def testMemorySafety(self):
# This is meant to be run with ASAN and checks that
# certain use-after-frees are not present.
cgi = neo_cgi.CGI()
hdf = cgi.hdf
del cgi
hdf.getValue("x", "y")
del hdf
cgi = neo_cgi.CGI()
cs = cgi.cs()
del cgi
cs.parseStr("x")
del cs
hdf = neo_util.HDF()
hdf.setValue("y.z", "1")
child = hdf.getChild("y")
del hdf
child.getValue("x", "y")
del child
hdf = neo_util.HDF()
cs = neo_cs.CS(hdf)
del hdf
cs.parseStr("x")
del cs
def execute(macro, args):
request = macro.request
content = []
page_name = macro.formatter.page.page_name
# get args
include_page_name = ''
if args is not None:
(include_page_name, _, hdf_text) = args.partition(',')
include_page_name = wikiutil.AbsPageName(page_name, include_page_name)
include_page = Page(request, include_page_name)
if include_page is None:
return ''
if not request.user.may.read(include_page_name):
return ''
cstemplate = include_page.getPageText()
pagename = macro.formatter.page.page_name
hdf = neo_util.HDF()
hdf.readString(hdf_text)
hdf.setValue("Config.WhiteSpaceStrip ", "0")
cs = neo_cs.CS(hdf)
cs.parseStr(cstemplate)
body = cs.render()
body = wikiutil.renderText(request, WikiParser, body)
return body
def __init__(self, loadpaths=[]):
"""Create a new HDF dataset.
The loadpaths parameter can be used to specify a sequence of paths under
which ClearSilver will search for template files:
>>> hdf = HDFWrapper(loadpaths=['/etc/templates',
... '/home/john/templates'])
>>> print hdf
hdf {
loadpaths {
0 = /etc/templates
1 = /home/john/templates
}
}
"""
try:
import neo_cgi
# The following line is needed so that ClearSilver can be loaded when
# we are being run in multiple interpreters under mod_python
neo_cgi.update()
import neo_util
self.hdf = neo_util.HDF()
except ImportError, e:
raise TracError, "ClearSilver not installed (%s)" % e
def _render(f, template_file, environ, start_response, *args, **kwds):
# call our original function with original args
try:
results = f(environ, start_response)
template_name, ext = template_file.split(".")
contenttype = "text/html"
if len(ext) > 1 and (ext[1] in extensions):
contenttype = extensions[ext[1]]
hdf = neo_util.HDF()
_set(hdf, '', results)
hdf.setValue('style', stylesheet_uri)
# shove the results into the template
clearsilver = neo_cs.CS(hdf)
clearsilver.parseFile(os.path.join('templates', template_name + '.cs'))
# XXX where is our error handling?
start_response("200 OK", [('Content-Type', contenttype)])
return [clearsilver.render()]
except DataNotFound:
start_response("404 Not Found", [('Content-Type', 'text/plain')])
return ['404 Error, Content not found']
except HTTP303, e:
url = str(e.value)
if not url.startswith(('http', '/')):
url = request_uri(environ) + url
start_response("302 Found", [('Location', url)])
return ['Redirect to url']
def GenreateCode(self):
items = urllib2.unquote(request.params['hdf']).split(';')
# .split(';')
hdf = neo_util.HDF()
print items
for item in items:
print item
v = item.split('=')
if (len(v) == 2):
hdf.setValue(v[0], v[1])
cs = neo_cs.CS(hdf)
cs.parseFile("./feedtpladmin/public/feed-java-code.cs"
) # parse a file from disk
gen_code = cs.render()
url_path = '/FeedBuilder_' + request.params[
'stype'] + '_' + request.params['version'] + '.java'
f = open("./feedtpladmin/public" + url_path, "w")
f.write(gen_code)
f.close()
return url_path
def dump_bug(level, etype, msg, location=None, nhdf=None):
global DISABLE_DUMP
if DISABLE_DUMP: return
now = int(time.time())
pid = os.getpid()
import neo_cgi, neo_util
hdf = neo_util.HDF()
hdf.setValue("Required.Level", level)
hdf.setValue("Required.When", str(int(time.time())))
hdf.setValue("Required.Type", etype)
hdf.setValue("Required.Title", msg)
hdf.setValue("Optional.Hostname", socket.gethostname())
if location:
hdf.setValue("Optional.Location", location)
for (key, value) in os.environ.items():
hdf.setValue("Environ.%s" % key, value)
global Count
Count = Count + 1
fname = "%d.%d_%d.%s" % (now, pid, Count, socket.gethostname())
checkPaths()
tpath = os.path.join(DUMP_DIR, "tmp", fname)
npath = os.path.join(DUMP_DIR, "new", fname)
hdf.writeFile(tpath)
os.rename(tpath, npath)
def format(self, formatter):
