Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 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 |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 36 | from bs4 import Tag |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 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 | |
| 51 | Returns: |
| 52 | A string with the full name, e.g. "android.lens.info.availableApertureSizes" |
| 53 | """ |
| 54 | filter_tags = ['namespace', 'section'] |
| 55 | parents = [i['name'] for i in entry.parents if i.name in filter_tags] |
| 56 | |
| 57 | name = entry['name'] |
| 58 | |
| 59 | parents.reverse() |
| 60 | parents.append(name) |
| 61 | |
| 62 | fqn = ".".join(parents) |
| 63 | |
| 64 | return fqn |
| 65 | |
| 66 | def find_parent_by_name(element, names): |
| 67 | """ |
| 68 | Find the ancestor for an element whose name matches one of those |
| 69 | in names. |
| 70 | |
| 71 | Args: |
| 72 | element: A BeautifulSoup Tag corresponding to an XML node |
| 73 | |
| 74 | Returns: |
| 75 | A BeautifulSoup element corresponding to the matched parent, or None. |
| 76 | |
| 77 | For example, assuming the following XML structure: |
| 78 | <static> |
| 79 | <anything> |
| 80 | <entry name="Hello" /> # this is in variable 'Hello' |
| 81 | </anything> |
| 82 | </static> |
| 83 | |
| 84 | el = find_parent_by_name(Hello, ['static']) |
| 85 | # el is now a value pointing to the '<static>' element |
| 86 | """ |
| 87 | matching_parents = [i.name for i in element.parents if i.name in names] |
| 88 | |
| 89 | if matching_parents: |
| 90 | return matching_parents[0] |
| 91 | else: |
| 92 | return None |
| 93 | |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 94 | def find_all_child_tags(element, tag): |
| 95 | """ |
| 96 | Finds all the children that are a Tag (as opposed to a NavigableString), |
| 97 | with a name of tag. This is useful to filter out the NavigableString out |
| 98 | of the children. |
| 99 | |
| 100 | Args: |
| 101 | element: A BeautifulSoup Tag corresponding to an XML node |
| 102 | tag: A string representing the name of the tag |
| 103 | |
| 104 | Returns: |
| 105 | A list of Tag instances |
| 106 | |
| 107 | For example, given the following XML structure: |
| 108 | <enum> # This is the variable el |
| 109 | Hello world # NavigableString |
| 110 | <value>Apple</value> # this is the variale apple (Tag) |
| 111 | <value>Orange</value> # this is the variable orange (Tag) |
| 112 | Hello world again # NavigableString |
| 113 | </enum> |
| 114 | |
| 115 | lst = find_all_child_tags(el, 'value') |
| 116 | # lst is [apple, orange] |
| 117 | |
| 118 | """ |
| 119 | matching_tags = [i for i in element.children if isinstance(i, Tag) and i.name == tag] |
| 120 | return matching_tags |
| 121 | |
| 122 | def find_child_tag(element, tag): |
| 123 | """ |
| 124 | Finds the first child that is a Tag with the matching name. |
| 125 | |
| 126 | Args: |
| 127 | element: a BeautifulSoup Tag |
| 128 | tag: A String representing the name of the tag |
| 129 | |
| 130 | Returns: |
| 131 | An instance of a Tag, or None if there was no matches. |
| 132 | |
| 133 | For example, given the following XML structure: |
| 134 | <enum> # This is the variable el |
| 135 | Hello world # NavigableString |
| 136 | <value>Apple</value> # this is the variale apple (Tag) |
| 137 | <value>Orange</value> # this is the variable orange (Tag) |
| 138 | Hello world again # NavigableString |
| 139 | </enum> |
| 140 | |
| 141 | res = find_child_tag(el, 'value') |
| 142 | # res is apple |
| 143 | """ |
| 144 | matching_tags = find_all_child_tags(element, tag) |
| 145 | if matching_tags: |
| 146 | return matching_tags[0] |
| 147 | else: |
| 148 | return None |
| 149 | |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 150 | def find_kind(element): |
| 151 | """ |
| 152 | Finds the kind Tag ancestor for an element. |
| 153 | |
| 154 | Args: |
| 155 | element: a BeautifulSoup Tag |
| 156 | |
| 157 | Returns: |
| 158 | a BeautifulSoup tag, or None if there was no matches |
| 159 | |
| 160 | Remarks: |
| 161 | This function only makes sense to be called for an Entry, Clone, or |
| 162 | InnerNamespace XML types. It will always return 'None' for other nodes. |
| 163 | """ |
| 164 | kinds = ['dynamic', 'static', 'controls'] |
| 165 | parent_kind = find_parent_by_name(element, kinds) |
| 166 | return parent_kind |
| 167 | |
| 168 | def validate_error(msg): |
| 169 | """ |
| 170 | Print a validation error to stderr. |
| 171 | |
| 172 | Args: |
| 173 | msg: a string you want to be printed |
| 174 | """ |
| 175 | print >> sys.stderr, "Validation error: " + msg |
| 176 | |
| 177 | |
| 178 | def validate_clones(soup): |
