From 036dd24098fafa518fd849bbaf57006b262f4d16 Mon Sep 17 00:00:00 2001 From: brent s Date: Mon, 28 Oct 2019 14:11:35 -0400 Subject: [PATCH] gorram it pycharm, stop warning me about this gorram dollar sign. it doesn't need to and shouldn't be escaped in XSD pattern expressions. --- aif.xsd | 8 ++--- aif/config.py | 99 ++++++++++++++++++++++++++++++++++----------------- docs/THANKS | 2 +- 3 files changed, 71 insertions(+), 38 deletions(-) diff --git a/aif.xsd b/aif.xsd index 800be9a..ac992ac 100644 --- a/aif.xsd +++ b/aif.xsd @@ -147,10 +147,10 @@ - - - - + + + + diff --git a/aif/config.py b/aif/config.py index 4a5b234..135a7f7 100644 --- a/aif/config.py +++ b/aif/config.py @@ -4,27 +4,29 @@ import re import requests from lxml import etree -# https://stackoverflow.com/questions/30232031/how-can-i-strip-namespaces-out-of-an-lxml-tree/30233635#30233635 ? - _patterns = {'raw': re.compile(r'^\s*(?P<(\?xml|aif)\s+.*)\s*$', re.DOTALL|re.MULTILINE), 'remote': re.compile(r'^(?P(?P(https?|ftps?)://)(?P.*))\s*$'), 'local': re.compile(r'^(file://)?(?P(/?[^/]+)+/?)$')} + class Config(object): - def __init__(self, *args, **kwargs): + def __init__(self, xsd_path = None, *args, **kwargs): + self.xsd_path = None self.tree = None self.namespaced_tree = None self.xml = None self.namespaced_xml = None self.raw = None self.xsd = None + self.defaultsParser = None - def main(self, validate = True): + def main(self, validate = True, populate_defaults = True): self.fetch() self.parseRaw() + if populate_defaults: + self.populateDefaults() if validate: self.validate() - return() def fetch(self): # Just a fail-safe; this is overridden by specific subclasses. @@ -32,6 +34,8 @@ class Config(object): return() def getXSD(self, xsdpath = None): + if not xsdpath: + xsdpath = self.xsd_path raw_xsd = None if xsdpath: xsdpath = os.path.abspath(os.path.expanduser(xsdpath)) @@ -45,20 +49,41 @@ class Config(object): schemaLocation = '{{{0}}}schemaLocation'.format(xsi) schemaURL = self.xml.attrib.get(schemaLocation, 'https://aif-ng.io/aif.xsd?ref={0}'.format(self.xml.attrib['version'])) + split_url = schemaURL.split() + if len(split_url) == 2: # a properly defined schemaLocation + schemaURL = split_url[1] + else: + schemaURL = split_url[0] req = requests.get(schemaURL) - if not req.ok(): + if not req.ok: + # TODO: logging! raise RuntimeError('Could not download XSD') raw_xsd = req.content self.xsd = etree.XMLSchema(etree.XML(raw_xsd)) return() - def parseRaw(self): - self.tree = etree.parse(self.raw) - self.namespaced_tree = etree.parse(self.raw) - self.xml = self.tree.getroot() - self.namespaced_xml = self.namespaced_tree.getroot() + def parseRaw(self, parser = None): + # self.xml = etree.parse(self.raw, parser = parser) + self.xml = etree.fromstring(self.raw, parser = parser) + # self.namespaced_xml = etree.parse(self.raw, parser = parser) + self.namespaced_xml = etree.fromstring(self.raw, parser = parser) + self.tree = self.xml.getroottree() + self.namespaced_tree = self.namespaced_xml.getroottree() self.tree.xinclude() self.namespaced_tree.xinclude() + self.stripNS() + return() + + def populateDefaults(self): + if not self.xsd: + self.getXSD() + if not self.defaultsParser: + self.defaultsParser = etree.XMLParser(schema = self.xsd, attribute_defaults = True) + self.parseRaw(parser = self.defaultsParser) + return() + + def removeDefaults(self): + self.parseRaw() return() def stripNS(self): @@ -71,31 +96,34 @@ class Config(object): def validate(self): if not self.xsd: self.getXSD() - + self.xsd.assertValid(self.tree) + self.xsd.assertValid(self.namespaced_tree) return() class LocalFile(Config): - def __init__(self, path, *args, **kwargs): - super().