Clean up in CLDR scripts for QLocale

Use python more fluently; DRY - use functions to avoid repetition; use
dict-comprehensions; use os.path.join() in preference to arithmetic
with path strings; use elsif to avoid the need for a local variable;
set() can take a generator directly, no need to go via a list; don't
end lines in semicolon (this is python).  Test isdir() once instead of
exists() twice on the same name.

Just, generally, use python's feature-set.

Change-Id: Ib114aa016f70b3be09e968d9cfc069b057f49d41
Reviewed-by: Lars Knoll <lars.knoll@qt.io>

2 files changed, 57 insertions, 85 deletions

diff --git a/util/local_database/cldr2qlocalexml.py b/util/local_database/cldr2qlocalexml.py
index 2f68b69a9f..fbc28ca712 100755
--- a/util/local_database/cldr2qlocalexml.py
+++ b/util/local_database/cldr2qlocalexml.py
@@ -46,14 +46,11 @@ import re
 
 import enumdata
 import xpathlite
-from  xpathlite import DraftResolution
+from xpathlite import DraftResolution, findAlias, findEntry, findTagsInFile
 from dateconverter import convert_date
 from localexml import Locale
 
-findAlias = xpathlite.findAlias
-findEntry = xpathlite.findEntry
 findEntryInFile = xpathlite._findEntryInFile
-findTagsInFile = xpathlite.findTagsInFile
 
 def parse_number_format(patterns, data):
     # this is a very limited parsing of the number format for currency only.
@@ -86,12 +83,8 @@ def parse_number_format(patterns, data):
     return result
 
 def parse_list_pattern_part_format(pattern):
-    # this is a very limited parsing of the format for list pattern part only.
-    result = ""
-    result = pattern.replace("{0}", "%1")
-    result = result.replace("{1}", "%2")
-    result = result.replace("{2}", "%3")
-    return result
+    # This is a very limited parsing of the format for list pattern part only.
+    return pattern.replace("{0}", "%1").replace("{1}", "%2").replace("{2}", "%3")
 
 def generateLocaleInfo(path):
     if not path.endswith(".xml"):
@@ -102,12 +95,11 @@ def generateLocaleInfo(path):
     if alias:
         raise xpathlite.Error('alias to "%s"' % alias)
 
-    language_code = findEntryInFile(path, "identity/language", attribute="type")[0]
-    country_code = findEntryInFile(path, "identity/territory", attribute="type")[0]
-    script_code = findEntryInFile(path, "identity/script", attribute="type")[0]
-    variant_code = findEntryInFile(path, "identity/variant", attribute="type")[0]
+    def code(tag):
+        return findEntryInFile(path, 'identity/' + tag, attribute="type")[0]
 
-    return _generateLocaleInfo(path, language_code, script_code, country_code, variant_code)
+    return _generateLocaleInfo(path, code('language'), code('script'),
+                               code('territory'), code('variant'))
 
 def _generateLocaleInfo(path, language_code, script_code, country_code, variant_code=""):
     if not path.endswith(".xml"):
@@ -126,12 +118,10 @@ def _generateLocaleInfo(path, language_code, script_code, country_code, variant_
     language_id = enumdata.languageCodeToId(language_code)
     if language_id <= 0:
         raise xpathlite.Error('unknown language code "%s"' % language_code)
-    language = enumdata.language_list[language_id][0]
 
     script_id = enumdata.scriptCodeToId(script_code)
     if script_id == -1:
         raise xpathlite.Error('unknown script code "%s"' % script_code)
-    script = enumdata.script_list[script_id][0]
 
     # we should handle fully qualified names with the territory
     if not country_code:
@@ -139,7 +129,6 @@ def _generateLocaleInfo(path, language_code, script_code, country_code, variant_
     country_id = enumdata.countryCodeToId(country_code)
     if country_id <= 0:
         raise xpathlite.Error('unknown country code "%s"' % country_code)
-    country = enumdata.country_list[country_id][0]
 
     # So we say we accept only those values that have "contributed" or
     # "approved" resolution. see http://www.unicode.org/cldr/process.html
@@ -147,36 +136,36 @@ def _generateLocaleInfo(path, language_code, script_code, country_code, variant_
     # compatibility.
     draft = DraftResolution.contributed
 
-    result = {}
-    result['language'] = language
-    result['script'] = script
-    result['country'] = country
-    result['language_code'] = language_code
-    result['country_code'] = country_code
-    result['script_code'] = script_code
-    result['variant_code'] = variant_code
-    result['language_id'] = language_id
-    result['script_id'] = script_id
-    result['country_id'] = country_id
+    result = dict(
+        language=enumdata.language_list[language_id][0],
+        language_code=language_code, language_id=language_id,
+        script=enumdata.script_list[script_id][0],
+        script_code=script_code, script_id=script_id,
+        country=enumdata.country_list[country_id][0],
+        country_code=country_code, country_id=country_id,
+        variant_code=variant_code)
 
