Updated desciption of function parse_wrmap.

[philipp/winterrodeln/wrpylib.git] / wrpylib / mwmarkup.py

diff --git a/wrpylib/mwmarkup.py b/wrpylib/mwmarkup.py
index 30b2f654fa4236190a17fb3b631a116a2c304365..beffbd26b8c9ea03ee8f8469f1a4fba4f0e675cc 100644 (file)
--- a/wrpylib/mwmarkup.py
+++ b/wrpylib/mwmarkup.py
@@ -106,6 +106,48 @@ def create_template(template_title, anonym_params=[], named_param_keys=[], named
     return pipe_char.join(parts) + end_char
 
 
+def find_tag(wikitext, tagname, pos=0):
+    """Returns the tuple (start, end) of the first occurence of the tag '<tag ...>...</tag>'
+    or '<tag ... />'.
+    (None, None) is returned if the tag is not found.
+    If you are sure that the wikitext contains the tag, the tag could be extracted like follows:
+
+    >>> wikitext = u'This is a <tag>mytag</tag> tag.'
+    >>> start, end = find_template(wikitext, u'tag')
+    >>> print wikitext[start:end]
+    <tag>mytag</tag>
+
+    :param wikitext: The text (preferalbe unicode) that has the template in it.
+    :param tagname: Name of the tag, e.g. u'tag' for <tag>.
+    :param pos: position within wikitext to start searching the tag.
+    :return:
+        (start, content, endtag, end). start is the position of '<' of the tag,
+        content is the beginning of the content (after '>'), enttag is the
+        beginning of the end tag ('</') and end is one position after the end tag.
+        For single tags, (start, None, None, end) is returned.
+        If the tag is not found (or only the start tag is present,
+        (None, None, None, None) is returned.
+    """
+    # Find start tag
+    regexp_starttag = re.compile(u"<{0}.*?(/?)>".format(tagname), re.DOTALL)
+    match_starttag = regexp_starttag.search(wikitext, pos)
+    if match_starttag is None:
+        return None, None, None, None
+
+    # does the tag have content?
+    if len(match_starttag.group(1)) == 1: # group(1) is either '' or '/'.
+        # single tag
+        return match_starttag.start(), None, None, match_starttag.end()
+
+    # tag with content
+    regexp_endtag = re.compile(u'</{0}>'.format(tagname), re.DOTALL)
+    match_endtag = regexp_endtag.search(wikitext, match_starttag.end())
+    if match_endtag is None:
+        # No closing tag - error in wikitext
+        return None, None, None, None
+    return match_starttag.start(), match_starttag.end(), match_endtag.start(), match_endtag.end()
+
+
 def parse_googlemap(wikitext, detail=False):
     """Parses the (unicode) u'<googlemap ...>content</googlemap>' of the googlemap extension
     out of a page. If wikitext does not contain the googlemap extension text None is returned.
@@ -159,13 +201,10 @@ def parse_googlemap(wikitext, detail=False):
         if not match is None: return (float(match.group(2)), float(match.group(1)), None, None)
         return RuntimeError(u'Could not parse line ' + line)
 
-    regexp = re.compile(u"(<googlemap[^>]*>)(.*?)(</googlemap>)", re.DOTALL)
-    match = regexp.search(wikitext)
-    if match is None: return None
-    start = match.start()
-    end = match.end()
-    content = match.group(2)
-    gm = xml.etree.ElementTree.XML((match.group(1)+match.group(3)).encode('UTF8'))
+    start, content, endtag, end = find_tag(wikitext, 'googlemap')
+    if content is None:
+        return None
+    gm = xml.etree.ElementTree.XML((wikitext[start:content]+wikitext[endtag:end]).encode('UTF8'))
     zoom = gm.get('zoom')
     lon = gm.get('lon')
     lat = gm.get('lat')
@@ -176,7 +215,7 @@ def parse_googlemap(wikitext, detail=False):
 
     coords = []
     paths = []
-    lines = content.split("\n")
+    lines = wikitext[content:endtag].split("\n")
     i = 0
     while i < len(lines):
         line = lines[i].strip()