| 1 |
# SOME DESCRIPTIVE TITLE. |
| 2 |
# Copyright (C) 2020, The JSpecify Authors |
| 3 |
# This file is distributed under the same license as the JSpecify package. |
| 4 |
# FIRST AUTHOR <EMAIL@ADDRESS>, 2021. |
| 5 |
# |
| 6 |
#, fuzzy |
| 7 |
msgid "" |
| 8 |
msgstr "" |
| 9 |
"Project-Id-Version: JSpecify \n" |
| 10 |
"Report-Msgid-Bugs-To: \n" |
| 11 |
"POT-Creation-Date: 2021-02-16 00:56+0000\n" |
| 12 |
"PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n" |
| 13 |
"Last-Translator: FULL NAME <EMAIL@ADDRESS>\n" |
| 14 |
"Language-Team: LANGUAGE <[email protected]>\n" |
| 15 |
"MIME-Version: 1.0\n" |
| 16 |
"Content-Type: text/plain; charset=utf-8\n" |
| 17 |
"Content-Transfer-Encoding: 8bit\n" |
| 18 |
"Generated-By: Babel 2.8.0\n" |
| 19 |
|
| 20 |
#: ../../tsttcpw.rst:2 |
| 21 |
msgid "The Simplest(?) Thing That Could Possibly Work for subtyping" |
| 22 |
msgstr "" |
| 23 |
|
| 24 |
#: ../../tsttcpw.rst:4 |
| 25 |
msgid "" |
| 26 |
"**This is not “the JSpecify spec.” This is an initial attempt to formally" |
| 27 |
" specify only a subset of the rules we’ll need for a subset of features " |
| 28 |
"we wish to cover. Additionally, it deviates from some of our current " |
| 29 |
"working decisions in an effort to remain simple.**" |
| 30 |
msgstr "" |
| 31 |
|
| 32 |
#: ../../tsttcpw.rst |
| 33 |
msgid "High-level overview" |
| 34 |
msgstr "" |
| 35 |
|
| 36 |
#: ../../tsttcpw.rst:12 |
| 37 |
msgid "" |
| 38 |
"It may be that some people will use this doc to guide their " |
| 39 |
"implementations in the near future. Please don’t hesitate to let cpovirk " |
| 40 |
"know of any confusing bits." |
| 41 |
msgstr "" |
| 42 |
|
| 43 |
#: ../../tsttcpw.rst:16 |
| 44 |
msgid "" |
| 45 |
"I should probably preemptively clarify at least one thing. In this doc, I" |
| 46 |
" have tried to distinguish explicitly between 3 “kinds of nullability” of" |
| 47 |
" a given type usage. Each kind is derived (at least in part) from the " |
| 48 |
"previous:" |
| 49 |
msgstr "" |
| 50 |
|
| 51 |
#: ../../tsttcpw.rst:21 |
| 52 |
msgid "What annotation (if any) appears directly on that type usage?" |
| 53 |
msgstr "" |
| 54 |
|
| 55 |
#: ../../tsttcpw.rst:22 |
| 56 |
msgid "What is the `nullness operator <#nullness-operator>`__ of that type usage?" |
| 57 |
msgstr "" |
| 58 |
|
| 59 |
#: ../../tsttcpw.rst:24 |
| 60 |
msgid "For that type usage…" |
| 61 |
msgstr "" |
| 62 |
|
| 63 |
#: ../../tsttcpw.rst:26 |
| 64 |
msgid "Is it safe to assume that is not ``null``?" |
| 65 |
msgstr "" |
| 66 |
|
| 67 |
#: ../../tsttcpw.rst:27 |
| 68 |
msgid "Is it safe to put a ``null`` into it?" |
| 69 |
msgstr "" |
| 70 |
|
| 71 |
#: ../../tsttcpw.rst:28 |
| 72 |
msgid "neither (as in “parametric nullness”)" |
| 73 |
msgstr "" |
| 74 |
|
| 75 |
#: ../../tsttcpw.rst:29 |
| 76 |
msgid "both (as in “unspecified nullness” in “lenient mode”)" |
| 77 |
msgstr "" |
| 78 |
|
| 79 |
#: ../../tsttcpw.rst:31 |
| 80 |
msgid "TODO(cpovirk): Link to my “Don’t say ‘nullable’” doc once I write it." |
| 81 |
msgstr "" |
| 82 |
|
| 83 |
#: ../../tsttcpw.rst:36 |
| 84 |
msgid "References to concepts defined by this spec" |
| 85 |
msgstr "" |
| 86 |
|
| 87 |
#: ../../tsttcpw.rst:38 |
| 88 |
msgid "" |
| 89 |
"When a rule in this spec refers to any concept that is defined in this " |
| 90 |
"spec (for example, `substitution <#substitution>`__ or `containment " |
| 91 |
"<#containment>`__), apply this spec’s definition (as opposed to other " |
| 92 |
"definitions, such as the ones in the JLS)." |
| 93 |
msgstr "" |
| 94 |
|
| 95 |
#: ../../tsttcpw.rst:43 |
| 96 |
msgid "" |
| 97 |
"Additionally, when a rule in this spec refers to a JLS rule that in turn " |
| 98 |
"refers to a concept that is defined in this spec, likewise apply this " |
| 99 |
"spec’s definition." |
| 100 |
msgstr "" |
| 101 |
|
| 102 |
#: ../../tsttcpw.rst:47 |
| 103 |
msgid "" |
| 104 |
"In particular, when a JLS rule refers to types, apply this spec’s " |
| 105 |
"definition of `augmented types <#augmented-type>`__ (as oppposed to `base" |
| 106 |
" types <https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 107 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__)." |
| 108 |
msgstr "" |
| 109 |
|
| 110 |
#: ../../tsttcpw.rst:53 |
| 111 |
msgid "Nullness operator" |
| 112 |
msgstr "" |
| 113 |
|
| 114 |
#: ../../tsttcpw.rst:55 |
| 115 |
msgid "An nullness operator is one of 4 values:" |
| 116 |
msgstr "" |
| 117 |
|
| 118 |
#: ../../tsttcpw.rst:57 |
| 119 |
msgid "``UNION_NULL``" |
| 120 |
msgstr "" |
| 121 |
|
| 122 |
#: ../../tsttcpw.rst:58 |
| 123 |
msgid "``NO_CHANGE``" |
| 124 |
msgstr "" |
| 125 |
|
| 126 |
#: ../../tsttcpw.rst:59 |
| 127 |
msgid "``UNSPECIFIED``" |
| 128 |
msgstr "" |
| 129 |
|
| 130 |
#: ../../tsttcpw.rst:60 |
| 131 |
msgid "``MINUS_NULL``" |
| 132 |
msgstr "" |
| 133 |
|
| 134 |
#: ../../tsttcpw.rst:64 |
| 135 |
msgid "" |
| 136 |
"The distinction among these 4 values is similar to the distinction among " |
| 137 |
"the Kotlin types ``Foo?``, ``Foo``, ``Foo!``, and ``Foo!!``, " |
| 138 |
"respectively." |
| 139 |
msgstr "" |
| 140 |
|
| 141 |
#: ../../tsttcpw.rst:69 |
| 142 |
msgid "Augmented type" |
| 143 |
msgstr "" |
| 144 |
|
| 145 |
#: ../../tsttcpw.rst:71 |
| 146 |
msgid "" |
| 147 |
"An augmented type consists of a `base type " |
| 148 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 149 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ and a " |