     (dir_name, file_name) = os.path.split(path)
-    supplementalPath = dir_name + "/../supplemental/supplementalData.xml"
-    currencies = findTagsInFile(supplementalPath, "currencyData/region[iso3166=%s]"%country_code);
+    def from_supplement(tag,
+                        path=os.path.join(dir_name, '..', 'supplemental',
+                                          'supplementalData.xml')):
+        return findTagsInFile(path, tag)
+    currencies = from_supplement('currencyData/region[iso3166=%s]' % country_code)
     result['currencyIsoCode'] = ''
     result['currencyDigits'] = 2
     result['currencyRounding'] = 1
     if currencies:
         for e in currencies:
             if e[0] == 'currency':
-                tender = True
-                t = [x for x in e[1] if x[0] == 'tender']
-                if t and t[0][1] == 'false':
-                    tender = False;
-                if tender and not any(x[0] == 'to' for x in e[1]):
+                t = [x[1] == 'false' for x in e[1] if x[0] == 'tender']
+                if t and t[0]:
+                    pass
+                elif not any(x[0] == 'to' for x in e[1]):
                     result['currencyIsoCode'] = (x[1] for x in e[1] if x[0] == 'iso4217').next()
                     break
         if result['currencyIsoCode']:
-            t = findTagsInFile(supplementalPath, "currencyData/fractions/info[iso4217=%s]"%result['currencyIsoCode']);
+            t = from_supplement("currencyData/fractions/info[iso4217=%s]"
+                                % result['currencyIsoCode'])
             if t and t[0][0] == 'info':
                 result['currencyDigits'] = (int(x[1]) for x in t[0][1] if x[0] == 'digits').next()
                 result['currencyRounding'] = (int(x[1]) for x in t[0][1] if x[0] == 'rounding').next()
@@ -210,7 +199,9 @@ def _generateLocaleInfo(path, language_code, script_code, country_code, variant_
     result['percent'] = get_number_in_system(path, "numbers/symbols/percentSign", numbering_system)
     try:
         numbering_systems = {}
-        for ns in findTagsInFile(cldr_dir + "/../supplemental/numberingSystems.xml", "numberingSystems"):
+        for ns in findTagsInFile(os.path.join(cldr_dir, '..', 'supplemental',
+                                              'numberingSystems.xml'),
+                                 'numberingSystems'):
             tmp = {}
             id = ""
             for data in ns[1:][0]: # ns looks like this: [u'numberingSystem', [(u'digits', u'0123456789'), (u'type', u'numeric'), (u'id', u'latn')]]
@@ -373,7 +364,9 @@ locale_database = {}
 
 # see http://www.unicode.org/reports/tr35/tr35-info.html#Default_Content
 defaultContent_locales = {}
-for ns in findTagsInFile(cldr_dir + "/../supplemental/supplementalMetadata.xml", "metadata/defaultContent"):
+for ns in findTagsInFile(os.path.join(cldr_dir, '..', 'supplemental',
+                                      'supplementalMetadata.xml'),
+                         'metadata/defaultContent'):
     for data in ns[1:][0]:
         if data[0] == u"locales":
             defaultContent_locales = data[1].split()
diff --git a/util/local_database/qlocalexml2cpp.py b/util/local_database/qlocalexml2cpp.py
index 44187ede59..baa5a60263 100755
--- a/util/local_database/qlocalexml2cpp.py
+++ b/util/local_database/qlocalexml2cpp.py
@@ -77,54 +77,40 @@ def eachEltInGroup(parent, group, key):
             yield element
         element = element.nextSibling
 
-def eltText(elt):
-    result = ""
+def eltWords(elt):
     child = elt.firstChild
     while child:
         if child.nodeType == elt.TEXT_NODE:
-            if result:
-                result += " "
-            result += child.nodeValue
+            yield child.nodeValue
         child = child.nextSibling
-    return result
+
+def firstChildText(elt, key):
+    return ' '.join(eltWords(firstChildElt(elt, key)))
 
 def loadMap(doc, category):
-    return dict((int(eltText(firstChildElt(element, 'id'))),
-                 (eltText(firstChildElt(element, 'name')),
-                  eltText(firstChildElt(element, 'code'))))
+    return dict((int(firstChildText(element, 'id')),
+                 (firstChildText(element, 'name'),
+                  firstChildText(element, 'code')))
                 for element in eachEltInGroup(doc.documentElement,
                                               category + 'List', category))
 
 def loadLikelySubtagsMap(doc):
-    result = {}
+    def triplet(element, keys=('language', 'script', 'country')):
+        return tuple(firstChildText(element, key) for key in keys)
 
