1 #!/usr/bin/python 2 3 # 4 # Copyright (C) 2012 The Android Open Source Project 5 # 6 # Licensed under the Apache License, Version 2.0 (the "License"); 7 # you may not use this file except in compliance with the License. 8 # You may obtain a copy of the License at 9 # 10 # http://www.apache.org/licenses/LICENSE-2.0 11 # 12 # Unless required by applicable law or agreed to in writing, software 13 # distributed under the License is distributed on an "AS IS" BASIS, 14 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 15 # See the License for the specific language governing permissions and 16 # limitations under the License. 17 # 18 19 """ 20 Usage: 21 metadata_validate.py <filename.xml> 22 - validates that the metadata properties defined in filename.xml are 23 semantically correct. 24 - does not do any XSD validation, use xmllint for that (in metadata-validate) 25 26 Module: 27 A set of helpful functions for dealing with BeautifulSoup element trees. 28 Especially the find_* and fully_qualified_name functions. 29 30 Dependencies: 31 BeautifulSoup - an HTML/XML parser available to download from 32 http://www.crummy.com/software/BeautifulSoup/ 33 """ 34 35 from bs4 import BeautifulSoup 36 from bs4 import Tag 37 import sys 38 39 40 ##################### 41 ##################### 42 43 def fully_qualified_name(entry): 44 """ 45 Calculates the fully qualified name for an entry by walking the path 46 to the root node. 47 48 Args: 49 entry: a BeautifulSoup Tag corresponding to an <entry ...> XML node, 50 or a <clone ...> XML node. 51 52 Raises: 53 ValueError: if entry does not correspond to one of the above XML nodes 54 55 Returns: 56 A string with the full name, e.g. "android.lens.info.availableApertureSizes" 57 """ 58 59 filter_tags = ['namespace', 'section'] 60 parents = [i['name'] for i in entry.parents if i.name in filter_tags] 61 62 if entry.name == 'entry': 63 name = entry['name'] 64 elif entry.name == 'clone': 65 name = entry['entry'].split(".")[-1] # "a.b.c" => "c" 66 else: 67 raise ValueError("Unsupported tag type '%s' for element '%s'" \ 68 %(entry.name, entry)) 69 70 parents.reverse() 71 parents.append(name) 72 73 fqn = ".".join(parents) 74 75 return fqn 76 77 def find_parent_by_name(element, names): 78 """ 79 Find the ancestor for an element whose name matches one of those 80 in names. 81 82 Args: 83 element: A BeautifulSoup Tag corresponding to an XML node 84 85 Returns: 86 A BeautifulSoup element corresponding to the matched parent, or None. 87 88 For example, assuming the following XML structure: 89 <static> 90 <anything> 91 <entry name="Hello" /> # this is in variable 'Hello' 92 </anything> 93 </static> 94 95 el = find_parent_by_name(Hello, ['static']) 96 # el is now a value pointing to the '<static>' element 97 """ 98 matching_parents = [i.name for i in element.parents if i.name in names] 99 100 if matching_parents: 101 return matching_parents[0] 102 else: 103 return None 104 105 def find_all_child_tags(element, tag): 106 """ 107 Finds all the children that are a Tag (as opposed to a NavigableString), 108 with a name of tag. This is useful to filter out the NavigableString out 109 of the children. 110 111 Args: 112 element: A BeautifulSoup Tag corresponding to an XML node 113 tag: A string representing the name of the tag 114 115 Returns: 116 A list of Tag instances 117 118 For example, given the following XML structure: 119 <enum> # This is the variable el 120 Hello world # NavigableString 121 <value>Apple</value> # this is the variale apple (Tag) 122 <value>Orange</value> # this is the variable orange (Tag) 123 Hello world again # NavigableString 124 </enum> 125 126 lst = find_all_child_tags(el, 'value') 127 # lst is [apple, orange] 128 129 """ 130 matching_tags = [i for i in element.children if isinstance(i, Tag) and i.name == tag] 131 return matching_tags 132 133 def find_child_tag(element, tag): 134 """ 135 Finds the first child that is a Tag with the matching name. 136 137 Args: 138 element: a BeautifulSoup Tag 139 tag: A String representing the name of the tag 140 141 Returns: 142 An instance of a Tag, or None if there was no matches. 143 144 For example, given the following XML structure: 145 <enum> # This is the variable el 146 Hello world # NavigableString 147 <value>Apple</value> # this is the variale apple (Tag) 148 <value>Orange</value> # this is the variable orange (Tag) 149 Hello world again # NavigableString 150 </enum> 151 152 res = find_child_tag(el, 'value') 153 # res is apple 154 """ 155 matching_tags = find_all_child_tags(element, tag) 156 if matching_tags: 157 return matching_tags[0] 158 else: 159 return None 160 161 def find_kind(element): 162 """ 163 Finds the kind Tag ancestor for an element. 