| 150 |
"`nullness operator <#nullness-operator>`__ corresponding to *each* of its" |
| 151 |
" `type components <https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf" |
| 152 |
"--w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.g7gl9fwq1tt5>`__." |
| 153 |
msgstr "" |
| 154 |
|
| 155 |
#: ../../tsttcpw.rst:77 |
| 156 |
msgid "" |
| 157 |
"This differs from our current `glossary definition " |
| 158 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 159 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=id.367l48xhsikk>`__, which is " |
| 160 |
"written in terms of a “nullness” for each component, not a “nullness " |
| 161 |
"operator.” Still, the glossary’s concept of the “nullness” of a type is " |
| 162 |
"derivable from the type’s nullness operator. Notably, the glossary’s " |
| 163 |
"“nullable” type is our `least convenient world <#multiple-worlds>`__\\ ’s" |
| 164 |
" `trusted null-inclusive under every parameterization <#trusted-null-" |
| 165 |
"inclusive-under-every-parameterization>`__, and the glossary’s “non-" |
| 166 |
"nullable” type is our least convenient world’s `trusted null-exclusive " |
| 167 |
"under every parameterization <#trusted-null-exclusive-under-every-" |
| 168 |
"parameterization>`__." |
| 169 |
msgstr "" |
| 170 |
|
| 171 |
#: ../../tsttcpw.rst:89 |
| 172 |
msgid "" |
| 173 |
"For our purposes, base types (and thus augmented types) include not just " |
| 174 |
"class and interface types, array types, and type variables but also " |
| 175 |
"`intersection types <#intersection-types>`__ and the null type. This is " |
| 176 |
"true even though the JLS sometimes does not supply rules for intersection" |
| 177 |
" types and sometimes has separate rules for the null type." |
| 178 |
msgstr "" |
| 179 |
|
| 180 |
#: ../../tsttcpw.rst:95 |
| 181 |
msgid "" |
| 182 |
"The goal of this spec is to define rules for augmented types compatible " |
| 183 |
"with those that the JLS defines for base types." |
| 184 |
msgstr "" |
| 185 |
|
| 186 |
#: ../../tsttcpw.rst:98 |
| 187 |
msgid "" |
| 188 |
"In almost all cases, this spec agrees with the JLS’s rules when " |
| 189 |
"specifying what *base* types appear in a piece of code. It makes an " |
| 190 |
"exception for `“Bound of an unbounded wildcard,” <#unbounded-wildcard>`__" |
| 191 |
" for which it specifies a bound of ``Object`` that the JLS does not " |
| 192 |
"specify." |
| 193 |
msgstr "" |
| 194 |
|
| 195 |
#: ../../tsttcpw.rst:104 |
| 196 |
msgid "" |
| 197 |
"When this spec uses capital letters, they refer to augmented types " |
| 198 |
"(unless otherwise noted). This is in contrast to the JLS, which typically" |
| 199 |
" uses them to refer to base types." |
| 200 |
msgstr "" |
| 201 |
|
| 202 |
#: ../../tsttcpw.rst:108 |
| 203 |
msgid "" |
| 204 |
"When this spec refers to “the nullness operator of” a type ``T``, it " |
| 205 |
"refers specifically to the nullness operator of the type component that " |
| 206 |
"is the entire type ``T``, without reference to the nullness operator of " |
| 207 |
"any other type components of ``T``." |
| 208 |
msgstr "" |
| 209 |
|
| 210 |
#: ../../tsttcpw.rst:113 |
| 211 |
msgid "" |
| 212 |
"For example, the nullness operator of ``List<@Nullable Object>`` would be" |
| 213 |
" ``NO_CHANGE`` (at least in a `null-aware context <#null-aware-" |
| 214 |
"context>`__), even though the nullness operator of its element type " |
| 215 |
"``Object`` is ``UNION_NULL``." |
| 216 |
msgstr "" |
| 217 |
|
| 218 |
#: ../../tsttcpw.rst:119 |
| 219 |
msgid "Null-aware context" |
| 220 |
msgstr "" |
| 221 |
|
| 222 |
#: ../../tsttcpw.rst:121 |
| 223 |
msgid "To determine whether a type usage appears in a null-aware context:" |
| 224 |
msgstr "" |
| 225 |
|
| 226 |
#: ../../tsttcpw.rst:123 |
| 227 |
msgid "" |
| 228 |
"Look for an ``@org.jspecify.annotations.NullAware`` annotation on any of " |
| 229 |
"the enclosing scopes surrounding the type usage." |
| 230 |
msgstr "" |
| 231 |
|
| 232 |
#: ../../tsttcpw.rst:126 |
| 233 |
msgid "" |
| 234 |
"Class members are enclosed by classes, which may be enclosed by other " |
| 235 |
"class members or classes. and top-level classes are enclosed by packages," |
| 236 |
" which may be enclosed by modules." |
| 237 |
msgstr "" |
| 238 |
|
| 239 |
#: ../../tsttcpw.rst:130 |
| 240 |
msgid "Packages are *not* enclosed by “parent” packages." |
| 241 |
msgstr "" |
| 242 |
|
| 243 |
#: ../../tsttcpw.rst:134 |
| 244 |
msgid "" |
| 245 |
"This definition of “enclosing” likely matches `the definition in the Java" |
| 246 |
" compiler API " |
| 247 |
"<https://docs.oracle.com/en/java/javase/14/docs/api/java.compiler/javax/lang/model/element/Element.html#getEnclosingElement()>`__." |
| 248 |
msgstr "" |
| 249 |
|
| 250 |
#: ../../tsttcpw.rst:138 |
| 251 |
msgid "" |
| 252 |
"If an ``@org.jspecify.annotations.NullAware`` annotation exists on one of" |
| 253 |
" these scopes, then the type usage is in a null-aware context. Otherwise," |
| 254 |
" it is not." |
| 255 |
msgstr "" |
| 256 |
|
| 257 |
#: ../../tsttcpw.rst:145 |
| 258 |
msgid "Augmented type of a type usage appearing in code" |
| 259 |
msgstr "" |
| 260 |
|
| 261 |
#: ../../tsttcpw.rst:147 |
| 262 |
msgid "" |
| 263 |
"For most type usages in source code or bytecode on which JSpecify " |
| 264 |
"nullness annotations are structurally valid, this section defines how to " |
