Minor fixes

* WikiTrans/wiki2html.py: Use absolute import names.
* WikiTrans/wiki2texi.py: Likewise.
* WikiTrans/wiki2text.py: Likewise.
* WikiTrans/wikimarkup.py: Likewise.
* bin/wikitrans: Enforce UTF-8 for Python 2
New option --base-name.
Optionally download the material from http

5 files changed, 58 insertions, 13 deletions

diff --git a/WikiTrans/wiki2html.py b/WikiTrans/wiki2html.py
index 0309ae3..67cb9c3 100644
--- a/WikiTrans/wiki2html.py
+++ b/WikiTrans/wiki2html.py
@@ -16,9 +16,9 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 from __future__ import print_function
-from wikimarkup import *
-from wikitoken import *
-from wikins import wiki_ns_re, wiki_ns
+from WikiTrans.wikimarkup import *
+from WikiTrans.wikitoken import *
+from WikiTrans.wikins import wiki_ns_re, wiki_ns
 import re
 try:
     from urllib import quote as url_quote
diff --git a/WikiTrans/wiki2texi.py b/WikiTrans/wiki2texi.py
index 106ad77..616fccb 100644
--- a/WikiTrans/wiki2texi.py
+++ b/WikiTrans/wiki2texi.py
@@ -15,9 +15,9 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-from wikimarkup import *
-from wikitoken import *
-from wikins import wiki_ns_re, wiki_ns
+from WikiTrans.wikimarkup import *
+from WikiTrans.wikitoken import *
+from WikiTrans.wikins import wiki_ns_re, wiki_ns
 import re
 import urllib
 
diff --git a/WikiTrans/wiki2text.py b/WikiTrans/wiki2text.py
index b5bd708..cd4937e 100644
--- a/WikiTrans/wiki2text.py
+++ b/WikiTrans/wiki2text.py
@@ -15,9 +15,9 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-from wikitoken import *
-from wikimarkup import *
-from wikins import wiki_ns_re, wiki_ns
+from WikiTrans.wikitoken import *
+from WikiTrans.wikimarkup import *
+from WikiTrans.wikins import wiki_ns_re, wiki_ns
 import re
 import urllib
 
diff --git a/WikiTrans/wikimarkup.py b/WikiTrans/wikimarkup.py
index ad0f675..e3bc26f 100644
--- a/WikiTrans/wikimarkup.py
+++ b/WikiTrans/wikimarkup.py
@@ -19,7 +19,7 @@ from __future__ import print_function
 import sys
 import re
 from types import *
-from wikitoken import *
+from WikiTrans.wikitoken import *
 
 __all__ = [ "BaseWikiMarkup", "WikiMarkup",
             "TagAttributes", "TagAttributeSyntax" ]
diff --git a/bin/wikitrans b/bin/wikitrans
index 0b2d867..e794dc2 100755
--- a/bin/wikitrans
+++ b/bin/wikitrans
@@ -16,7 +16,12 @@
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 from __future__ import print_function
+from __future__ import unicode_literals
 import sys
+import re
+import urllib
+import tempfile
+import xml.etree.ElementTree as etree
 from optparse import OptionParser
 try:
     from StringIO import StringIO
@@ -28,6 +33,19 @@ from WikiTrans.wiki2texi  import TexiWikiMarkup
 from WikiTrans.wikimarkup import WikiMarkup
 from WikiTrans.wikidump   import DumpWikiMarkup
 
+# Set utf-8 as the default encoding. 
+# Trying to do so using encode('utf_8')/unicode, which is 
+# supposed to be the right way, does not work in Python 2.7
+# Simply calling sys.setdefaultencoding is not possible,
+# because, for some obscure reason, Python chooses to delete 
+# this symbol from the namespace after setting its default 
+# encoding in site.py. That's why reload is needed. 
+try:
+    reload(sys)
+    sys.setdefaultencoding('utf-8')
+except:
+    pass
+
 handlers = {
     'dump': {
         'default': DumpWikiMarkup
@@ -52,10 +70,30 @@ def setkw(option, opt, value, parser):
     if val:
         parser.values.kwdict[kw] = val
 
+def getwiki(url):
+    tmp = tempfile.NamedTemporaryFile()            
+    u = urllib.URLopener()
+    u.retrieve(url, tmp.name)
+    tree = etree.parse(tmp.name)
+    root = tree.getroot()
+    ns = { 'wiki':'' }
+    if 'version' in root.attrib:
+        ns['wiki'] = 'http://www.mediawiki.org/xml/export-%s/' % root.attrib['version']
+        text = root.find('wiki:page/wiki:revision/wiki:text',ns)
+    if text is None:
+        print("no page/revision/text element in the downloaded page")
+        exit(0)
+    return text.text
+        
 def main():
-    usage = '%prog [OPTIONS] FILE'
+    usage = '%prog [OPTIONS] ARG'
     version = '%prog 1.0'
     description = """Translates MediaWiki documents markup to various other formats.
+If ARG looks like a URL, the wiki text to be converted will be downloaded
+from that URL.
+Otherwise, if --base-url is given, ARG is treated as the name of the page to
+get from the WikiMedia istallation at that URL.
+Otherwise, ARG is name of the file to read wiki material from.    
 """
     epilog = "Report bugs to: <gray+wikitrans@gnu.org.ua>"
     
@@ -91,12 +129,19 @@ def main():
                       action='store_const', const='dump',
                       dest='otype',
                       help='dump parse tree and exit; similar to --type=dump')
+    parser.add_option('-b', '--base-url',
+                      action='store', type='string', dest='base_url',
+                      help='set base url')
+    
 
     (options, args) = parser.parse_args()
-    
     if len(args) == 1:
-        if args[0] == '-':
+        if options.base_url:
+            options.kwdict['text'] = getwiki(options.base_url + '/wiki/Special:Export/' + args[0])
+        elif args[0] == '-':
             options.kwdict['file'] = sys.stdin
+        elif re.match('^(http|ftp)s?://',args[0]):
+            options.kwdict['text'] = getwiki(args[0])
         else:
             options.kwdict['filename'] = args[0]
     else: