diff options
Diffstat (limited to 'sleekxmpp/xmlstream/matcher')
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/__init__.py | 16 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/base.py | 31 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/id.py | 29 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/many.py | 40 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/stanzapath.py | 43 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/xmlmask.py | 158 | ||||
| -rw-r--r-- | sleekxmpp/xmlstream/matcher/xpath.py | 86 |
7 files changed, 403 insertions, 0 deletions
diff --git a/sleekxmpp/xmlstream/matcher/__init__.py b/sleekxmpp/xmlstream/matcher/__init__.py new file mode 100644 index 00000000..1038d1bd --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/__init__.py @@ -0,0 +1,16 @@ +""" + SleekXMPP: The Sleek XMPP Library + Copyright (C) 2010 Nathanael C. Fritz + This file is part of SleekXMPP. + + See the file LICENSE for copying permission. +""" + +from sleekxmpp.xmlstream.matcher.id import MatcherId +from sleekxmpp.xmlstream.matcher.many import MatchMany +from sleekxmpp.xmlstream.matcher.stanzapath import StanzaPath +from sleekxmpp.xmlstream.matcher.xmlmask import MatchXMLMask +from sleekxmpp.xmlstream.matcher.xpath import MatchXPath + +__all__ = ['MatcherId', 'MatchMany', 'StanzaPath', + 'MatchXMLMask', 'MatchXPath'] diff --git a/sleekxmpp/xmlstream/matcher/base.py b/sleekxmpp/xmlstream/matcher/base.py new file mode 100644 index 00000000..83c26688 --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/base.py @@ -0,0 +1,31 @@ +# -*- coding: utf-8 -*- +""" + sleekxmpp.xmlstream.matcher.base + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + Part of SleekXMPP: The Sleek XMPP Library + + :copyright: (c) 2011 Nathanael C. Fritz + :license: MIT, see LICENSE for more details +""" + + +class MatcherBase(object): + + """ + Base class for stanza matchers. Stanza matchers are used to pick + stanzas out of the XML stream and pass them to the appropriate + stream handlers. + + :param criteria: Object to compare some aspect of a stanza against. + """ + + def __init__(self, criteria): + self._criteria = criteria + + def match(self, xml): + """Check if a stanza matches the stored criteria. + + Meant to be overridden. + """ + return False diff --git a/sleekxmpp/xmlstream/matcher/id.py b/sleekxmpp/xmlstream/matcher/id.py new file mode 100644 index 00000000..11ab70bb --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/id.py @@ -0,0 +1,29 @@ +# -*- coding: utf-8 -*- +""" + sleekxmpp.xmlstream.matcher.id + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + Part of SleekXMPP: The Sleek XMPP Library + + :copyright: (c) 2011 Nathanael C. Fritz + :license: MIT, see LICENSE for more details +""" + +from sleekxmpp.xmlstream.matcher.base import MatcherBase + + +class MatcherId(MatcherBase): + + """ + The ID matcher selects stanzas that have the same stanza 'id' + interface value as the desired ID. + """ + + def match(self, xml): + """Compare the given stanza's ``'id'`` attribute to the stored + ``id`` value. + + :param xml: The :class:`~sleekxmpp.xmlstream.stanzabase.ElementBase` + stanza to compare against. + """ + return xml['id'] == self._criteria diff --git a/sleekxmpp/xmlstream/matcher/many.py b/sleekxmpp/xmlstream/matcher/many.py new file mode 100644 index 00000000..f470ec9c --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/many.py @@ -0,0 +1,40 @@ +""" + SleekXMPP: The Sleek XMPP Library + Copyright (C) 2010 Nathanael C. Fritz + This file is part of SleekXMPP. + + See the file LICENSE for copying permission. +""" + +from sleekxmpp.xmlstream.matcher.base import MatcherBase + + +class MatchMany(MatcherBase): + + """ + The MatchMany matcher may compare a stanza against multiple + criteria. It is essentially an OR relation combining multiple + matchers. + + Each of the criteria must implement a match() method. + + Methods: + match -- Overrides MatcherBase.match. + """ + + def match(self, xml): + """ + Match a stanza against multiple criteria. The match is successful + if one of the criteria matches. + + Each of the criteria must implement a match() method. + + Overrides MatcherBase.match. + + Arguments: + xml -- The stanza object to compare against. + """ + for m in self._criteria: + if m.match(xml): + return True + return False diff --git a/sleekxmpp/xmlstream/matcher/stanzapath.py b/sleekxmpp/xmlstream/matcher/stanzapath.py new file mode 100644 index 00000000..a4c0fda0 --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/stanzapath.py @@ -0,0 +1,43 @@ +# -*- coding: utf-8 -*- +""" + sleekxmpp.xmlstream.matcher.stanzapath + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + Part of SleekXMPP: The Sleek XMPP Library + + :copyright: (c) 2011 Nathanael C. Fritz + :license: MIT, see LICENSE for more details +""" + +from sleekxmpp.xmlstream.matcher.base import MatcherBase +from sleekxmpp.xmlstream.stanzabase import fix_ns + + +class StanzaPath(MatcherBase): + + """ + The StanzaPath matcher selects stanzas that match a given "stanza path", + which is similar to a normal XPath except that it uses the interfaces and + plugins of the stanza instead of the actual, underlying XML. + + :param criteria: Object to compare some aspect of a stanza against. + """ + + def __init__(self, criteria): + self._criteria = fix_ns(criteria, split=True, + propagate_ns=False, + default_ns='jabber:client') + self._raw_criteria = criteria + + def match(self, stanza): + """ + Compare a stanza against a "stanza path". A stanza path is similar to + an XPath expression, but uses the stanza's interfaces and plugins + instead of the underlying XML. See the documentation for the stanza + :meth:`~sleekxmpp.xmlstream.stanzabase.ElementBase.match()` method + for more information. + + :param stanza: The :class:`~sleekxmpp.xmlstream.stanzabase.ElementBase` + stanza to compare against. + """ + return stanza.match(self._criteria) or stanza.match(self._raw_criteria) diff --git a/sleekxmpp/xmlstream/matcher/xmlmask.py b/sleekxmpp/xmlstream/matcher/xmlmask.py new file mode 100644 index 00000000..7977e767 --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/xmlmask.py @@ -0,0 +1,158 @@ +""" + SleekXMPP: The Sleek XMPP Library + Copyright (C) 2010 Nathanael C. Fritz + This file is part of SleekXMPP. + + See the file LICENSE for copying permission. +""" + +import logging + +from xml.parsers.expat import ExpatError + +from sleekxmpp.xmlstream.stanzabase import ET +from sleekxmpp.xmlstream.matcher.base import MatcherBase + + +# Flag indicating if the builtin XPath matcher should be used, which +# uses namespaces, or a custom matcher that ignores namespaces. +# Changing this will affect ALL XMLMask matchers. +IGNORE_NS = False + + +log = logging.getLogger(__name__) + + +class MatchXMLMask(MatcherBase): + + """ + The XMLMask matcher selects stanzas whose XML matches a given + XML pattern, or mask. For example, message stanzas with body elements + could be matched using the mask: + + .. code-block:: xml + + <message xmlns="jabber:client"><body /></message> + + Use of XMLMask is discouraged, and + :class:`~sleekxmpp.xmlstream.matcher.xpath.MatchXPath` or + :class:`~sleekxmpp.xmlstream.matcher.stanzapath.StanzaPath` + should be used instead. + + The use of namespaces in the mask comparison is controlled by + ``IGNORE_NS``. Setting ``IGNORE_NS`` to ``True`` will disable namespace + based matching for ALL XMLMask matchers. + + :param criteria: Either an :class:`~xml.etree.ElementTree.Element` XML + object or XML string to use as a mask. + """ + + def __init__(self, criteria): + MatcherBase.__init__(self, criteria) + if isinstance(criteria, str): + self._criteria = ET.fromstring(self._criteria) + self.default_ns = 'jabber:client' + + def setDefaultNS(self, ns): + """Set the default namespace to use during comparisons. + + :param ns: The new namespace to use as the default. + """ + self.default_ns = ns + + def match(self, xml): + """Compare a stanza object or XML object against the stored XML mask. + + Overrides MatcherBase.match. + + :param xml: The stanza object or XML object to compare against. + """ + if hasattr(xml, 'xml'): + xml = xml.xml + return self._mask_cmp(xml, self._criteria, True) + + def _mask_cmp(self, source, mask, use_ns=False, default_ns='__no_ns__'): + """Compare an XML object against an XML mask. + + :param source: The :class:`~xml.etree.ElementTree.Element` XML object + to compare against the mask. + :param mask: The :class:`~xml.etree.ElementTree.Element` XML object + serving as the mask. + :param use_ns: Indicates if namespaces should be respected during + the comparison. + :default_ns: The default namespace to apply to elements that + do not have a specified namespace. + Defaults to ``"__no_ns__"``. + """ + use_ns = not IGNORE_NS + + if source is None: + # If the element was not found. May happend during recursive calls. + return False + + # Convert the mask to an XML object if it is a string. + if not hasattr(mask, 'attrib'): + try: + mask = ET.fromstring(mask) + except