| 265 |
"determine their `augmented types <#augmented-type>`__. Note, however, " |
| 266 |
"that rules for specific cases below take precedence over the general rule" |
| 267 |
" here." |
| 268 |
msgstr "" |
| 269 |
|
| 270 |
#: ../../tsttcpw.rst:153 |
| 271 |
msgid "" |
| 272 |
"Because the JLS already has rules for determining the `base type " |
| 273 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 274 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ for a type " |
| 275 |
"usage, this section covers only how to determine its `nullness operator " |
| 276 |
"<#nullness-operator>`__." |
| 277 |
msgstr "" |
| 278 |
|
| 279 |
#: ../../tsttcpw.rst:158 |
| 280 |
msgid "" |
| 281 |
"To determine the nullness operator, apply the following rules in order. " |
| 282 |
"Once one condition is met, skip the remaining conditions." |
| 283 |
msgstr "" |
| 284 |
|
| 285 |
#: ../../tsttcpw.rst:161 |
| 286 |
msgid "" |
| 287 |
"If the type usage is annotated with " |
| 288 |
"``@org.jspecify.annotations.Nullable``, its nullness operator is " |
| 289 |
"``UNION_NULL``." |
| 290 |
msgstr "" |
| 291 |
|
| 292 |
#: ../../tsttcpw.rst:164 |
| 293 |
msgid "" |
| 294 |
"If the type usage appears in a `null-aware context <#null-aware-" |
| 295 |
"context>`__, its nullness operator is ``NO_CHANGE``." |
| 296 |
msgstr "" |
| 297 |
|
| 298 |
#: ../../tsttcpw.rst:167 |
| 299 |
msgid "Its nullness operator is ``UNSPECIFIED``." |
| 300 |
msgstr "" |
| 301 |
|
| 302 |
#: ../../tsttcpw.rst:171 |
| 303 |
msgid "" |
| 304 |
"The choice of nullness operator is *not* affected by any nullness " |
| 305 |
"operator that appears in a corresponding location in a supertype. For " |
| 306 |
"example, if one type declares a method whose return type is annotated " |
| 307 |
"``@Nullable``, and if another type overrides that method but does not " |
| 308 |
"declare the return type as ``@Nullable``, then the override’s return type" |
| 309 |
" will *not* have nullness operator ``UNION_NULL``." |
| 310 |
msgstr "" |
| 311 |
|
| 312 |
#: ../../tsttcpw.rst:178 |
| 313 |
msgid "" |
| 314 |
"The rules here never produce the fourth nullness operator, " |
| 315 |
"``MINUS_NULL``. (It will appear later in `substitution " |
| 316 |
"<#substitution>`__. Additionally, we expect for tool authors to produce " |
| 317 |
"``MINUS_NULL`` based on the results of null checks in implementation " |
| 318 |
"code.) However, if tool authors prefer, they can safely produce " |
| 319 |
"``MINUS_NULL`` in any case in which it is equivalent to ``NO_CHANGE``. " |
| 320 |
"For example, there is no difference between a ``String`` with " |
| 321 |
"``NO_CHANGE`` and a ``String`` with ``MINUS_NULL``." |
| 322 |
msgstr "" |
| 323 |
|
| 324 |
#: ../../tsttcpw.rst:190 |
| 325 |
msgid "Augmented type of an intersection type" |
| 326 |
msgstr "" |
| 327 |
|
| 328 |
#: ../../tsttcpw.rst:192 |
| 329 |
msgid "" |
| 330 |
"Technically speaking, the JLS does not define syntax for an intersection " |
| 331 |
"type. Instead, it defines a syntax for type parameters and casts that " |
| 332 |
"supports multiple types. Then the intersection type is derived from " |
| 333 |
"those. Intersection types can also arise from operations like `capture " |
| 334 |
"conversion <#capture-conversion>`__. See `JLS 4.9 " |
| 335 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-4.html#jls-4.9>`__." |
| 336 |
msgstr "" |
| 337 |
|
| 338 |
#: ../../tsttcpw.rst:199 |
| 339 |
msgid "" |
| 340 |
"One result of all this is that it’s never possible for a programmer to " |
| 341 |
"write an annotation “on an intersection type.”" |
| 342 |
msgstr "" |
| 343 |
|
| 344 |
#: ../../tsttcpw.rst:202 |
| 345 |
msgid "" |
| 346 |
"This spec assigns a `nullness operator <#nullness-operator>`__ to each " |
| 347 |
"individual element of an intersection type, following our normal rules " |
| 348 |
"for type usages. It also assigns a nullness operator to the intersection " |
| 349 |
"type as a whole. The nullness operator of the type as a whole is always " |
| 350 |
"``NO_CHANGE``." |
| 351 |
msgstr "" |
| 352 |
|
| 353 |
#: ../../tsttcpw.rst:208 |
| 354 |
msgid "" |
| 355 |
"This lets us provide, for every `base type " |
| 356 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 357 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__, a rule for " |
| 358 |
"computing its `augmented type <#augmented-type>`__. But we require " |
| 359 |
"``NO_CHANGE`` so as to avoid questions like whether “a ``UNION_NULL`` " |
| 360 |
"intersection type whose members are ``UNION_NULL`` ``Foo`` and " |
| 361 |
"``UNION_NULL`` ``Bar``” is a subtype of “a ``NO_CHANGE`` intersection " |
| 362 |
"type with those same members.” Plus, it would be difficult for tools to " |
| 363 |
"output the nullness operator of an intersection type in a human-readable " |
| 364 |
"way." |
| 365 |
msgstr "" |
| 366 |
|
| 367 |
#: ../../tsttcpw.rst:220 |
| 368 |
msgid "" |
| 369 |
"To avoid ever creating an intersection type with a nullness operator " |
| 370 |
"other than ``NO_CHANGE``, we define special handling for intersection " |
| 371 |
"types under `“Applying a nullness operator to an augmented type.” " |
| 372 |
"<#applying-operator>`__" |
| 373 |
msgstr "" |
| 374 |
|
| 375 |
#: ../../tsttcpw.rst:228 |
| 376 |
msgid "Bound of an “unbounded” wildcard" |
| 377 |
msgstr "" |
| 378 |
|
| 379 |
#: ../../tsttcpw.rst:230 |
| 380 |
msgid "" |