164 165 Args: 166 element: a BeautifulSoup Tag 167 168 Returns: 169 a BeautifulSoup tag, or None if there was no matches 170 171 Remarks: 172 This function only makes sense to be called for an Entry, Clone, or 173 InnerNamespace XML types. It will always return 'None' for other nodes. 174 """ 175 kinds = ['dynamic', 'static', 'controls'] 176 parent_kind = find_parent_by_name(element, kinds) 177 return parent_kind 178 179 def validate_error(msg): 180 """ 181 Print a validation error to stderr. 182 183 Args: 184 msg: a string you want to be printed 185 """ 186 print >> sys.stderr, "ERROR: " + msg 187 188 189 def validate_clones(soup): 190 """ 191 Validate that all <clone> elements point to an existing <entry> element. 192 193 Args: 194 soup - an instance of BeautifulSoup 195 196 Returns: 197 True if the validation succeeds, False otherwise 198 """ 199 success = True 200 201 for clone in soup.find_all("clone"): 202 clone_entry = clone['entry'] 203 clone_kind = clone['kind'] 204 205 parent_kind = find_kind(clone) 206 207 find_entry = lambda x: x.name == 'entry' \ 208 and find_kind(x) == clone_kind \ 209 and fully_qualified_name(x) == clone_entry 210 matching_entry = soup.find(find_entry) 211 212 if matching_entry is None: 213 error_msg = ("Did not find corresponding clone entry '%s' " + \ 214 "with kind '%s'") %(clone_entry, clone_kind) 215 validate_error(error_msg) 216 success = False 217 218 clone_name = fully_qualified_name(clone) 219 if clone_name != clone_entry: 220 error_msg = ("Clone entry target '%s' did not match fully qualified " + \ 221 "name '%s'.") %(clone_entry, clone_name) 222 validate_error(error_msg) 223 success = False 224 225 return success 226 227 # All <entry> elements with container=$foo have a <$foo> child 228 # If type="enum", <enum> tag is present 229 # In <enum> for all <value id="$x">, $x is numeric 230 def validate_entries(soup): 231 """ 232 Validate all <entry> elements with the following rules: 233 * If there is a container="$foo" attribute, there is a <$foo> child 234 * If there is a type="enum" attribute, there is an <enum> child 235 * In the <enum> child, all <value id="$x"> have a numeric $x 236 237 Args: 238 soup - an instance of BeautifulSoup 239 240 Returns: 241 True if the validation succeeds, False otherwise 242 """ 243 success = True 244 for entry in soup.find_all("entry"): 245 entry_container = entry.attrs.get('container') 246 247 if entry_container is not None: 248 container_tag = entry.find(entry_container) 249 250 if container_tag is None: 251 success = False 252 validate_error(("Entry '%s' in kind '%s' has type '%s' but " + \ 253 "missing child element <%s>") \ 254 %(fully_qualified_name(entry), find_kind(entry), \ 255 entry_container, entry_container)) 256 257 enum = entry.attrs.get('enum') 258 if enum and enum == 'true': 259 if entry.enum is None: 260 validate_error(("Entry '%s' in kind '%s' is missing enum") \ 261 % (fully_qualified_name(entry), find_kind(entry), 262 )) 263 success = False 264 265 else: 266 for value in entry.enum.find_all('value'): 267 value_id = value.attrs.get('id') 268 269 if value_id is not None: 270 try: 271 id_int = int(value_id, 0) #autoguess base 272 except ValueError: 273 validate_error(("Entry '%s' has id '%s', which is not" + \ 274 " numeric.") \ 275 %(fully_qualified_name(entry), value_id)) 276 success = False 277 else: 278 if entry.enum: 279 validate_error(("Entry '%s' kind '%s' has enum el, but no enum attr") \ 280 % (fully_qualified_name(entry), find_kind(entry), 281 )) 282 success = False 283 284 return success 285 286 def validate_xml(xml): 287 """ 288 Validate all XML nodes according to the rules in validate_clones and 289 validate_entries. 290 291 Args: 292 xml - A string containing a block of XML to validate 293 294 Returns: 295 a BeautifulSoup instance if validation succeeds, None otherwise 296 """ 297 298 soup = BeautifulSoup(xml, features='xml') 299 300 succ = validate_clones(soup) 301 succ = validate_entries(soup) and succ 302 303 if succ: 304 return soup 305 else: 306 return None 307 308 ##################### 309 ##################### 310 311 if __name__ == "__main__": 312 if len(sys.argv) <= 1: 313 print >> sys.stderr, "Usage: %s <filename.xml>" % (sys.argv[0]) 314 sys.exit(0) 315 316 file_name = sys.argv[1] 317 succ = validate_xml(file(file_name).read()) is not None 318 319 if succ: 320 print "%s: SUCCESS! Document validated" %(file_name) 321 sys.exit(0) 322 else: 323 print >> sys.stderr, "%s: ERRORS: Document failed to validate" %(file_name) 324 sys.exit(1) 325