# print "format"
# format is also called for each !# command. its called after __init__
# is called. this is where parsers do most of their work.
# they write their results into the Httprequest object
# which is usually stored from __init__ in self.request.
# print "formatter",dir(formatter)
# formatter is a special object in MoinMoin that
# is supposed to help people who write extensions to have
# sort of a uniform looking thing going on.
# see http://moinmoin.wikiwikiweb.de/ApplyingFormatters?highlight=%28formatter%29
# but formatter is not documented well. you have to look at
# moinmoin/formatter/base.py. And if you do, you will see that half of
# the methods raise a 'not implemented' error.
# formatter is also being refactored alot so dont get used to it.
# if all else fails just use formatter.rawHTML which will
# screw up XML output but at least it will work.
page_name = formatter.page.page_name
cs_template_page = wikiutil.AbsPageName(page_name,
self.kw["format_args"])
cs_template = Page(self.request, cs_template_page).getPageText()
hdf = neo_util.HDF()
hdf.readString(self.raw)
hdf.setValue("Config.WhiteSpaceStrip", "0")
cs = neo_cs.CS(hdf)
cs.parseStr(cs_template)
body = cs.render()
body = wikiutil.renderText(self.request, WikiParser, body)
self.request.write(formatter.rawHTML(body))
def error(self, ecode, reason=None):
import httpResponses
message = httpResponses.gHTTPResponses[ecode]
template = httpResponses.errorMessage_Default
if ecode == 404:
template = httpResponses.errorMessage_404
hdf = neo_util.HDF()
hdf.setValue("code", str(ecode))
if message: hdf.setValue("message", message)
if reason: hdf.setValue("reason", reason)
for key, val in self.context.environ.items():
hdf.setValue("environ." + key, str(val))
self.context.stdout.write("Content-Type: text/html\r\n")
self.context.setStatus(None, ecode)
self.context.stdout.write("Status: %s\r\n" % ecode)
self.context.stdout.write("\r\n")
cs = neo_cs.CS(hdf)
cs.parseStr(template)
page = cs.render()
self.context.stdout.write(page)
warn("Error", message, reason)
def loadMap(self, file, prefix, lang):
log("Loading map for language %s" % lang)
hdf = neo_util.HDF()
hdf.readFile(file)
obj = hdf.getChild(prefix)
updates = 0
new_r = 0
while obj is not None:
s_id = obj.name()
str = obj.value()
try:
map_r = self.tdb.maps.fetchRow([('string_id', s_id),
('lang', lang)])
except odb.eNoMatchingRows:
map_r = self.tdb.maps.newRow()
map_r.string_id = s_id
map_r.lang = lang
new_r = new_r + 1
if map_r.string != str:
updates = updates + 1
map_r.string = str
map_r.save()
obj = obj.next()
log("New maps: %d Updates: %d" % (new_r, updates - new_r))
def testCsRenderStrip(self):
hdf = neo_util.HDF()
cs = neo_cs.CS(hdf)
hdf.setValue("Foo.Bar", "1")
cs.parseStr("This is my file <?cs var:Foo.Bar ?> ")
assert cs.render() == 'This is my file 1 '
hdf.setValue("ClearSilver.WhiteSpaceStrip", "1")
assert cs.render() == 'This is my file 1 '
def ValidateTemplate(self):
tpl = request.params['tpl']
hdf = neo_util.HDF() # create an HDF dataset
cs = neo_cs.CS(hdf)
try:
cs.parseStr(urllib2.unquote(str(tpl)))
cs.render()
except neo_util.ParseError, e:
return "错误:", str(e)
def __init__(self, env, msg_template):
self.env = env
self.db = env.get_db_cnx()
self.hdf = neo_util.HDF()
core.populate_hdf(self.hdf, env, self.db, None)
tmpl = os.path.join(env.get_config('general', 'templates_dir'),
msg_template)
self.cs = neo_cs.CS(self.hdf)
self.cs.parseFile(tmpl)
def render():
hdf = neo_util.HDF()
hdf.setValue('hdf.loadpaths.0', dirname)
hdf.setValue('title', escape('Just a test'))
hdf.setValue('user', escape('joe'))
for num in range(1, 15):
hdf.setValue('items.%d' % (num - 1), escape('Number %d' % num))
cs = neo_cs.CS(hdf)
cs.parseFile('template.cs')
return cs.render()
def render():
hdf = neo_util.HDF()
hdf.setValue("hdf.loadpaths.0", dirname)
hdf.setValue("title", escape("Just a test"))
hdf.setValue("user", escape("joe"))
for num in range(1, 15):
hdf.setValue("items.%d" % (num - 1), escape("Number %d" % num))
cs = neo_cs.CS(hdf)
cs.parseFile("template.cs")