| 381 |
"In source, an unbounded wildcard is written as ``<?>``. This section does" |
| 382 |
" **not** apply to ``<? extends Object>``, even though that is often " |
| 383 |
"equivalent to ``<?>``. See `JLS 4.5.1 " |
| 384 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-4.html#jls-4.5.1>`__." |
| 385 |
msgstr "" |
| 386 |
|
| 387 |
#: ../../tsttcpw.rst:235 |
| 388 |
msgid "" |
| 389 |
"In bytecode, such a wildcard is represented as a wildcard type with an " |
| 390 |
"empty list of upper bounds and an empty list of lower bounds. This " |
| 391 |
"section does **not** apply to a wildcard with any bounds in either list, " |
| 392 |
"even a sole upper bound of ``Object``." |
| 393 |
msgstr "" |
| 394 |
|
| 395 |
#: ../../tsttcpw.rst:240 |
| 396 |
msgid "" |
| 397 |
"For a wildcard with an explicit bound of ``Object`` (that is, ``<? " |
| 398 |
"extends Object>``, perhaps with an annotation on ``Object``), instead " |
| 399 |
"apply `the normal rules <#augmented-type-of-usage>`__ for the explicit " |
| 400 |
"bound type." |
| 401 |
msgstr "" |
| 402 |
|
| 403 |
#: ../../tsttcpw.rst:245 |
| 404 |
msgid "" |
| 405 |
"If an unbounded wildcard appears in a `null-aware context <#null-aware-" |
| 406 |
"context>`__, then it has a single upper bound whose `base type " |
| 407 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 408 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ is ``Object`` " |
| 409 |
"and whose `nullness operator <#nullness-operator>`__ is ``UNION_NULL``." |
| 410 |
msgstr "" |
| 411 |
|
| 412 |
#: ../../tsttcpw.rst:252 |
| 413 |
msgid "" |
| 414 |
"If an unbounded wildcard appears outside a null-aware context, then it " |
| 415 |
"has a single upper bound whose base type is ``Object`` and whose nullness" |
| 416 |
" operator is ``UNSPECIFIED``." |
| 417 |
msgstr "" |
| 418 |
|
| 419 |
#: ../../tsttcpw.rst:256 |
| 420 |
msgid "" |
| 421 |
"In both cases, we specify a bound that does not exist in the source or " |
| 422 |
"bytecode, deviating from the JLS. Because the base type of the bound is " |
| 423 |
"``Object``, this should produce no user-visible differences except to " |
| 424 |
"tools that implement JSpecify nullness analysis." |
| 425 |
msgstr "" |
| 426 |
|
| 427 |
#: ../../tsttcpw.rst:261 |
| 428 |
msgid "" |
| 429 |
"Whenever a JLS rule refers specifically to ``<?>``, disregard it, and " |
| 430 |
"instead apply the rules for ``<? extends T>``, where ``T`` has a base " |
| 431 |
"type of ``Object`` and the nullness operator defined by this section." |
| 432 |
msgstr "" |
| 433 |
|
| 434 |
#: ../../tsttcpw.rst:268 |
| 435 |
msgid "Bound of an ``Object``-bounded type parameter" |
| 436 |
msgstr "" |
| 437 |
|
| 438 |
#: ../../tsttcpw.rst:270 |
| 439 |
msgid "" |
| 440 |
"In source, an ``Object``-bounded type parameter can be writen in either " |
| 441 |
"of 2 ways:" |
| 442 |
msgstr "" |
| 443 |
|
| 444 |
#: ../../tsttcpw.rst:273 |
| 445 |
msgid "``<T>``" |
| 446 |
msgstr "" |
| 447 |
|
| 448 |
#: ../../tsttcpw.rst:274 |
| 449 |
msgid "" |
| 450 |
"``<T extends Object>`` with no JSpecify nullness type annotations on the " |
| 451 |
"bound" |
| 452 |
msgstr "" |
| 453 |
|
| 454 |
#: ../../tsttcpw.rst:277 |
| 455 |
msgid "" |
| 456 |
"See `JLS 4.4 " |
| 457 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-4.html#jls-4.4>`__." |
| 458 |
msgstr "" |
| 459 |
|
| 460 |
#: ../../tsttcpw.rst:280 |
| 461 |
msgid "" |
| 462 |
"In bytecode, ``<T>`` and ``<T extends Object>`` are both represented as a" |
| 463 |
" type parameter with only a single upper bound, ``Object``, and no " |
| 464 |
"JSpecify nullness type annotations on the bound." |
| 465 |
msgstr "" |
| 466 |
|
| 467 |
#: ../../tsttcpw.rst:284 |
| 468 |
msgid "" |
| 469 |
"If an ``Object``-bounded type parameter appears in a `null-aware context " |
| 470 |
"<#null-aware-context>`__, then its bound has a `base type " |
| 471 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 472 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ of ``Object`` " |
| 473 |
"and a `nullness operator <#nullness-operator>`__ of ``NO_CHANGE``." |
| 474 |
msgstr "" |
| 475 |
|
| 476 |
#: ../../tsttcpw.rst:290 |
| 477 |
msgid "" |
| 478 |
"Note that this gives ``<T>`` a different bound than ``<?>`` (though only " |
| 479 |
"in a null-aware context)." |
| 480 |
msgstr "" |
| 481 |
|
| 482 |
#: ../../tsttcpw.rst:293 |
| 483 |
msgid "" |
| 484 |
"If an ``Object``-bounded type parameter appears outside a null-aware " |
| 485 |
"context, then its bound has a base type of ``Object`` and a nullness " |
| 486 |
"operator of ``UNSPECIFIED``." |
| 487 |
msgstr "" |
| 488 |
|
| 489 |
#: ../../tsttcpw.rst:297 |
| 490 |
msgid "" |
| 491 |
"All these rules match the behavior of `our normal rules <#augmented-type-" |
| 492 |
"of-usage>`__ for determining the `augmented type <#augmented-type>`__ of " |
| 493 |
"the bound ``Object``. The only “special” part is that we consider the " |
| 494 |
"source code ``<T>`` to have a bound of ``Object``, just as it does when " |
| 495 |
"compiled to bytecode." |
| 496 |
msgstr "" |
| 497 |
|
| 498 |
#: ../../tsttcpw.rst:306 |
| 499 |
msgid "Augmented null types" |
| 500 |
msgstr "" |
| 501 |
|
| 502 |
#: ../../tsttcpw.rst:308 |
| 503 |
msgid "" |
| 504 |
"The JLS refers to “the null type.” In this spec, we assign a `nullness " |
| 505 |
"operator <#nullness-operator>`__ to all types, including the null type. " |
