Python 3.3 port

12 files changed, 53 insertions, 39 deletions

diff --git a/ChangeLog b/ChangeLog
index a35e6ae..3113fab 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,6 @@
+2012-12-04 László Németh <nemeth@numbertext.org>:
+        - Python 3.3 port
+
 2012-11-23 László Németh <nemeth@numbertext.org>:
         - fix strange false positives resulted by a synchronization problem (missing
           alternatives with unloaded dictionary), the problem reported by Olivier Ronez
diff --git a/Dialog.py b/Dialog.py
index b652618..a405cc9 100644
--- a/Dialog.py
+++ b/Dialog.py
@@ -1,14 +1,14 @@
 # -*- encoding: UTF-8 -*-
+from __future__ import unicode_literals
 import sys
 import re
-from string import split
 import os
 import codecs
 
-comment = re.compile(ur"[\n#]")
-ids = re.compile(ur"\w+:\s*\*?\w+(,\s*\*?\w+)*")
-langu = re.compile(ur"\[.+=.+\]\s*")
-titl = re.compile(ur"\w+\s*=\s*")
+comment = re.compile(r"[\n#]")
+ids = re.compile(r"\w+:\s*\*?\w+(,\s*\*?\w+)*")
+langu = re.compile(r"\[.+=.+\]\s*")
+titl = re.compile(r"\w+\s*=\s*")
 helptexts = []
 
 xdl_header = """<?xml version="1.0" encoding="UTF-8"?>
@@ -30,17 +30,17 @@ oor:name="%s" oor:package="org.openoffice" xml:lang="en-US">
 </info>
 <templates>
 """
-xcs_leaf_header = ur"""
+xcs_leaf_header = r"""
                 <group oor:name="%s">
                         <info>
                                 <desc>The data for one leaf.</desc>
                         </info>
 """
-xcs_leaf = ur"""<prop oor:name="%s" oor:type="xs:string">
+xcs_leaf = r"""<prop oor:name="%s" oor:type="xs:string">
                                 <value></value>
 </prop>
 """
-xcs_leaf_footer = ur"""                </group>
+xcs_leaf_footer = r"""                </group>
 """
 xcs_component_header = """        </templates>
         <component>
@@ -128,13 +128,13 @@ def create_xdl(pkg, lines, target, lang):
             if state == 0:
                 ok = True
                 if ids.match(i.strip()):
-                    j = split(i.strip(),":")
+                    j = i.strip().split(":")
                     f2 = f2 + xdl_group%(j[0].strip(), k, k2 * 10 + 5, j[0].strip())
-                    for l in split(j[1],","):
+                    for l in j[1].split(","):
                         k = k + 1
                         k2 = k2 + 1
                         l = l.strip()
-                        la = split(l, " ")
+                        la = l.split(" ")
                         l3 = 0
                         itemlen = int(240 / len(la)) 
                         for l2 in la:
@@ -175,7 +175,7 @@ def create_xdl(pkg, lines, target, lang):
                     hlp[0] = hlp[0] + "\n"
                 f2 = f2 + hlp[0]
             elif not ok:
-                print "Syntax error in line %d: %s" %(lin, i)
+                print ("Syntax error in line %d: %s" %(lin, i))
     if "xdl" in f2n:
         f2 = f2 + xdl_footer
     target.writestr(f2n, f2)
diff --git a/Lightproof.py b/Lightproof.py
index 71d8be4..11afeab 100644
--- a/Lightproof.py
+++ b/Lightproof.py
@@ -1,6 +1,6 @@
 # -*- encoding: UTF-8 -*-
 # Lightproof grammar checker for LibreOffice and OpenOffice.org
-# 2009-2012 (c) László Németh (nemeth at numbertext org), license: MPL 1.1 / GPLv3+ / LGPLv3+
+# 2009-2012 (c) Laszlo Nemeth (nemeth at numbertext org), license: MPL 1.1 / GPLv3+ / LGPLv3+
 
 import uno, unohelper, os, sys, traceback
 from lightproof_impl_${implname} import locales
diff --git a/make.py b/make.py
index f836e8a..c9381f3 100644
--- a/make.py
+++ b/make.py
@@ -1,6 +1,13 @@
 # -*- encoding: UTF-8 -*-
+#from __future__ import unicode_literals
 import sys, os, zipfile, traceback, Dialog
-import ConfigParser as cp    # configparser in Python 3
+try:
+    import ConfigParser as cp    # configparser in Python 3
+    uni = lambda i: unicode(i, "UTF-8")
+except:
+    import configparser as cp
+    uni = str
+
 import pythonpath.lightproof_compile___implname__
 from string import Template
 