| 179 | """ |
| 180 | Validate that all <clone> elements point to an existing <entry> element. |
| 181 | |
| 182 | Args: |
| 183 | soup - an instance of BeautifulSoup |
| 184 | |
| 185 | Returns: |
| 186 | True if the validation succeeds, False otherwise |
| 187 | """ |
| 188 | success = True |
| 189 | |
| 190 | for clone in soup.find_all("clone"): |
| 191 | clone_entry = clone['entry'] |
| 192 | clone_kind = clone['kind'] |
| 193 | |
| 194 | parent_kind = find_kind(clone) |
| 195 | |
| 196 | find_entry = lambda x: x.name == 'entry' \ |
| 197 | and find_kind(x) == clone_kind \ |
| 198 | and fully_qualified_name(x) == clone_entry |
| 199 | matching_entry = soup.find(find_entry) |
| 200 | |
| 201 | if matching_entry is None: |
| 202 | error_msg = ("Did not find corresponding clone entry '%s' " + \ |
| 203 | "with kind '%s'") %(clone_entry, clone_kind) |
| 204 | validate_error(error_msg) |
| 205 | success = False |
| 206 | |
| 207 | return success |
| 208 | |
| 209 | # All <entry> elements with container=$foo have a <$foo> child |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 210 | # If type="enum", <enum> tag is present |
| 211 | # In <enum> for all <value id="$x">, $x is numeric |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 212 | def validate_entries(soup): |
| 213 | """ |
| 214 | Validate all <entry> elements with the following rules: |
| 215 | * If there is a container="$foo" attribute, there is a <$foo> child |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 216 | * If there is a type="enum" attribute, there is an <enum> child |
| 217 | * In the <enum> child, all <value id="$x"> have a numeric $x |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 218 | |
| 219 | Args: |
| 220 | soup - an instance of BeautifulSoup |
| 221 | |
| 222 | Returns: |
| 223 | True if the validation succeeds, False otherwise |
| 224 | """ |
| 225 | success = True |
| 226 | for entry in soup.find_all("entry"): |
| 227 | entry_container = entry.attrs.get('container') |
| 228 | |
| 229 | if entry_container is not None: |
| 230 | container_tag = entry.find(entry_container) |
| 231 | |
| 232 | if container_tag is None: |
| 233 | success = False |
| 234 | validate_error(("Entry '%s' in kind '%s' has type '%s' but " + \ |
| 235 | "missing child element <%s>") \ |
| 236 | %(fully_qualified_name(entry), find_kind(entry), \ |
| 237 | entry_container, entry_container)) |
| 238 | |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 239 | typ = entry.attrs.get('type') |
| 240 | if typ == 'enum': |
| 241 | if entry.enum is None: |
| 242 | validate_error(("Entry '%s' in kind '%s' is missing enum") \ |
| 243 | % (fully_qualified_name(entry), find_kind(entry), |
| 244 | )) |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 245 | |
Igor Murashkin | 96bd019 | 2012-11-19 16:49:37 -0800 | [diff] [blame^] | 246 | if typ == 'enum' and entry.enum is not None: |
| 247 | |
| 248 | for value in entry.enum.find_all('value'): |
| 249 | value_id = value.attrs.get('id') |
| 250 | |
| 251 | if value_id is not None: |
| 252 | try: |
| 253 | id_int = int(value_id, 0) #autoguess base |
| 254 | except ValueError: |
| 255 | validate_error(("Entry '%s' has id '%s', which is not" + \ |
| 256 | " numeric.") \ |
| 257 | %(fully_qualified_name(entry), value_id)) |
| 258 | success = False |
| 259 | |
| 260 | return success |
Igor Murashkin | 77b63ca | 2012-11-09 16:15:02 -0800 | [diff] [blame] | 261 | |
| 262 | def validate_xml(file_name): |
| 263 | """ |
| 264 | Validate all XML nodes according to the rules in validate_clones and |
| 265 | validate_entries. |
| 266 | |
| 267 | Args: |
| 268 | file_name - a string path to an XML file we wish to validate |
| 269 | |
| 270 | Returns: |
| 271 | a BeautifulSoup instance if validation succeeds, None otherwise |
| 272 | """ |
| 273 | |
| 274 | xml = file(file_name).read() |
| 275 | soup = BeautifulSoup(xml, features='xml') |
| 276 | |
| 277 | succ = validate_clones(soup) |
| 278 | succ = validate_entries(soup) and succ |
| 279 | |
| 280 | if succ: |
| 281 | return soup |
| 282 | else: |
| 283 | return None |
| 284 | |
| 285 | ##################### |
| 286 | ##################### |
| 287 | |
| 288 | if __name__ == "__main__": |
| 289 | if len(sys.argv) <= 1: |
| 290 | print >> sys.stderr, "Usage: %s <filename.xml>" % (sys.argv[0]) |
| 291 | sys.exit(0) |
| 292 | |
| 293 | file_name = sys.argv[1] |
| 294 | succ = validate_xml(file_name) is not None |
| 295 | |
| 296 | if succ: |
| 297 | print "%s: SUCCESS! Document validated" %(file_name) |
| 298 | sys.exit(0) |
| 299 | else: |
| 300 | print >> sys.stderr, "%s: ERRORS: Document failed to validate" %(file_name) |
| 301 | sys.exit(1) |