| 506 |
"This produces multiple null types:" |
| 507 |
msgstr "" |
| 508 |
|
| 509 |
#: ../../tsttcpw.rst:312 |
| 510 |
msgid "" |
| 511 |
"the null `base type " |
| 512 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 513 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ with nullness " |
| 514 |
"operator ``NO_CHANGE``: the “bottom”/“nothing” type used in `capture " |
| 515 |
"conversion <#capture-conversion>`__" |
| 516 |
msgstr "" |
| 517 |
|
| 518 |
#: ../../tsttcpw.rst:317 |
| 519 |
msgid "No value, including ``null`` itself, has this type." |
| 520 |
msgstr "" |
| 521 |
|
| 522 |
#: ../../tsttcpw.rst:319 |
| 523 |
msgid "the null base type with nullness operator ``MINUS_NULL``" |
| 524 |
msgstr "" |
| 525 |
|
| 526 |
#: ../../tsttcpw.rst:321 |
| 527 |
msgid "" |
| 528 |
"This is equivalent to the previous type. Tools may use the 2 " |
| 529 |
"interchangeably." |
| 530 |
msgstr "" |
| 531 |
|
| 532 |
#: ../../tsttcpw.rst:324 |
| 533 |
msgid "" |
| 534 |
"the null base type with nullness operator ``UNION_NULL``: the type of the" |
| 535 |
" null reference" |
| 536 |
msgstr "" |
| 537 |
|
| 538 |
#: ../../tsttcpw.rst:327 |
| 539 |
msgid "the null base type with nullness operator ``UNSPECIFIED``" |
| 540 |
msgstr "" |
| 541 |
|
| 542 |
#: ../../tsttcpw.rst:329 |
| 543 |
msgid "This may be relevant only in implementation code." |
| 544 |
msgstr "" |
| 545 |
|
| 546 |
#: ../../tsttcpw.rst:334 |
| 547 |
msgid "The least convenient world and the most convenient world" |
| 548 |
msgstr "" |
| 549 |
|
| 550 |
#: ../../tsttcpw.rst:336 |
| 551 |
msgid "" |
| 552 |
"Some of the rules in this spec come in 2 versions, 1 for “the least " |
| 553 |
"convenient world” and 1 for “the most convenient world.”" |
| 554 |
msgstr "" |
| 555 |
|
| 556 |
#: ../../tsttcpw.rst:339 |
| 557 |
msgid "Tools may implement either or both versions of the rules." |
| 558 |
msgstr "" |
| 559 |
|
| 560 |
#: ../../tsttcpw.rst:341 |
| 561 |
msgid "" |
| 562 |
"Our goal is to allow tools and their users to choose their desired level " |
| 563 |
"of strictness in the presence of ``UNSPECIFIED``. “The least convenient " |
| 564 |
"world” usually assumes that types are incompatible unless it has enough " |
| 565 |
"information to prove they are compatible; “the most convenient world” " |
| 566 |
"assumes that types are compatible unless it has enough information to " |
| 567 |
"prove they are incompatible." |
| 568 |
msgstr "" |
| 569 |
|
| 570 |
#: ../../tsttcpw.rst:348 |
| 571 |
msgid "" |
| 572 |
"Thus, strict tools may want to implement the least-convenient-world " |
| 573 |
"version of rules, and lenient tools may wish to implement the most-" |
| 574 |
"convenient-world version. Or a tool might implement both and let users " |
| 575 |
"select which rules to apply." |
| 576 |
msgstr "" |
| 577 |
|
| 578 |
#: ../../tsttcpw.rst:353 |
| 579 |
msgid "" |
| 580 |
"Another possibility is for a tool to implement both versions and to use " |
| 581 |
"that to distinguish between “errors” and “warnings.” Such a tool might " |
| 582 |
"run each check first in the least convenient world and then, if the check" |
| 583 |
" fails, run it again in the most convenient world. If the check fails in " |
| 584 |
"both worlds, the tool would produce an error. If it passes only because " |
| 585 |
"of the most convenient interpretation, the tool would produce a warning." |
| 586 |
msgstr "" |
| 587 |
|
| 588 |
#: ../../tsttcpw.rst:361 |
| 589 |
msgid "" |
| 590 |
"The main body of each section describes the *least*-convenient-world " |
| 591 |
"rule. If the most-convenient-world rule differs, the differences are " |
| 592 |
"explained at the end." |
| 593 |
msgstr "" |
| 594 |
|
| 595 |
#: ../../tsttcpw.rst:368 |
| 596 |
msgid "Propagating the most/least convenient world" |
| 597 |
msgstr "" |
| 598 |
|
| 599 |
#: ../../tsttcpw.rst:370 |
| 600 |
msgid "" |
| 601 |
"When one rule in this spec refers to another, it refers to the rule for " |
| 602 |
"the same “world.” For example, when the rules for `containment " |
| 603 |
"<#containment>`__ refer to the rules for `subtyping <#subtyping>`__, the " |
| 604 |
"most-convenient-world containment check applies the most-convenient-world" |
| 605 |
" subtyping check, and the least-convenient-world containment check " |
| 606 |
"applies the least-convenient-world subtyping check." |
| 607 |
msgstr "" |
| 608 |
|
| 609 |
#: ../../tsttcpw.rst:378 |
| 610 |
msgid "" |
| 611 |
"This applies even if a rule says it is the same for both worlds: It means" |
| 612 |
" “the same except that any other rules are applied in the corresponding " |
| 613 |
"world.”" |
| 614 |
msgstr "" |
| 615 |
|
| 616 |
#: ../../tsttcpw.rst:383 |
| 617 |
msgid "Same type" |
| 618 |
msgstr "" |
| 619 |
|
| 620 |
#: ../../tsttcpw.rst:385 |
| 621 |
msgid "" |
| 622 |
"``S`` and ``T`` are the same type if ``S`` is a `subtype <#subtyping>`__ " |
| 623 |
"of ``T`` and ``T`` is a subtype of ``S``." |
| 624 |
msgstr "" |
| 625 |
|
| 626 |
#: ../../tsttcpw.rst:389 |
| 627 |
msgid "Subtyping" |
| 628 |
msgstr "" |
| 629 |
|
| 630 |
#: ../../tsttcpw.rst:391 |
| 631 |
msgid "``A`` is a subtype of ``F`` if both of the following conditions are met:" |
| 632 |
msgstr "" |
| 633 |
|
| 634 |
#: ../../tsttcpw.rst:393 |
| 635 |
msgid "" |
| 636 |
"``A`` is a subtype of ``F`` according to the `nullness-delegating " |
| 637 |