@@ -10,14 +17,14 @@ def dist(fn, a):
     distname = a['implname'] + "-" + a['version'] + '.oxt'
     z = zipfile.ZipFile(distname, mode='w', compression = zipfile.ZIP_DEFLATED)
     f = open(fn + ".dat",'r')
-    code = pythonpath.lightproof_compile___implname__.c(unicode(f.read(), "UTF-8"), a['lang'])
+    code = pythonpath.lightproof_compile___implname__.c(uni(f.read()), a['lang'])
     a["code"] = code["code"]
     a['data'] = code["rules"]
 
     for i in ["META-INF/manifest.xml", "description.xml", "Linguistic.xcu", "Lightproof.py", \
         "pythonpath/lightproof_handler___implname__.py", "pythonpath/lightproof_impl___implname__.py", \
         "pythonpath/lightproof___implname__.py" ]:
-        z.writestr(i.replace("__implname__", a["implname"]), Template(open(i, "r").read()).safe_substitute(a))
+        z.writestr(i.replace("__implname__", a["implname"]), Template(uni(open(i, "r").read())).safe_substitute(a))
 
     for i in a["extras"].split(","):
         z.writestr(i.strip().replace("../", "").replace("__implname__", a["implname"]), \
@@ -30,8 +37,8 @@ def dist(fn, a):
         z.writestr("pythonpath/lightproof_opts_%s.py"%a["implname"], "")
 
 if len(sys.argv) == 1:
-    print """Synopsis: python make.py config_file
-eg. python make.py src/en/en.cfg"""
+    print ("""Synopsis: python make.py config_file
+eg. python make.py src/en/en.cfg""")
     sys.exit(0)
 
 fArgs = cp.SafeConfigParser()
@@ -40,6 +47,6 @@ for i in sys.argv[1:]:
         fArgs.read(i)
         dist(i[:-4], fArgs._sections['args'])
     except:
-        print traceback.format_exc()
-        print "missing config file or options: ", i
+        print (traceback.format_exc())
+        print ("missing config file or options: " + str(i))
         sys.exit(0)
diff --git a/pythonpath/lightproof___implname__.py b/pythonpath/lightproof___implname__.py
index c1fe0ce..a0699be 100644
--- a/pythonpath/lightproof___implname__.py
+++ b/pythonpath/lightproof___implname__.py
@@ -1,3 +1,4 @@
 # -*- encoding: UTF-8 -*-
+from __future__ import unicode_literals
 dic = ${data}
 
diff --git a/pythonpath/lightproof_compile___implname__.py b/pythonpath/lightproof_compile___implname__.py
index 3e9e598..c3d1e53 100644
--- a/pythonpath/lightproof_compile___implname__.py
+++ b/pythonpath/lightproof_compile___implname__.py
@@ -1,5 +1,5 @@
+from __future__ import unicode_literals
 import sys, re, traceback
-from string import split
 
 repl = {}
 tests = []
@@ -105,9 +105,9 @@ def mysplit(s, line, oldline, debug):
         # modes
         if mode == "[Word]" or mode == "[word]":
             if s1[0] == '^':
-                s1 = ur"((?<=[!?.] )|^)" + s1[1:] + ur"(?![-\w\u2013\u00AD])"
+                s1 = r"((?<=[!?.] )|^)" + s1[1:] + r"(?![-\w\u2013\u00AD])"
             else:
-                s1 = ur"(?<![-\w\u2013.,\u00AD])" + s1 + ur"(?![-\w\u2013\u00AD])"
+                s1 = r"(?<![-\w\u2013.,\u00AD])" + s1 + r"(?![-\w\u2013\u00AD])"
         # modes with casing ([Word] or [Char])
         if mode == "[Word]" or mode == "[Char]":
             s1 = "(?i)" + s1
@@ -274,7 +274,7 @@ def c(rules, lang, debug = False):
                 dic = dic + [item]
 