return cs.render()
def init_request(self):
import neo_cgi
# The following line is needed so that ClearSilver can be loaded when
# we are being run in multiple interpreters under mod_python
neo_cgi.update()
import neo_cs
import neo_util
import Cookie
self.hdf = neo_util.HDF()
self.incookie = Cookie.SimpleCookie()
self.outcookie = Cookie.SimpleCookie()
def ValidateTemplate(self):
stype = int(request.params['stype'])
tpl = request.params['tpl']
if stype <= 0 or not tpl:
return '请求参数错误'
hdf = neo_util.HDF() # create an HDF dataset
cs = neo_cs.CS(hdf)
try:
cs.parseStr(urllib2.unquote(str(tpl)))
cs.render()
except neo_util.ParseError, e:
return "错误:", str(e)
def test_clearsilver():
"""ClearSilver"""
hdf = neo_util.HDF()
for i, row in enumerate(table):
for j, c in enumerate(row.values()):
hdf.setValue("rows.%d.cell.%d" % (i, j), cgi.escape(str(c)))
cs = neo_cs.CS(hdf)
cs.parseStr("""
<table><?cs
each:row=rows
?><tr><?cs each:c=row.cell
?><td><?cs var:c ?></td><?cs /each
?></tr><?cs /each?>
</table>""")
cs.render()
def execute(macro, args):
request = macro.request
content = []
page_name = macro.formatter.page.page_name
# get args
include_page_name = ''
if args is not None:
include_page_name = args
include_page_name = wikiutil.AbsPageName(page_name, include_page_name)
include_page = Page(request, include_page_name)
if include_page is None:
return ''
if not request.user.may.read(include_page_name):
return ''
cstemplate = include_page.getPageText()
pagename = macro.formatter.page.page_name
kv_page = Page(request, pagename)
kv_body = kv_page.get_body()
hdf = neo_util.HDF()
for line in kv_body.split("\n"):
if line.startswith("##"):
line = line[2:].strip()
parts = line.split("=", 1)
if len(parts) == 2:
val = parts[1].strip()
#val = parts[1].strip().encode('utf-8') # swh
hdf.setValue(parts[0].strip(), val)
hdf.setValue("Config.WhiteSpaceStrip ", "0")
cs = neo_cs.CS(hdf)
cs.parseStr(cstemplate)
body = cs.render()
body = wikiutil.renderText(request, WikiParser, body)
open("/tmp/out.html", "w").write(body)
return body
def testHdf(self):
hdf = neo_util.HDF()
hdf.setValue("Foo", "bar")
assert hdf.getValue("Foo", "baz") == "bar"
assert hdf.getValue("Foo.1", "baz") == "baz"
hdf.setValue("Numbers.1", "1")
hdf.setValue("Numbers.2", "2")
hdf.setValue("Numbers.3", "3")
assert hdf.getIntValue("Numbers.2", -1) == 2
assert hdf.getIntValue("Numbers.5", -1) == -1
hdf_num = hdf.getObj("Numbers")
assert hdf_num.getIntValue("2", -1) == 2
assert hdf_num.name() == "Numbers"
assert hdf_num.child().name() == "1"
assert hdf_num.child().value() == "1"
hdf.setAttr("Numbers", "type", "integers")
hdf.setAttr("Numbers", "k", "v")
assert hdf.getAttrs("Numbers") == [('type', 'integers'), ('k', 'v')]
assert hdf_num.attrs() == [('type', 'integers'), ('k', 'v')]
def dump_hdf(self, directory, owner):
global QUIET
sys.path.insert(0, "../python")
sys.path.insert(0, "python")
import neo_cgi, neo_util
hdf = neo_util.HDF()
date = time.strftime("%d %B %Y", time.localtime(time.time()))
if not self._funcs.items(): return
for name, f in self._funcs.items():
if f._title is None and f._desc is None and f._args is None and f._retr is None:
if not QUIET:
sys.stderr.write('-W- No info for function "%s()"\n' %
name)
continue
if f._defn is None:
if not QUIET:
sys.stderr.write('-W- No defn for function "%s()"\n' %
name)
hdf.setValue("Code.%s" % name, name)
obj = hdf.getObj("Code.%s" % name)
obj.setValue("Name", name)
obj.setValue("filename", self._filename)
if f._title: obj.setValue("Title", f._title)
if f._defn: obj.setValue("Define", neo_cgi.text2html(f._defn))
if f._args: obj.setValue("Args", neo_cgi.text2html(f._args))
if f._desc: obj.setValue("Desc", neo_cgi.text2html(f._desc))
if string.strip(f._other):
obj.setValue("Other",
neo_cgi.text2html(string.strip(f._other)))
if f._output: obj.setValue("Output", neo_cgi.text2html(f._output))
n = 0
for func in self._funcs.keys():
obj.setValue("related.%d" % n, func)
n = n + 1
fname = self._filename
x = string.rindex(fname, "/")
if x != -1: fname = fname[x + 1:]
x = string.rindex(fname, '.')
if x != -1: fname = fname[:x]
hdf.writeFile("%s/%s.hdf" % (directory, fname))
def getItemReferences(self, key):
if self.loggedin == False:
self.login()
key = key.strip()
url = self.site + "invent/api.py?Action.getItemReferences=1&key=%s" % (key,)
fp = self.opener.open(url)
body = fp.read()
fp.close()
import neo_cgi, neo_util
import simple_hdfhelp as hdfhelp
hdf = neo_util.HDF()
hdf.readString(body)
ret = {}
for k,o in hdfhelp.hdf_ko_iterator(hdf.getObj("CGI.cur.refs")):
ret[o.getValue("name", "")] = o.getValue("reference", "")
return ret
def stringsHDF(self, prefix, locations, lang='en', exist=0, tiered=0):
hdf = neo_util.HDF()
if exist and lang == 'en': return hdf
done = {}
locations.sort()
maps = self.tdb.maps.fetchRows(('lang', lang))
maps_d = {}
for map in maps:
maps_d[int(map.string_id)] = map
strings = self.tdb.strings.fetchRows()
strings_d = {}
for string in strings:
strings_d[int(string.string_id)] = string
count = 0
for loc in locations:
s_id = int(loc.string_id)
if done.has_key(s_id): continue
try:
s_row = maps_d[s_id]
if exist: continue
except KeyError:
try:
s_row = strings_d[s_id]
except KeyError:
log("Missing string_id %d, skipping" % s_id)
continue
count = count + 1
if tiered:
hdf.setValue(
"%s.%d.%d.%s" % (prefix, int(s_id) / TIER1_DIV,
int(s_id) / TIER2_DIV, s_id),
s_row.string)
else:
hdf.setValue("%s.%s" % (prefix, s_id), s_row.string)
done[s_id] = 1
if exist == 1: log("Missing %d strings for lang %s" % (count, lang))
return hdf
import neo_cgi
import neo_util
import neo_cs
hdf = neo_util.HDF()
hdf.setValue("hdf.loadpaths.0", ".")
hdf.readFile("units.hdf")
hdf.readFile("slots.hdf")
cs = neo_cs.CS(hdf)
cs.parseFile("dhcpd.conf.cst")
print cs.render()
def generate_po_files(hdf_file, po_dir, textDomain):
## prepare hdf
if ((not os.path.isfile(hdf_file)) or (not os.access(hdf_file, os.R_OK))):
sys.stderr.write("Unable to read the hdf file: %s\n" % hdf_file)
return
if not os.path.isdir(po_dir):
os.mkdir(po_dir)
pot_file = os.path.join(po_dir, "%s.pot" % textDomain)
hdf = neo_util.HDF()
hdf.readFile(hdf_file)
## update pot
if not os.path.isfile(pot_file):
sys.stdout.write("Creating: %s\n" % pot_file)
pot = translate.storage.po.pofile(encoding="utf-8")
pot.makeheader(pot_creation_date=True)
pot.updateheader(add=True,
Project_Id_Version='ezmlm-web 3.3',
pot_creation_date=True,
language_team='Lars Kruse <%s>' % MAIL_ADDRESS,
Report_Msgid_Bugs_To=MAIL_ADDRESS,
encoding='utf-8',
Plural_Forms=['nplurals=2', 'plural=(n != 1)'])
#TODO: somehow we need 'updateheaderplural'
else:
sys.stdout.write("Loading: %s\n" % pot_file)
pot = translate.storage.po.pofile.parsefile(pot_file)
## remove all msgids - we will add them later
pot.units = []
## add new entries
def walk_hdf(prefix, node):
def addPoItem(hdf_node):
## ignore hdf values with a "LINK" attribute
for (key, value) in hdf_node.attrs():
if key == "LINK":
return
if not hdf_node.value():
return
item = pot.findunit(hdf_node.value())
if not item:
item = pot.addsourceunit(hdf_node.value())
item.addlocation("%s%s" % (prefix, hdf_node.name()))
while node:
if node.name():
new_prefix = prefix + node.name() + '.'