"subtyping rules for Java <#nullness-delegating-subtyping>`__." |
| 638 |
msgstr "" |
| 639 |
|
| 640 |
#: ../../tsttcpw.rst:395 |
| 641 |
msgid "``A`` is a `nullness subtype <#nullness-subtyping>`__ of ``F``." |
| 642 |
msgstr "" |
| 643 |
|
| 644 |
#: ../../tsttcpw.rst:400 |
| 645 |
msgid "Nullness-delegating subtyping rules for Java" |
| 646 |
msgstr "" |
| 647 |
|
| 648 |
#: ../../tsttcpw.rst:402 |
| 649 |
msgid "" |
| 650 |
"The Java subtyping rules are defined in `JLS 4.10 " |
| 651 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-4.html#jls-4.10>`__." |
| 652 |
" We add to them as follows:" |
| 653 |
msgstr "" |
| 654 |
|
| 655 |
#: ../../tsttcpw.rst:406 |
| 656 |
msgid "" |
| 657 |
"`As always <#concept-references>`__, interpret the Java rules as " |
| 658 |
"operating on `augmented types <#augmented-type>`__, not `base types " |
| 659 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 660 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__. However, when" |
| 661 |
" applying the Java direct-supertype rules themselves, *ignore* the " |
| 662 |
"`nullness operator <#nullness-operator>`__ of the input types and output " |
| 663 |
"types. The augmented types matter only when the Java rules refer to " |
| 664 |
"*other* rules that are defined in this spec. *Those* rules respect the " |
| 665 |
"nullness operator of some type components — but never the nullness " |
| 666 |
"operator of the type component that represents the whole input or output " |
| 667 |
"type." |
| 668 |
msgstr "" |
| 669 |
|
| 670 |
#: ../../tsttcpw.rst:417 |
| 671 |
msgid "" |
| 672 |
"To “ignore” the output’s nullness operator, we recommend outputting a " |
| 673 |
"value of ``NO_CHANGE``, since that is valid for all types, including " |
| 674 |
"`intersection types <#intersection-types>`__." |
| 675 |
msgstr "" |
| 676 |
|
| 677 |
#: ../../tsttcpw.rst:421 |
| 678 |
msgid "" |
| 679 |
"When the Java array rules require one type to be a *direct* supertype of " |
| 680 |
"another, consider the direct supertypes of ``T`` to be *every* type that " |
| 681 |
"``T`` is a `subtype <#subtyping>`__ of (as always, applying the " |
| 682 |
"definition of subtyping in this spec)." |
| 683 |
msgstr "" |
| 684 |
|
| 685 |
#: ../../tsttcpw.rst:427 |
| 686 |
msgid "Nullness subtyping" |
| 687 |
msgstr "" |
| 688 |
|
| 689 |
#: ../../tsttcpw.rst:429 |
| 690 |
msgid "" |
| 691 |
"The primary complication in subtyping comes from type-variable usages. " |
| 692 |
"Our rules for them must account for every combination of type arguments " |
| 693 |
"with which a given generic type can be parameterized." |
| 694 |
msgstr "" |
| 695 |
|
| 696 |
#: ../../tsttcpw.rst:433 |
| 697 |
msgid "" |
| 698 |
"``A`` is a nullness subtype of ``F`` if any of the following conditions " |
| 699 |
"are met:" |
| 700 |
msgstr "" |
| 701 |
|
| 702 |
#: ../../tsttcpw.rst:436 |
| 703 |
msgid "" |
| 704 |
"``F`` is `trusted null-inclusive under every parameterization <#trusted-" |
| 705 |
"null-inclusive-under-every-parameterization>`__." |
| 706 |
msgstr "" |
| 707 |
|
| 708 |
#: ../../tsttcpw.rst:438 |
| 709 |
msgid "" |
| 710 |
"``A`` is `trusted null-exclusive under every parameterization <#trusted-" |
| 711 |
"null-exclusive-under-every-parameterization>`__." |
| 712 |
msgstr "" |
| 713 |
|
| 714 |
#: ../../tsttcpw.rst:440 |
| 715 |
msgid "" |
| 716 |
"``A`` has a `nullness-subtype-establishing path <#nullness-subtype-" |
| 717 |
"establishing-path>`__ to any type whose base type is the same as the base" |
| 718 |
" type of ``F``." |
| 719 |
msgstr "" |
| 720 |
|
| 721 |
#: ../../tsttcpw.rst:444 |
| 722 |
msgid "Nullness subtyping (and thus subtyping itself) is *not* transitive." |
| 723 |
msgstr "" |
| 724 |
|
| 725 |
#: ../../tsttcpw.rst:446 |
| 726 |
msgid "" |
| 727 |
"(Contrast this with our `nullness-delegating subtyping <#nullness-" |
| 728 |
"delegating-subtyping>`__ rules and `containment <#containment>`__ rules: " |
| 729 |
"Each of those is defined as a transitive closure. However, technically " |
| 730 |
"speaking, `there are cases in which those should not be transitive, " |
| 731 |
"either <https://groups.google.com/d/msg/jspecify-" |
| 732 |
"dev/yPnkx_GSb0Q/hLgS_431AQAJ>`__. Fortunately, this “mostly transitive” " |
| 733 |
"behavior is exactly the behavior that implementations are likely to " |
| 734 |
"produce naturally. Maybe someday we will find a way to specify this fully" |
| 735 |
" correctly.)" |
| 736 |
msgstr "" |
| 737 |
|
| 738 |
#: ../../tsttcpw.rst:456 |
| 739 |
msgid "Nullness subtyping (and thus subtyping itself) is *not* reflexive." |
| 740 |
msgstr "" |
| 741 |
|
| 742 |
#: ../../tsttcpw.rst:458 |
| 743 |
msgid "" |
| 744 |
"It does end up being reflexive in the `most convenient world <#multiple-" |
| 745 |
"worlds>`__. We don’t state that as a rule for 2 reasons: First, it arises" |
| 746 |
" naturally from the definitions in that world. Second, we don’t want to " |
| 747 |
"suggest that subtyping is reflexive in the `least convenient world " |
| 748 |
"<#multiple-worlds>`__." |
| 749 |
msgstr "" |
| 750 |
|
| 751 |
#: ../../tsttcpw.rst:465 |
| 752 |
msgid "Trusted null-inclusive under every parameterization" |
| 753 |
msgstr "" |
| 754 |
|
| 755 |
#: ../../tsttcpw.rst:467 |
| 756 |
msgid "" |
| 757 |
"A type is trusted null-inclusive under every parameterization if it meets" |
| 758 |
" either of the following conditions:" |
| 759 |