ExpatError: + log.warning("Expat error: %s\nIn parsing: %s", '', mask) + if not use_ns: + # Compare the element without using namespaces. + source_tag = source.tag.split('}', 1)[-1] + mask_tag = mask.tag.split('}', 1)[-1] + if source_tag != mask_tag: + return False + else: + # Compare the element using namespaces + mask_ns_tag = "{%s}%s" % (self.default_ns, mask.tag) + if source.tag not in [mask.tag, mask_ns_tag]: + return False + + # If the mask includes text, compare it. + if mask.text and source.text and \ + source.text.strip() != mask.text.strip(): + return False + + # Compare attributes. The stanza must include the attributes + # defined by the mask, but may include others. + for name, value in mask.attrib.items(): + if source.attrib.get(name, "__None__") != value: + return False + + # Recursively check subelements. + matched_elements = {} + for subelement in mask: + if use_ns: + matched = False + for other in source.findall(subelement.tag): + matched_elements[other] = False + if self._mask_cmp(other, subelement, use_ns): + if not matched_elements.get(other, False): + matched_elements[other] = True + matched = True + if not matched: + return False + else: + if not self._mask_cmp(self._get_child(source, subelement.tag), + subelement, use_ns): + return False + + # Everything matches. + return True + + def _get_child(self, xml, tag): + """Return a child element given its tag, ignoring namespace values. + + Returns ``None`` if the child was not found. + + :param xml: The :class:`~xml.etree.ElementTree.Element` XML object + to search for the given child tag. + :param tag: The name of the subelement to find. + """ + tag = tag.split('}')[-1] + try: + children = [c.tag.split('}')[-1] for c in xml.getchildren()] + index = children.index(tag) + except ValueError: + return None + return xml.getchildren()[index] diff --git a/sleekxmpp/xmlstream/matcher/xpath.py b/sleekxmpp/xmlstream/matcher/xpath.py new file mode 100644 index 00000000..b6af0609 --- /dev/null +++ b/sleekxmpp/xmlstream/matcher/xpath.py @@ -0,0 +1,86 @@ +# -*- coding: utf-8 -*- +""" + sleekxmpp.xmlstream.matcher.xpath + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + Part of SleekXMPP: The Sleek XMPP Library + + :copyright: (c) 2011 Nathanael C. Fritz + :license: MIT, see LICENSE for more details +""" + +from sleekxmpp.xmlstream.stanzabase import ET +from sleekxmpp.xmlstream.matcher.base import MatcherBase + + +# Flag indicating if the builtin XPath matcher should be used, which +# uses namespaces, or a custom matcher that ignores namespaces. +# Changing this will affect ALL XPath matchers. +IGNORE_NS = False + + +class MatchXPath(MatcherBase): + + """ + The XPath matcher selects stanzas whose XML contents matches a given + XPath expression. + + .. warning:: + + Using this matcher may not produce expected behavior when using + attribute selectors. For Python 2.6 and 3.1, the ElementTree + :meth:`~xml.etree.ElementTree.Element.find()` method does + not support the use of attribute selectors. If you need to + support Python 2.6 or 3.1, it might be more useful to use a + :class:`~sleekxmpp.xmlstream.matcher.stanzapath.StanzaPath` matcher. + + If the value of :data:`IGNORE_NS` is set to ``True``, then XPath + expressions will be matched without using namespaces. + """ + + def match(self, xml): + """ + Compare a stanza's XML contents to an XPath expression. + + If the value of :data:`IGNORE_NS` is set to ``True``, then XPath + expressions will be matched without using namespaces. + + .. warning:: + + In Python 2.6 and 3.1 the ElementTree + :meth:`~xml.etree.ElementTree.Element.find()` method does not + support attribute selectors in the XPath expression. + + :param xml: The :class:`~sleekxmpp.xmlstream.stanzabase.ElementBase` + stanza to compare against. + """ + if hasattr(xml, 'xml'): + xml = xml.xml + x = ET.Element('x') + x.append(xml) + + if not IGNORE_NS: + # Use builtin, namespace respecting, XPath matcher. + if x.find(self._criteria) is not None: + return True + return False + else: + # Remove namespaces from the XPath expression. + criteria = [] + for ns_block in self._criteria.split('{'): + criteria.extend(ns_block.split('}')[-1].split('/')) + + # Walk the XPath expression. + xml = x + for tag in criteria: + if not tag: + # Skip empty tag name artifacts from the cleanup phase. + continue + + children = [c.tag.split('}')[-1] for c in xml.getchildren()] + try: + index = children.index(tag) + except ValueError: + return False + xml = xml.getchildren()[index] + return True |