else:
new_prefix = prefix
## as the attribute feature of clearsilver does not work yet, we
## have to rely on magic names to prevent the translation of links
if not (new_prefix.endswith(".Link.Rel.") \
or new_prefix.endswith(".Link.Prot.") \
or new_prefix.endswith(".Link.Abs.") \
or new_prefix.endswith(".Link.Attr1.name.") \
or new_prefix.endswith(".Link.Attr1.value.") \
or new_prefix.endswith(".Link.Attr2.name.") \
or new_prefix.endswith(".Link.Attr2.value.") \
or new_prefix == "Lang.Name."):
addPoItem(node)
walk_hdf(new_prefix, node.child())
node = node.next()
walk_hdf("", hdf)
pot.savefile(pot_file)
## create po files
for ld in ALL_LANGUAGES:
if not os.path.isdir(os.path.join(po_dir, ld)):
os.mkdir(os.path.join(po_dir, ld))
if not os.path.isdir(os.path.join(po_dir, ld, 'LC_MESSAGES')):
os.mkdir(os.path.join(po_dir, ld, 'LC_MESSAGES'))
po_file = os.path.join(po_dir, ld, 'LC_MESSAGES', "%s.po" % textDomain)
if not os.path.isfile(po_file):
translate.convert.pot2po.convertpot(file(pot_file),
file(po_file, 'w'), None)
else:
po2_file = po_file + '.new'
translate.convert.pot2po.convertpot(file(pot_file),
file(po2_file, 'w'),
file(po_file))
os.rename(po2_file, po_file)
if ld == DEFAULT_LANG:
## set every msgstr to the respective msgid
po_data = translate.storage.po.pofile.parsefile(po_file)
po_data.removeduplicates()
for po_unit in po_data.units:
po_unit.settarget(po_unit.getsource())
po_data.savefile(po_file)
else:
po_data = translate.storage.po.pofile.parsefile(po_file)
po_data.removeduplicates()
## go through all msgstr and remove empty ones
for index in range(len(po_data.units) - 1, 0, -1):
if po_data.units[index].isfuzzy() and \
(po_data.units[index].msgidlen() == 0):
po_data.units.remove(po_data.units[index])
po_data.savefile(po_file)
if USE_SVN:
revert_if_unchanged(po_file)
## make it writeable for pootle
os.chmod(po_file, 0666)
## compile po file
mo_file = po_file[:-3] + '.mo'
translate.tools.pocompile.convertmo(file(po_file), file(mo_file, 'w'),
file(pot_file))
def generate_translated_hdf_file(orig_hdf_file, po_dir, hdf_dir, textdomain,
language, src_languages):
import gettext
## prepare original hdf
if ((not os.path.isfile(orig_hdf_file))
or (not os.access(orig_hdf_file, os.R_OK))):
sys.stderr.write("Unable to read the hdf file: %s\n" % orig_hdf_file)
return
hdf = neo_util.HDF()
hdf.readFile(orig_hdf_file)
## name of new hdf file
new_hdf_file = os.path.join(hdf_dir, language + '.hdf')
## create translation object
translator = gettext.translation(textdomain,
localedir=po_dir,
languages=src_languages)
## translate entries
## count the number of translated items - so we can decide later, if we
## want to create the language file
def walk_hdf(prefix, node):
translate_count = 0
def addHdfItem(hdf_node):
## ignore hdf values with a "LINK" attribute
for (key, value) in hdf_node.attrs():
if key == "LINK":
return
if not hdf_node.value():
return
translated = translator.gettext(hdf_node.value())
if translated:
hdf.setValue("%s%s" % (prefix, hdf_node.name()), translated)
return True
else:
hdf.setValue("%s%s" % (prefix, hdf_node.name()),
hdf_node.value())
return False
while node:
if node.name():
new_prefix = prefix + node.name() + '.'
else:
new_prefix = prefix
## as the attribute feature of clearsilver does not work yet, we
## have to rely on magic names to prevent the translation of links
if (new_prefix.endswith(".Link.Rel.") \
or new_prefix.endswith(".Link.Prot.") \
or new_prefix.endswith(".Link.Abs.") \
or new_prefix.endswith(".Link.Attr1.name.") \
or new_prefix.endswith(".Link.Attr1.value.") \
or new_prefix.endswith(".Link.Attr2.name.") \
or new_prefix.endswith(".Link.Attr2.value.")):
pass
elif new_prefix == "Lang.Name.":
# set the "Lang.Name" attribute properly
# remove trailing dot
new_prefix = new_prefix.strip(".")
if language in LANGUAGE_NAMES:
hdf.setValue(new_prefix, LANGUAGE_NAMES[language])
else:
hdf.setValue(new_prefix, language)
else:
if addHdfItem(node):
translate_count += 1
translate_count += walk_hdf(new_prefix, node.child())
node = node.next()
return translate_count
translated_items_count = walk_hdf("", hdf)
## if there was at least one valid translation, then we should write
## the language file
if translated_items_count > 0:
print "Writing translation: %s" % language
hdf.writeFile(new_hdf_file)
else:
print "Skipping empty translation: %s" % language
def test():
import neo_util
hdf = neo_util.HDF()
hdf.setValue("foo","1")
print eval_cs(hdf,"this should say 1 ===> <?cs var:foo ?>")
class Translator:
_HTML_TAG_RE = None
_HTML_TAG_REGEX = '<[^!][^>]*?>'
_HTML_CMT_RE = None
_HTML_CMT_REGEX = '<!--.*?-->'
_CS_TAG_RE = None
_CS_TAG_REGEX = '<\\?.+?\\?>'
def __init__(self):
self.tdb = db_trans.trans_connect()
# configuration data ......
# - we should stop hardcoding this... - jeske
self.root = "testroot"
self.languages = ['es', 'en']
self.ignore_paths = ['tmpl/m'] # common place for mockups
self.ignore_files = ['blah_ignore.cs'] # ignore clearsilver file
# ignore clearsilver javascript files
self.ignore_patterns = ['tmpl/[^ ]*_js.cs']
# ............................
if self.root is None:
raise "Unable to determine installation root"
if Translator._HTML_TAG_RE is None:
Translator._HTML_TAG_RE = re.compile(Translator._HTML_TAG_REGEX,
re.MULTILINE | re.DOTALL)
if Translator._HTML_CMT_RE is None:
Translator._HTML_CMT_RE = re.compile(Translator._HTML_CMT_REGEX,
re.MULTILINE | re.DOTALL)
if Translator._CS_TAG_RE is None:
Translator._CS_TAG_RE = re.compile(Translator._CS_TAG_REGEX,
re.MULTILINE | re.DOTALL)
self._html_state = 0
def parseHTMLTag(self, data):