msgstr "" |
| 760 |
|
| 761 |
#: ../../tsttcpw.rst:470 |
| 762 |
msgid "Its `nullness operator <#nullness-operator>`__ is ``UNION_NULL``." |
| 763 |
msgstr "" |
| 764 |
|
| 765 |
#: ../../tsttcpw.rst:471 |
| 766 |
msgid "" |
| 767 |
"It is an `intersection type <#intersection-types>`__ whose elements all " |
| 768 |
"are trusted null-inclusive under every parameterization." |
| 769 |
msgstr "" |
| 770 |
|
| 771 |
#: ../../tsttcpw.rst:474 |
| 772 |
msgid "" |
| 773 |
"**Most convenient world:** The rule is the same except that the " |
| 774 |
"requirement for “``UNION_NULL``” is loosened to “``UNION_NULL`` or " |
| 775 |
"``UNSPECIFIED``.”" |
| 776 |
msgstr "" |
| 777 |
|
| 778 |
#: ../../tsttcpw.rst:479 |
| 779 |
msgid "Trusted null-exclusive under every parameterization" |
| 780 |
msgstr "" |
| 781 |
|
| 782 |
#: ../../tsttcpw.rst:481 |
| 783 |
msgid "" |
| 784 |
"A type is trusted null-exclusive under every parameterization if it has a" |
| 785 |
" `nullness-subtype-establishing path <#nullness-subtype-establishing-" |
| 786 |
"path>`__ to either of the following:" |
| 787 |
msgstr "" |
| 788 |
|
| 789 |
#: ../../tsttcpw.rst:486 |
| 790 |
msgid "" |
| 791 |
"any type whose `nullness operator <#nullness-operator>`__ is " |
| 792 |
"``MINUS_NULL``" |
| 793 |
msgstr "" |
| 794 |
|
| 795 |
#: ../../tsttcpw.rst:489 |
| 796 |
msgid "any augmented class or array type" |
| 797 |
msgstr "" |
| 798 |
|
| 799 |
#: ../../tsttcpw.rst:491 |
| 800 |
msgid "" |
| 801 |
"This rule refers specifically to a “class or array type,” as distinct " |
| 802 |
"from other types like type variables and `intersection types " |
| 803 |
"<#intersection-types>`__." |
| 804 |
msgstr "" |
| 805 |
|
| 806 |
#: ../../tsttcpw.rst:496 |
| 807 |
msgid "Nullness-subtype-establishing path" |
| 808 |
msgstr "" |
| 809 |
|
| 810 |
#: ../../tsttcpw.rst:498 |
| 811 |
msgid "" |
| 812 |
"``A`` has a nullness-subtype-establishing path to ``F`` if both of the " |
| 813 |
"following hold:" |
| 814 |
msgstr "" |
| 815 |
|
| 816 |
#: ../../tsttcpw.rst:501 |
| 817 |
msgid "" |
| 818 |
"``A`` has `nullness operator <#nullness-operator>`__ ``NO_CHANGE`` or " |
| 819 |
"``MINUS_NULL``." |
| 820 |
msgstr "" |
| 821 |
|
| 822 |
#: ../../tsttcpw.rst:504 |
| 823 |
msgid "" |
| 824 |
"There is a path from ``A`` to ``F`` through `nullness-subtype-" |
| 825 |
"establishing direct-supertype edges <#nullness-subtype-establishing-" |
| 826 |
"direct-supertype-edges>`__." |
| 827 |
msgstr "" |
| 828 |
|
| 829 |
#: ../../tsttcpw.rst:508 |
| 830 |
msgid "" |
| 831 |
"The path may be empty. That is, ``A`` has a nullness-subtype-establishing" |
| 832 |
" path to itself — as long as it has one of the required nullness " |
| 833 |
"operators." |
| 834 |
msgstr "" |
| 835 |
|
| 836 |
#: ../../tsttcpw.rst:512 |
| 837 |
msgid "" |
| 838 |
"**Most convenient world:** The rules are the same except that the " |
| 839 |
"requirement for “``NO_CHANGE`` or ``MINUS_NULL``” is loosened to " |
| 840 |
"“``NO_CHANGE``, ``MINUS_NULL``, or ``UNSPECIFIED``.”" |
| 841 |
msgstr "" |
| 842 |
|
| 843 |
#: ../../tsttcpw.rst:517 |
| 844 |
msgid "Nullness-subtype-establishing direct-supertype edges" |
| 845 |
msgstr "" |
| 846 |
|
| 847 |
#: ../../tsttcpw.rst:519 |
| 848 |
msgid "" |
| 849 |
"``T`` has nullness-subtype-establishing direct-supertype edges to the " |
| 850 |
"union of the nodes computed by the following 2 rules:" |
| 851 |
msgstr "" |
| 852 |
|
| 853 |
#: ../../tsttcpw.rst:522 |
| 854 |
msgid "Upper-bound rule:" |
| 855 |
msgstr "" |
| 856 |
|
| 857 |
#: ../../tsttcpw.rst:524 |
| 858 |
msgid "" |
| 859 |
"if ``T`` is an augmented `intersection type <#intersection-types>`__: all" |
| 860 |
" the intersection type’s elements whose `nullness operator <#nullness-" |
| 861 |
"operator>`__ is ``NO_CHANGE`` or ``MINUS_NULL``" |
| 862 |
msgstr "" |
| 863 |
|
| 864 |
#: ../../tsttcpw.rst:527 |
| 865 |
msgid "" |
| 866 |
"if ``T`` is an augmented type variable: all the corresponding type " |
| 867 |
"parameter’s upper bounds whose nullness operator is ``NO_CHANGE`` or " |
| 868 |
"``MINUS_NULL``" |
| 869 |
msgstr "" |
| 870 |
|
| 871 |
#: ../../tsttcpw.rst:530 ../../tsttcpw.rst:544 |
| 872 |
msgid "otherwise: no nodes" |
| 873 |
msgstr "" |
| 874 |
|
| 875 |
#: ../../tsttcpw.rst:532 |
| 876 |
msgid "Lower-bound rule:" |
| 877 |
msgstr "" |
| 878 |
|
| 879 |
#: ../../tsttcpw.rst:534 |
| 880 |
msgid "" |
| 881 |
"for every type parameter ``P`` that has a lower bound whose `base type " |
| 882 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 883 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ is the same as" |
| 884 |
" ``T``\\ ’s base type and whose nullness operator is ``NO_CHANGE``: the " |
| 885 |
"type variable ``P``" |
| 886 |
msgstr "" |
| 887 |
|
| 888 |
#: ../../tsttcpw.rst:539 |
| 889 |
msgid "" |
| 890 |
"TODO(cpovirk): What if the lower bound has some other nullness operator? " |
| 891 |
"I’m pretty sure that we want to allow ``UNSPECIFIED`` in the most " |
| 892 |
"convenient world (as we did before my recent edits), and we may want to " |
| 893 |
"allow more." |
| 894 |
msgstr "" |
| 895 |
|
| 896 |
#: ../../tsttcpw.rst:546 |
| 897 |
msgid "" |
| 898 |
"**Most convenient world:** The rules are the same except that the " |
| 899 |
"requirements for “``NO_CHANGE`` or ``MINUS_NULL``” are loosened to " |
| 900 |
"“``NO_CHANGE``, ``MINUS_NULL``, or ``UNSPECIFIED``.”" |