     result["rules"] = dic
-    code = re.sub(r"(?<![\\])\\n", "\n", code.encode("unicode-escape"))
+    code = re.sub(r"(?<![\\])\\n", "\n", code.encode("unicode-escape").decode('ascii'))
 
     result["code"] = re.sub(r"(?<![\\])\\t", "\t", code).replace(r"\\n", r"\n").replace(r"\\t", r"\t")
 
diff --git a/pythonpath/lightproof_impl___implname__.py b/pythonpath/lightproof_impl___implname__.py
index 11996bd..bab4e9d 100644
--- a/pythonpath/lightproof_impl___implname__.py
+++ b/pythonpath/lightproof_impl___implname__.py
@@ -1,6 +1,6 @@
 # -*- encoding: UTF-8 -*-
+from __future__ import unicode_literals
 import uno, re, sys, os, traceback
-from string import join
 from com.sun.star.text.TextMarkupType import PROOFREADING
 from com.sun.star.beans import PropertyValue
 
@@ -121,7 +121,7 @@ def suggest(rLoc, word):
         if not x:
             return word
         t = x.getAlternatives()
-        suggestions[word] = join(t, "|")
+        suggestions[word] = "|".join(t)
     return suggestions[word]
 
 # get the nth word of the input string or None
@@ -218,7 +218,7 @@ def compile_rules(dic):
             i[0] = re.compile(i[0])
         except:
             if 'PYUNO_LOGLEVEL' in os.environ:
-                print("Lightproof: bad regular expression: ", traceback.format_exc())
+                print("Lightproof: bad regular expression: " + str(traceback.format_exc()))
             i[0] = None
 
 def get_rule(loc):
diff --git a/src/en/en.cfg b/src/en/en.cfg
index 0f038cb..6169224 100644
--- a/src/en/en.cfg
+++ b/src/en/en.cfg
@@ -3,7 +3,7 @@ lang = en
 locales = en_GB en_US en_PH en_ZA en_NA en_ZW en_AU en_CA en_IE en_IN en_BZ en_BS en_GH en_JM en_NZ en_TT
 name = Lightproof grammar checker (English)
 version = 0.4.3
-author = László Németh
+author = Laszlo Nemeth
 provider = LibreOffice
 implname = lightproof_en
 link = http://www.libreoffice.org
diff --git a/src/en/en.dat b/src/en/en.dat
index 95b0d68..789a405 100644
--- a/src/en/en.dat
+++ b/src/en/en.dat
@@ -58,17 +58,17 @@ etc [-–'’\w]*
 
 {a}n{_}{vow}{etc} <- {vow} in aA or {vow}.lower() in aA -> {a}{_}{vow}{etc} # Did you mean: \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
 
-a{_}{vow}{etc} <- ({vow} <> {vow}.upper()) and not ({vow} in aA or
+a{_}{vow}{etc} <- ({vow} != {vow}.upper()) and not ({vow} in aA or
 	{vow}.lower() in aA) and spell({vow}) -> an{_}{vow}{etc} # Bad article? \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
 
 a{_}{con}{etc} <- {con} in aAN or {con}.lower() in aAN -> an{_}{con}{etc} # Did you mean: \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
 
-{a}n{_}{con}{etc} <- ({con} <> {con}.upper()) and not ({con} in aA or
+{a}n{_}{con}{etc} <- ({con} != {con}.upper()) and not ({con} in aA or
 	{con}.lower() in aAN) and not {con} in aB and spell({con}) -> {a}{_}{con}{etc} # Bad article? \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
 
 # rules for sentences beginning with "A"
 
-^A{_}{vow}{etc} <- ({vow} <> {vow}.upper()) and not ({vow} in aA or
+^A{_}{vow}{etc} <- ({vow} != {vow}.upper()) and not ({vow} in aA or
 	{vow}.lower() in aA) and spell({vow}) -> An{_}{vow}{etc} # Bad article? \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
 