__init__(*args, **kwargs) - self.source = ['local', _patterns['local'].search(path).group('path')] + def __init__(self, path, xsd_path = None, *args, **kwargs): + super().__init__(xsd_path = xsd_path, *args, **kwargs) + self.type = 'local' + self.source = _patterns['local'].search(path).group('path') def fetch(self): - self.source[1] = os.path.abspath(os.path.expanduser(self.source[1])) - if not os.path.isfile(self.source[1]): - raise ValueError('{0} does not exist'.format(self.source[1])) - with open(self.source[1], 'rb') as fh: + self.source = os.path.realpath(self.source) + if not os.path.isfile(self.source): + raise ValueError('{0} does not exist'.format(self.source)) + with open(self.source, 'rb') as fh: self.raw = fh.read() return() class RemoteFile(Config): - def __init__(self, uri, *args, **kwargs): + def __init__(self, uri, xsd_path = None, *args, **kwargs): super().__init__(*args, **kwargs) - self.source = ('remote', uri) + self.type = 'remote' + self.source = uri def fetch(self): - r = requests.get(self.source[1]) + r = requests.get(self.source) if not r.ok(): raise RuntimeError('Could not download XML') self.raw = r.content @@ -103,36 +131,39 @@ class RemoteFile(Config): class ConfigStr(Config): - def __init__(self, rawxml, *args, **kwargs): + def __init__(self, rawxml, xsd_path = None, *args, **kwargs): super().__init__(*args, **kwargs) - self.source = ('raw', rawxml) + self.type = 'raw_str' + self.source = rawxml def fetch(self): - self.raw = self.source[1].encode('utf-8') + self.raw = self.source.encode('utf-8') return() + class ConfigBin(Config): - def __init__(self, rawbinaryxml, *args, **kwargs): + def __init__(self, rawbinaryxml, xsd_path = None, *args, **kwargs): super().__init__(*args, **kwargs) - self.source = ('raw_binary', rawbinaryxml) + self.type = 'raw_bin' + self.source = rawbinaryxml def fetch(self): - self.raw = self.source[1] + self.raw = self.source return() -def getConfig(cfg_ref, validate = True): +def getConfig(cfg_ref, validate = True, populate_defaults = True, xsd_path = None): cfgobj = None # This is kind of gross. for configtype, pattern in _patterns.items(): try: if pattern.search(cfg_ref): if configtype == 'raw': - cfgobj = ConfigStr(cfg_ref) + cfgobj = ConfigStr(cfg_ref, xsd_path = xsd_path) elif configtype == 'remote': - cfgobj = RemoteFile(cfg_ref) + cfgobj = RemoteFile(cfg_ref, xsd_path = xsd_path) elif configtype == 'local': - cfgobj = LocalFile(cfg_ref) + cfgobj = LocalFile(cfg_ref, xsd_path = xsd_path) if cfgobj: break except TypeError: @@ -140,6 +171,8 @@ def getConfig(cfg_ref, validate = True): if not ptrn.search(cfg_ref): raise ValueError('Received junk data for cfg_ref') else: - cfgobj = ConfigBin(cfg_ref) + cfgobj = ConfigBin(cfg_ref, xsd_path = xsd_path) break + if cfgobj: + cfgobj.main(validate = validate, populate_defaults = populate_defaults) return(cfgobj) diff --git a/docs/THANKS b/docs/THANKS index 547a037..ae35677 100644 --- a/docs/THANKS +++ b/docs/THANKS @@ -1,3 +1,3 @@ AIF-NG owes thanks to: -* jthan for being a good rubber ducky \ No newline at end of file +* jthan for being a good rubber ducky