# this is only called if we see a full tag in one parse...
i = 0
if len(data) == 0: return []
if data[0] in '/?': return []
while i < len(data) and data[i] not in ' \n\r\t>':
i = i + 1
if i == len(data): return []
tag = data[:i].lower()
#print "Searching tag: %s" % data
#print "Found tag: %s" % tag
results = []
attrfind = re.compile(r'\s*([a-zA-Z_][-.a-zA-Z_0-9]*)(\s*=\s*'
r'(\'[^\']*\'|"[^"]*"|[^ \t\n<>]*))?')
k = i
attrs = {}
attrs_beg = {}
while k < len(data):
match = attrfind.match(data, k)
if not match: break
attrname, rest, attrvalue = match.group(1, 2, 3)
if not rest:
attrvalue = attrname
elif attrvalue[:1] == '\'' == attrvalue[-1:] or \
attrvalue[:1] == '"' == attrvalue[-1:]:
attrvalue = attrvalue[1:-1]
attrname = attrname.lower()
if attrs.has_key(attrname):
log("Can't handle duplicate attrs: %s" % attrname)
attrs[attrname] = attrvalue
attrs_beg[attrname] = match.start(3)
k = match.end(0)
find_l = []
if tag == "input":
if attrs.get('type', "").lower() in ["submit", "button"]:
find_l.append((attrs.get('value',
''), attrs_beg.get('value', 0)))
for s, k in find_l:
if s:
x = data[k:].find(s)
if x != -1: results.append((s, x + k, 1))
return results
def parseHTML(self, data, reset=1):
if reset: self._html_state = 0
if DEBUG:
print "- %d ---------\n%s\n- E ---------" % (self._html_state,
data)
results = []
i = 0
n = len(data)
# if we had state from the last parse... find it
if self._html_state:
if self._html_state == 2:
x = string.find(data[i:], '-->')
l = 3
else:
x = string.find(data[i:], '>')
l = 1
if x == -1: return results
i = i + x + l
self._html_state = 0
while i < n:
if DEBUG: print "MATCHING>%s<MATCHING" % data[i:]
cmt_b = string.find(data[i:], '<!--')
cmt_e = string.find(data[i:], '-->')
tag_b = string.find(data[i:], '<')
tag_e = string.find(data[i:], '>')
if DEBUG: print "B> %d %d %d %d <B" % (cmt_b, cmt_e, tag_b, tag_e)
if cmt_b != -1 and cmt_b <= tag_b:
x = i
y = i + cmt_b - 1
while x < y and data[x] in string.whitespace:
x += 1
while y > x and data[y] in string.whitespace:
y -= 1
results.append((data[x:y + 1], x, 1))
if cmt_e == -1: # partial comment:
self._html_state = 2
break
i = i + cmt_e + 3
elif tag_b != -1:
x = i
y = i + tag_b - 1
while x < y and data[x] in string.whitespace:
x += 1
while y > x and data[y] in string.whitespace:
y -= 1
results.append((data[x:y + 1], x, 1))
if tag_e == -1: # partial tag
self._html_state = 1
break
h_results = self.parseHTMLTag(data[i + tag_b + 1:i + tag_e])
h_results = map(lambda x: (x[0], x[1] + i + tag_b + 1, x[2]),
h_results)
results = results + h_results
i = i + tag_e + 1
else:
x = i
y = n - 1
while x < y and data[x] in string.whitespace:
x += 1
while y > x and data[y] in string.whitespace:
y -= 1
results.append((data[x:y + 1], x, 1))
break
return results
def parseCS(self, data):
results = []
i = 0
n = len(data)
while i < n:
m = Translator._CS_TAG_RE.search(data, i)
if not m:
# search for a partial...
x = string.find(data[i:], '<?')
if x == -1:
results.append((data[i:], i))
else:
results.append((data[i:x], i))
break
(b, e) = m.span()
if i != b: results.append((data[i:b], i))
i = e
t_results = []
self._html_in = 0
for (s, ofs) in results:
r = self.parseHTML(s, reset=0)
r = map(lambda x: (x[0], x[1] + ofs, x[2]), r)
t_results = t_results + r
return t_results
def descendHDF(self, obj, prefix):
results = []
while obj is not None:
if obj.value():
attrs = obj.attrs()
attrs = map(lambda x: x[0], attrs)
if "Lang" in attrs:
if prefix:
results.append(
(obj.value(), "%s.%s" % (prefix, obj.name()), 0))
else:
results.append((obj.value(), "%s" % (obj.name()), 0))
if obj.child():
if prefix:
results = results + self.descendHDF(
obj.child(), "%s.%s" % (prefix, obj.name()))
else:
results = results + self.descendHDF(
obj.child(), (obj.name()))
obj = obj.next()
return results
def parseHDF(self, data):
# Ok, we handle HDF files specially.. the theory is, we only
# extract entire HDF elements which have the attribute Lang
hdf = neo_util.HDF()
hdf.readString(data, 1)
return self.descendHDF(hdf, "")
def handleFile(self, file):
if file in self.ignore_files: return []
for a_re in self.ignore_patterns:
if re.match(a_re, file):
return []
fpath = self.root + '/' + file
x = string.rfind(file, '.')