 ^A{_}{con}{etc} <- {con} in aAN or {con}.lower() in aAN -> An{_}{con}{etc} # Did you mean: \n http://en.wikipedia.org/wiki/English_articles#Distinction_between_a_and_an
@@ -379,5 +379,5 @@ def measurement(mnum, min, mout, mstr, decimal, remove):
     m = calc("CONVERT_ADD", (float(eval(mnum.replace(remove, "").replace(decimal, ".").replace(u"−", "-"))), min, mout))
     a = list(set([str(calc("ROUND", (m, 0)))[:-2], str(calc("ROUND", (m, 1))), str(calc("ROUND", (m, 2))), str(m)])) # remove duplicated rounded items
     a.sort(lambda x, y: len(x) - len(y)) # sort by string length
-    return join(a, mstr + "\n").replace(".", decimal).replace("-", u"−") + mstr
+    return (mstr + "\n").join(a).replace(".", decimal).replace("-", u"−") + mstr
 
diff --git a/src/hu_HU/ChangeLog b/src/hu_HU/ChangeLog
index a93dab9..e469b0f 100644
--- a/src/hu_HU/ChangeLog
+++ b/src/hu_HU/ChangeLog
@@ -1,3 +1,6 @@
+2012-12-04 László Németh <nemeth@numbertext.org>:
+- Python 3.3 támogatás
+
 2012-08-23  László Németh <nemeth@numbertext.org>:
 - „a 1,5” („a másfél”) nem névelőhasználati hiba. Pénzes Dávid hívta fel a
   figyelmet a problémára.
diff --git a/src/hu_HU/hu_HU.cfg b/src/hu_HU/hu_HU.cfg
index b64ed39..f73efa9 100644
--- a/src/hu_HU/hu_HU.cfg
+++ b/src/hu_HU/hu_HU.cfg
@@ -3,7 +3,7 @@ lang = hu_HU
 locales = hu_HU
 name = Lightproof grammar checker (magyar)
 version = 1.4.4
-author = László Németh
+author = Laszlo Nemeth
 provider = FSF.hu
 implname = lightproof_hu
 link = http://www.fsf.hu
diff --git a/src/hu_HU/hu_HU.dat b/src/hu_HU/hu_HU.dat
index b0d3a46..ae623ca 100644
--- a/src/hu_HU/hu_HU.dat
+++ b/src/hu_HU/hu_HU.dat
@@ -30,7 +30,7 @@ hogy-hogy -> hogyhogy # Egybeírás.
 ###################### nagybetűsítés #################
 [code]
 
