_xetex true
   and, in that case, return "true". (Note that whitespace
   is stripped.) The final argument, default, defaults to "", 
   and is what is returned if we do not find anything. So you
@@ -80,10 +80,10 @@ del_token(lines, token, start[, end]):
 
 find_beginning_of(lines, i, start_token, end_token):
   Here, start_token and end_token are meant to be a matching 
-  pair, like "\begin_layout" and "\end_layout". We look for 
+  pair, like "\\begin_layout" and "\\end_layout". We look for 
   the start_token that pairs with the end_token that occurs
   on or after line i. Returns -1 if not found.
-  So, in the layout case, this would find the \begin_layout 
+  So, in the layout case, this would find the \\begin_layout 
   for the layout line i is in. 
   Example:
     ec = find_token(document.body, "</cell", i)
@@ -187,7 +187,7 @@ def find_token(lines, token, start, end = 0, ignorews = False):
     if end == 0 or end > len(lines):
         end = len(lines)
     m = len(token)
-    for i in xrange(start, end):
+    for i in range(start, end):
         if ignorews:
             x = lines[i].split()
             y = token.split()
@@ -215,7 +215,7 @@ def find_tokens(lines, tokens, start, end = 0, ignorews = False):
     if end == 0 or end > len(lines):
         end = len(lines)
 
-    for i in xrange(start, end):
+    for i in range(start, end):
         for token in tokens:
             if ignorews:
                 x = lines[i].split()
@@ -244,7 +244,7 @@ def find_re(lines, rexp, start, end = 0):
 
     if end == 0 or end > len(lines):
         end = len(lines)
-    for i in xrange(start, end):
+    for i in range(start, end):
         if rexp.match(lines[i]):
                 return i
     return -1
@@ -258,7 +258,7 @@ def find_token_backwards(lines, token, start):
 
     Return -1 on failure."""
     m = len(token)
-    for i in xrange(start, -1, -1):
+    for i in range(start, -1, -1):
         line = lines[i]
         if line[:m] == token:
             return i
@@ -272,7 +272,7 @@ def find_tokens_backwards(lines, tokens, start):
     element, in lines[end, start].
 
     Return -1 on failure."""
-    for i in xrange(start, -1, -1):
+    for i in range(start, -1, -1):
         line = lines[i]
         for token in tokens:
             if line[:len(token)] == token:
@@ -381,7 +381,7 @@ def find_end_of(lines, i, start_token, end_token):
 def find_nonempty_line(lines, start, end = 0):
     if end == 0:
         end = len(lines)
-    for i in xrange(start, end):
+    for i in range(start, end):
         if is_nonempty_line(lines[i]):
             return i
     return -1
diff --git a/lib/lyx2lyx/unicode_symbols.py b/lib/lyx2lyx/unicode_symbols.py
index 9df65e83f7..9171c180ef 100644
--- a/lib/lyx2lyx/unicode_symbols.py
+++ b/lib/lyx2lyx/unicode_symbols.py
@@ -20,6 +20,12 @@
 
 import sys, os, re
 
+# Provide support for both python 2 and 3
+PY2 = sys.version_info[0] == 2
+if not PY2:
+    unichr = chr
+# End of code to support for both python 2 and 3
+
 def read_unicodesymbols():
     " Read the unicodesymbols list of unicode characters and corresponding commands."
     pathname = os.path.abspath(os.path.dirname(sys.argv[0]))