if x == -1: return []
data = open(fpath, 'r').read()
ext = file[x:]
strings = []
if ext in ['.cst', '.cs']:
strings = self.parseCS(data)
elif ext in ['.html', '.htm']:
strings = self.parseHTML(data)
elif ext in ['.hdf']:
strings = self.parseHDF(data)
if len(strings):
print "Found %d strings in %s" % (len(strings), file)
return strings
return []
def walkDirectory(self, path):
if path in self.ignore_paths: return []
fpath = self.root + '/' + path
files = os.listdir(fpath)
dirs = []
results = []
for file in files:
if file[0] == '.': continue
fname = fpath + '/' + file
if os.path.isdir(fname):
dirs.append(file)
else:
strings = self.handleFile(path + '/' + file)
if len(strings):
results.append((path + '/' + file, strings))
for dir in dirs:
if dir not in ["release"]:
results = results + self.walkDirectory(path + '/' + dir)
return results
def cleanHtmlString(self, s):
s = re.sub("\s+", " ", s)
return string.strip(s)
def containsWords(self, s, ishtml):
if ishtml:
s = string.replace(s, ' ', ' ')
s = string.replace(s, '"', '"')
s = string.replace(s, '©', '')
s = string.replace(s, '<', '<')
s = string.replace(s, '>', '>')
s = string.replace(s, '&', '&')
for x in range(len(s)):
n = ord(s[x])
if (n > 47 and n < 58) or (n > 64 and n < 91) or (n > 96
and n < 123):
return 1
return 0
def findString(self, s):
rows = self.tdb.strings.fetchRows(('string', s))
if len(rows) == 0:
row = self.tdb.strings.newRow()
row.string = s
row.save()
return row.string_id
elif len(rows) > 1:
raise eTransError, "String %s exists multiple times!" % s
else:
return rows[0].string_id
def loadStrings(self, one_file=None, verbose=0):
if one_file is not None:
strings = self.handleFile(one_file)
results = [(one_file, strings)]
else:
results = self.walkDirectory('tmpl')
uniq = {}
cnt = 0
seen_hdf = {}
for fname, strings in results:
for (s, ofs, ishtml) in strings:
if s and string.strip(s):
l = len(s)
if ishtml:
s = self.cleanHtmlString(s)
if self.containsWords(s, ishtml):
if type(ofs) == type(""): # HDF
if seen_hdf.has_key(ofs):
if seen_hdf[ofs][0] != s:
log("Duplicate HDF Name %s:\n\t file %s = %s\n\t file %s = %s"
% (ofs, seen_hdf[ofs][1],
seen_hdf[ofs][0], fname, s))
else:
seen_hdf[ofs] = (s, fname)
try:
uniq[s].append((fname, ofs, l))
except KeyError:
uniq[s] = [(fname, ofs, l)]
cnt = cnt + 1
print "%d strings, %d unique" % (cnt, len(uniq.keys()))
fp = open("map", 'w')
for (s, locs) in uniq.items():
locs = map(lambda x: "%s:%s:%d" % x, locs)
fp.write('#: %s\n' % (string.join(locs, ',')))
fp.write('msgid=%s\n\n' % repr(s))
log("Loading strings/locations into database")
locations = []
for (s, locs) in uniq.items():
s_id = self.findString(s)
for (fname, ofs, l) in locs:
if type(ofs) == type(""): # ie, its HDF
location = "hdf:%s" % ofs
else:
location = "ofs:%d:%d" % (ofs, l)
loc_r = TransLoc(s_id, fname, location)
locations.append(loc_r)
return locations
def stringsHDF(self, prefix, locations, lang='en', exist=0, tiered=0):
hdf = neo_util.HDF()
if exist and lang == 'en': return hdf
done = {}
locations.sort()
maps = self.tdb.maps.fetchRows(('lang', lang))
maps_d = {}
for map in maps:
maps_d[int(map.string_id)] = map
strings = self.tdb.strings.fetchRows()
strings_d = {}
for string in strings:
strings_d[int(string.string_id)] = string
count = 0
for loc in locations:
s_id = int(loc.string_id)
if done.has_key(s_id): continue
try:
s_row = maps_d[s_id]
if exist: continue
except KeyError:
try:
s_row = strings_d[s_id]
except KeyError:
log("Missing string_id %d, skipping" % s_id)
continue
count = count + 1
if tiered:
hdf.setValue(
"%s.%d.%d.%s" % (prefix, int(s_id) / TIER1_DIV,
int(s_id) / TIER2_DIV, s_id),
s_row.string)
else:
hdf.setValue("%s.%s" % (prefix, s_id), s_row.string)
done[s_id] = 1