-abbrev=re.compile(ur"(?i)\b([a-zöüóőúéáűíÖÜÓŐÚÉÁŰÍ]|Áe|Áht|AkH|al|ált|ápr|aug|Avtv|bek|Bp|br|bt|Btk|cca|ci(i|ii|v|x)?|cl(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii|xxx|xxxi|xxxii|xxxiii|xxxiv|xxxix|xxxv|xxxvi|xxxvii|xxxviii)?|Co|cv(i|ii|iii)?|cx(c|ci|cii|ciii|civ|cix|cv|cvi|cvii|cviii|i|ii|iii|iv|ix|l|li|lii|liii|liv|lix|lv|lvi|lvii|lviii|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii)?|cs|Csjt|Cstv|csüt|dec|dk|dny|dr|du|dz(s)?|egy|ék|ÉKsz|em|ény|Épt|érk|etc|Etv|eü|ev|évf|febr|felv|Flt|ford|főisk|fsz(la|t)?|Ftv|gimn|gör|gr|Gt|gy|Gyvt|habil|hg|hiv|Hjt|honv|Hpt|hrsz|hsz|Hszt|htb|id|ifj|ig(h)?|ii(i)?|ill|Inc|ind|isk|iv|ix|izr|jan|jegyz|júl|jún|kat|kb|Kbt|ker|kft|kgy|kht|kir|kiv|Kjt|kk(t)?|koll|korm|köv|kp|Kr|krt|Kt(v)?|ld|li(i|ii|v|x)?|Ltd|ltp|Ltv|luth|lv(i|ii|iii)?|lx(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii)?|ly|máj|márc|mat|max|mb|megh|megj|MHSz|min|mk|Mo|Mt|NB|nov|ny(á)?|Nyilv|nyrt|okl|okt|olv|op|orsz|ort|ov(h)?|össz|Ötv|özv|Pf|pl(d)?|prof|prot|Ptk|pu|ref|rk(p)?|róm|röv|rt|sgt|spec|stb|sz(ept|erk)?|Szjt|szoc|Szt(v)?|szül|Tbj|tc|tel|tkp|tszf|tvr|ty|ua|ui|úm|ún|uo|Ve|Vhr|vi(i|ii)?|vö|vsz|Vt(v)?|xc(i|ii|iii|iv|ix|v|vi|vii|viii)?|xi(i|ii|v|x)?|xl(i|ii|iii|iv|ix|v|vi|vii|viii)?|xv(i|ii|iii)?|xx(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii)?|zrt)\.")
+abbrev=re.compile(r"(?i)\b([a-zöüóőúéáűíÖÜÓŐÚÉÁŰÍ]|Áe|Áht|AkH|al|ált|ápr|aug|Avtv|bek|Bp|br|bt|Btk|cca|ci(i|ii|v|x)?|cl(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii|xxx|xxxi|xxxii|xxxiii|xxxiv|xxxix|xxxv|xxxvi|xxxvii|xxxviii)?|Co|cv(i|ii|iii)?|cx(c|ci|cii|ciii|civ|cix|cv|cvi|cvii|cviii|i|ii|iii|iv|ix|l|li|lii|liii|liv|lix|lv|lvi|lvii|lviii|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii)?|cs|Csjt|Cstv|csüt|dec|dk|dny|dr|du|dz(s)?|egy|ék|ÉKsz|em|ény|Épt|érk|etc|Etv|eü|ev|évf|febr|felv|Flt|ford|főisk|fsz(la|t)?|Ftv|gimn|gör|gr|Gt|gy|Gyvt|habil|hg|hiv|Hjt|honv|Hpt|hrsz|hsz|Hszt|htb|id|ifj|ig(h)?|ii(i)?|ill|Inc|ind|isk|iv|ix|izr|jan|jegyz|júl|jún|kat|kb|Kbt|ker|kft|kgy|kht|kir|kiv|Kjt|kk(t)?|koll|korm|köv|kp|Kr|krt|Kt(v)?|ld|li(i|ii|v|x)?|Ltd|ltp|Ltv|luth|lv(i|ii|iii)?|lx(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii|xx|xxi|xxii|xxiii|xxiv|xxix|xxv|xxvi|xxvii|xxviii)?|ly|máj|márc|mat|max|mb|megh|megj|MHSz|min|mk|Mo|Mt|NB|nov|ny(á)?|Nyilv|nyrt|okl|okt|olv|op|orsz|ort|ov(h)?|össz|Ötv|özv|Pf|pl(d)?|prof|prot|Ptk|pu|ref|rk(p)?|róm|röv|rt|sgt|spec|stb|sz(ept|erk)?|Szjt|szoc|Szt(v)?|szül|Tbj|tc|tel|tkp|tszf|tvr|ty|ua|ui|úm|ún|uo|Ve|Vhr|vi(i|ii)?|vö|vsz|Vt(v)?|xc(i|ii|iii|iv|ix|v|vi|vii|viii)?|xi(i|ii|v|x)?|xl(i|ii|iii|iv|ix|v|vi|vii|viii)?|xv(i|ii|iii)?|xx(i|ii|iii|iv|ix|v|vi|vii|viii|x|xi|xii|xiii|xiv|xix|xv|xvi|xvii|xviii)?|zrt)\.")
 