-    i = 0
-    for elt in eachEltInGroup(doc.documentElement, "likelySubtags", "likelySubtag"):
-        elt_from = firstChildElt(elt, "from")
-        from_language = eltText(firstChildElt(elt_from, "language"));
-        from_script = eltText(firstChildElt(elt_from, "script"));
-        from_country = eltText(firstChildElt(elt_from, "country"));
-
-        elt_to = firstChildElt(elt, "to")
-        to_language = eltText(firstChildElt(elt_to, "language"));
-        to_script = eltText(firstChildElt(elt_to, "script"));
-        to_country = eltText(firstChildElt(elt_to, "country"));
-
-        tmp = {}
-        tmp["from"] = (from_language, from_script, from_country)
-        tmp["to"] = (to_language, to_script, to_country)
-        result[i] = tmp;
-        i += 1
-    return result
+    return dict((i, {'from': triplet(firstChildElt(elt, "from")),
+                     'to': triplet(firstChildElt(elt, "to"))})
+                for i, elt in enumerate(eachEltInGroup(doc.documentElement,
+                                                       'likelySubtags', 'likelySubtag')))
 
 def fixedScriptName(name, dupes):
     # Don't .capitalize() as some names are already camel-case (see enumdata.py):
     name = ''.join(word[0].upper() + word[1:] for word in name.split())
     if name[-6:] != "Script":
-        name = name + "Script";
+        name = name + "Script"
     if name in dupes:
         sys.stderr.write("\n\n\nERROR: The script name '%s' is messy" % name)
-        sys.exit(1);
+        sys.exit(1)
     return name
 
 def fixedCountryName(name, dupes):
@@ -138,8 +124,8 @@ def fixedLanguageName(name, dupes):
     return name.replace(" ", "")
 
 def findDupes(country_map, language_map):
-    country_set = set([ v[0] for a, v in country_map.iteritems() ])
-    language_set = set([ v[0] for a, v in language_map.iteritems() ])
+    country_set = set(v[0] for a, v in country_map.iteritems())
+    language_set = set(v[0] for a, v in language_map.iteritems())
     return country_set & language_set
 
 def languageNameToId(name, language_map):
@@ -164,7 +150,7 @@ def loadLocaleMap(doc, language_map, script_map, country_map, likely_subtags_map
     result = {}
 
     for locale_elt in eachEltInGroup(doc.documentElement, "localeList", "locale"):
-        locale = Locale.fromXmlData(lambda k: eltText(firstChildElt(locale_elt, k)))
+        locale = Locale.fromXmlData(lambda k: firstChildText(locale_elt, k))
         language_id = languageNameToId(locale.language, language_map)
         if language_id == -1:
             sys.stderr.write("Cannot find a language id for '%s'\n" % locale.language)
@@ -324,8 +310,7 @@ def escapedString(s):
     return result
 
 def printEscapedString(s):
-    print escapedString(s);
-
+    print escapedString(s)
 
 def currencyIsoCodeData(s):
     if s:
@@ -346,13 +331,9 @@ def main():
     localexml = sys.argv[1]
     qtsrcdir = sys.argv[2]
 
-    if not os.path.exists(qtsrcdir) or not os.path.exists(qtsrcdir):
-        usage()
-    if not os.path.isfile(qtsrcdir + "/src/corelib/tools/qlocale_data_p.h"):
-        usage()
-    if not os.path.isfile(qtsrcdir + "/src/corelib/tools/qlocale.h"):
-        usage()
-    if not os.path.isfile(qtsrcdir + "/src/corelib/tools/qlocale.qdoc"):
+    if not (os.path.isdir(qtsrcdir)
+            and all(os.path.isfile(os.path.join(qtsrcdir, 'src', 'corelib', 'tools', leaf))
+                    for leaf in ('qlocale_data_p.h', 'qlocale.h', 'qlocale.qdoc'))):
         usage()
 
     (data_temp_file, data_temp_file_path) = tempfile.mkstemp("qlocale_data_p", dir=qtsrcdir)
@@ -377,7 +358,7 @@ def main():
     locale_map = loadLocaleMap(doc, language_map, script_map, country_map, likely_subtags_map)
     dupes = findDupes(language_map, country_map)
 
-    cldr_version = eltText(firstChildElt(doc.documentElement, "version"))
+    cldr_version = firstChildText(doc.documentElement, "version")
 
     data_temp_file.write("""
 /*
@@ -455,9 +436,7 @@ def main():
             index += count
         data_temp_file.write("%6d, // %s\n" % (i, language_map[key][0]))
     data_temp_file.write("     0 // trailing 0\n")
-    data_temp_file.write("};\n")
-
-    data_temp_file.write("\n")
+    data_temp_file.write("};\n\n")
 
     list_pattern_part_data = StringData('list_pattern_part_data')
     date_format_data = StringData('date_format_data')