if exist == 1: log("Missing %d strings for lang %s" % (count, lang))
return hdf
def dumpStrings(self, locations, lang=None):
log("Dumping strings to HDF")
if lang is None:
langs = ['en']
sql = "select lang from nt_trans_maps group by lang"
cursor = self.tdb.defaultCursor()
cursor.execute(sql)
rows = cursor.fetchall()
for row in rows:
langs.append(row[0])
else:
langs = [lang]
for a_lang in langs:
hdf = self.stringsHDF('S', locations, a_lang)
hdf.writeFile("strings_%s.hdf" % a_lang)
for a_lang in langs:
hdf = self.stringsHDF('S', locations, a_lang, exist=1)
if hdf.child():
hdf.writeFile("strings_missing_%s.hdf" % a_lang)
def fetchString(self, s_id, lang):
if lang == "hdf":
return "<?cs var:Lang.Extracted.%d.%d.%s ?>" % (
int(s_id) / TIER1_DIV, int(s_id) / TIER2_DIV, s_id)
rows = self.tdb.maps.fetchRows([('string_id', s_id), ('lang', lang)])
if len(rows) == 0:
try:
row = self.tdb.strings.fetchRow(('string_id', s_id))
except odb.eNoMatchingRows:
log("Unable to find string id %s" % s_id)
raise eNoString
if lang != 'en':
log("Untranslated string for id %s" % s_id)
return row.string
else:
return rows[0].string
def dumpFiles(self, locations, lang):
log("Dumping files for %s" % lang)
files = {}
for row in locations:
try:
files[row.filename].append(row)
except KeyError:
files[row.filename] = [row]
hdf_map = []
os.system("rm -rf %s/gen/tmpl" % (self.root))
for file in files.keys():
fname = "%s/gen/%s" % (self.root, file)
try:
os.makedirs(os.path.dirname(fname))
except OSError, reason:
if reason[0] != 17:
raise
do_hdf = 0
x = string.rfind(file, '.')
if x != -1 and file[x:] == '.hdf':
do_hdf = 1
ofs = []
for loc in files[file]:
parts = string.split(loc.location, ':')
if len(parts) == 3 and parts[0] == 'ofs' and do_hdf == 0:
ofs.append((int(parts[1]), int(parts[2]), loc.string_id))
elif len(parts) == 2 and parts[0] == 'hdf' and do_hdf == 1:
hdf_map.append((parts[1], loc.string_id))
else:
log("Invalid location for loc_id %s" % loc.loc_id)
continue
if not do_hdf:
ofs.sort()
data = open(self.root + '/' + file).read()
# ok, now we split up the original data into sections
x = 0
n = len(data)
out = []
#sys.stderr.write("%s\n" % repr(ofs))
while len(ofs):
if ofs[0][0] > x:
out.append(data[x:ofs[0][0]])
x = ofs[0][0]
elif ofs[0][0] == x:
out.append(self.fetchString(ofs[0][2], lang))
x = ofs[0][0] + ofs[0][1]
ofs = ofs[1:]
else:
log("How did we get here? %s x=%d ofs=%d sid=%d" %
(file, x, ofs[0][0], ofs[0][2]))
log("Data[x:20]: %s" % data[x:20])
log("Data[ofs:20]: %s" % data[ofs[0][0]:20])
break
if n > x:
out.append(data[x:])
odata = string.join(out, '')
open(fname, 'w').write(odata)
if lang == "hdf":
langs = self.languages
else:
langs = [lang]
for d_lang in langs:
# dumping the extracted strings
hdf = self.stringsHDF('Lang.Extracted',
locations,
d_lang,
tiered=1)
fname = "%s/gen/tmpl/lang_%s.hdf" % (self.root, d_lang)
hdf.writeFile(fname)
data = open(fname).read()
fp = open(fname, 'w')
fp.write('## AUTOMATICALLY GENERATED -- DO NOT EDIT\n\n')
fp.write(data)
fp.write('\n#include "lang_map.hdf"\n')
# dumping the hdf strings file
if d_lang == "en":
map_file = "%s/gen/tmpl/lang_map.hdf" % (self.root)
else:
map_file = "%s/gen/tmpl/%s/lang_map.hdf" % (self.root, d_lang)
try:
os.makedirs(os.path.dirname(map_file))
except OSError, reason:
if reason[0] != 17: raise
map_hdf = neo_util.HDF()
for (name, s_id) in hdf_map:
str = hdf.getValue(
'Lang.Extracted.%d.%d.%s' %
(int(s_id) / TIER1_DIV, int(s_id) / TIER2_DIV, s_id), '')
map_hdf.setValue(name, str)
map_hdf.writeFile(map_file)
def parseHDF(self, data):
# Ok, we handle HDF files specially.. the theory is, we only
# extract entire HDF elements which have the attribute Lang
hdf = neo_util.HDF()
hdf.readString(data, 1)
return self.descendHDF(hdf, "")