 # pattern for paragraph checking
 paralcap = re.compile(u"(?u)^[a-zöüóőúéáűí].*[.?!] [A-ZÖÜÓŐÚÉÁŰÍ].*[.?!][)”]?$")
@@ -139,7 +139,7 @@ b {abc}*
 	calc("NUMBERTEXT", ("1", "hu")) and
 	calc("NUMBERTEXT", (re.sub(u"[   .]", "", \1), "hu")).replace(u"kettő", u"két").replace(u"ezeregyszáz", u"ezerszáz") !=
 	# változatok: száz vagy egyszáz, ezer vagy egyezer, kettő vagy két
-	re.sub(ur"\begy(száz|ezer)", r"\g<1>", \3).replace(u"ezeregyszáz", u"ezerszáz").replace(u"kettő", u"két")
+	re.sub(r"\begy(száz|ezer)", r"\g<1>", \3).replace(u"ezeregyszáz", u"ezerszáz").replace(u"kettő", u"két")
 	-> = \1 + \2 + calc("NUMBERTEXT", (re.sub(u"[   .]", "", \1), "hu")) +  \4
 	# A két összeg nem egyezik.
 
@@ -379,7 +379,7 @@ fő(irány\w*|közlekedés\w*|szabály\w*|szervező\w*) -> fő \1 # A „fő mű
 hátba (támadás\w*) -> hátba\1			# Egybeírás. 
 időről-időre -> időről időre			# Különírás.
 jing és jang -> jin és jang			# Helyesen jin és jang.
-kőr((öz|út|ut)\w*) <- morph(\0, ur"st:kőr\b") -> kör\1 # Rövid ö-vel a kör összetételeiben.
+kőr((öz|út|ut)\w*) <- morph(\0, r"st:kőr\b") -> kör\1 # Rövid ö-vel a kör összetételeiben.
 légyszíves -> légy szíves|legyél szíves	# Legyél szíves értelemben különírjuk.
 legalább is -> legalábbis			# Egybeírás.
 Lichten(stein\w*) -> Liechten\1			# Helyesen Liechtenstein.
@@ -414,7 +414,7 @@ aword [aáeéiíoóöőuúüűAÁEÉIÍOÓÖŐUÚÜŰ]\w*
 a(?! 1,5\b) <- word(1) and re.match(r"(1|5\d*|[15]\d\d\d)[.]?$", word(1)) -> az # Hibás névelő.
 
 az <- not option("grammar") and word(2) and
-	re.match(ur"(alfejezet|alszakasz|ábr|bekezdés|diagram|fejezet|kép|lap|oldal|paragrafus|szakasz|táblázat)\w*", word(2)) and
+	re.match(r"(alfejezet|alszakasz|ábr|bekezdés|diagram|fejezet|kép|lap|oldal|paragrafus|szakasz|táblázat)\w*", word(2)) and
 	re.match(r"([02-46-9]|[1-46-9]\d|[1-46-9]\d\d|[2346-9]\d\d\d)[.]", word(1)) -> a # Hibás névelő?
 
 az <- option("grammar") and word(1) and
@@ -758,7 +758,7 @@ W \w{4,5}
 # rövid (kétbetűs) tagot tartalmazó összetett szó
 
 W \w{4,}
-{W} <- option("compound") and morph({W}, ur"(?u)pa:\w\w\b") -> =suggest({W}) # Biztos, hogy helyes összetett szó?
+{W} <- option("compound") and morph({W}, r"(?u)pa:\w\w\b") -> =suggest({W}) # Biztos, hogy helyes összetett szó?
 
 # minden képzett összetett szó
 
@@ -912,4 +912,4 @@ def measurement(mnum, min, mout, mstr):
     m = calc("CONVERT_ADD", (float(mnum.replace(",", ".").replace(u"−", "-")), min, mout))
     a = list(set([str(calc("ROUND", (m, 0)))[:-2], str(calc("ROUND", (m, 1))), str(calc("ROUND", (m, 2))), str(m)])) # remove duplicated rounded items
     a.sort(lambda x, y: len(x) - len(y)) # sort by string length
-    return join(a, mstr + "|").replace(".", ",").replace("-", u"−") + mstr
+    return (mstr + "|").join(a).replace(".", ",").replace("-", u"−") + mstr