| 901 |
msgstr "" |
| 902 |
|
| 903 |
#: ../../tsttcpw.rst:551 |
| 904 |
msgid "Containment" |
| 905 |
msgstr "" |
| 906 |
|
| 907 |
#: ../../tsttcpw.rst:553 |
| 908 |
msgid "" |
| 909 |
"The Java rules are defined in `JLS 4.5.1 " |
| 910 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-4.html#jls-4.5.1>`__." |
| 911 |
" We add to them as follows:" |
| 912 |
msgstr "" |
| 913 |
|
| 914 |
#: ../../tsttcpw.rst:557 |
| 915 |
msgid "" |
| 916 |
"Disregard the 2 rules that refer to a bare ``?``. Instead, treat ``?`` " |
| 917 |
"like ``? extends Object``, where the `nullness operator <#nullness-" |
| 918 |
"operator>`__ of the ``Object`` bound is specified by `“Bound of an " |
| 919 |
"unbounded wildcard.” <#unbounded-wildcard>`__" |
| 920 |
msgstr "" |
| 921 |
|
| 922 |
#: ../../tsttcpw.rst:562 ../../tsttcpw.rst:686 |
| 923 |
msgid "" |
| 924 |
"This is just a part of our universal rule to treat a bare ``?`` like ``? " |
| 925 |
"extends Object``." |
| 926 |
msgstr "" |
| 927 |
|
| 928 |
#: ../../tsttcpw.rst:565 |
| 929 |
msgid "" |
| 930 |
"The rule written specifically for ``? extends Object`` applies only if " |
| 931 |
"the nullness operator of the ``Object`` bound is ``UNION_NULL``." |
| 932 |
msgstr "" |
| 933 |
|
| 934 |
#: ../../tsttcpw.rst:568 |
| 935 |
msgid "" |
| 936 |
"When the JLS refers to the same type ``T`` on both sides of a rule, the " |
| 937 |
"rule applies if and only if this spec defines the 2 types to be the `same" |
| 938 |
" type <#same-type>`__." |
| 939 |
msgstr "" |
| 940 |
|
| 941 |
#: ../../tsttcpw.rst:572 |
| 942 |
msgid "" |
| 943 |
"**Most convenient world:** The rules are the same except that the " |
| 944 |
"requirement for “``UNION_NULL``” is loosened to “``UNION_NULL`` or " |
| 945 |
"``UNSPECIFIED``.”" |
| 946 |
msgstr "" |
| 947 |
|
| 948 |
#: ../../tsttcpw.rst:577 |
| 949 |
msgid "Substitution" |
| 950 |
msgstr "" |
| 951 |
|
| 952 |
#: ../../tsttcpw.rst:579 |
| 953 |
msgid "" |
| 954 |
"Substitution on Java base types barely requires an explanation: See `JLS " |
| 955 |
"1.3 " |
| 956 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-1.html#jls-1.3>`__." |
| 957 |
" Substitution on `augmented types <#augmented-type>`__, however, is " |
| 958 |
"trickier: If ``Map.get`` returns “``V`` with `nullness operator " |
| 959 |
"<#nullness-operator>`__ ``UNION_NULL``,” and if a user has a map whose " |
| 960 |
"value type is “``String`` with nullness operator ``UNSPECIFIED``,” then " |
| 961 |
"what does its ``get`` method return? Naive substitution would produce " |
| 962 |
"“``String`` with nullness operator ``UNSPECIFIED`` with nullness operator" |
| 963 |
" ``UNION_NULL``.” To reduce that to a proper augmented type with a single" |
| 964 |
" nullness operator, we define this process." |
| 965 |
msgstr "" |
| 966 |
|
| 967 |
#: ../../tsttcpw.rst:592 |
| 968 |
msgid "" |
| 969 |
"To substitute each type argument ``Aᵢ`` for each corresponding type " |
| 970 |
"parameter ``Pᵢ``:" |
| 971 |
msgstr "" |
| 972 |
|
| 973 |
#: ../../tsttcpw.rst:595 |
| 974 |
msgid "" |
| 975 |
"For every type-variable usage ``V`` whose `base type " |
| 976 |
"<https://docs.google.com/document/d/1KQrBxwaVIPIac_6SCf--" |
| 977 |
"w-vZBeHkTvtaqPSU_icIccc/edit#bookmark=kix.k81vs7t5p45i>`__ is ``Pᵢ``, " |
| 978 |
"replace ``V`` with the result of the following operation:" |
| 979 |
msgstr "" |
| 980 |
|
| 981 |
#: ../../tsttcpw.rst:599 |
| 982 |
msgid "" |
| 983 |
"If ``V`` is `trusted null-exclusive under every parameterization " |
| 984 |
"<#trusted-null-exclusive-under-every-parameterization>`__ in the `least " |
| 985 |
"convenient world <#multiple-worlds>`__, then replace it with the result " |
| 986 |
"of `applying <#applying-operator>`__ ``MINUS_NULL`` to ``Aᵢ``." |
| 987 |
msgstr "" |
| 988 |
|
| 989 |
#: ../../tsttcpw.rst:605 |
| 990 |
msgid "" |
| 991 |
"This is the one instance in which a rule references another rule to be " |
| 992 |
"run under a *different* “world.” Normally, all rules are run `under the " |
| 993 |
"same “world.” <#propagating-multiple-worlds>`__ But in this instance, the" |
| 994 |
" null-exclusivity rule (and all rules that it in turn applies) are always" |
| 995 |
" run in the least convenient world." |
| 996 |
msgstr "" |
| 997 |
|
| 998 |
#: ../../tsttcpw.rst:613 |
| 999 |
msgid "" |
| 1000 |
"This special case improves behavior in “the ``ImmutableList.Builder`` " |
| 1001 |
"case”: Consider an unannotated user of that class. Its builder will have " |
| 1002 |
"an element type whose `nullness operator <#nullness-operator>`__ is " |
| 1003 |
"``UNSPECIFIED``. Without this special case, " |
| 1004 |
"``builder.add(objectUnionNull)`` would pass the subtyping check in the " |
| 1005 |
"`most convenient world <#multiple-worlds>`__. This would happen even " |
| 1006 |
"though we have enough information to know that the parameter to ``add`` " |
| 1007 |
"is universally null-exclusive — even in the most convenient world. The " |
| 1008 |
"special case here makes that subtyping check fail." |
| 1009 |
msgstr "" |
| 1010 |
|
| 1011 |
#: ../../tsttcpw.rst:624 |
| 1012 |
msgid "" |
| 1013 |
"Otherwise, replace ``V`` with the result of applying the nullness " |
| 1014 |
"operator of ``V`` to ``Aᵢ``." |
| 1015 |
msgstr "" |
| 1016 |
|
| 1017 |
#: ../../tsttcpw.rst:630 |
| 1018 |
msgid "Applying a nullness operator to an augmented type" |
| 1019 |
msgstr "" |
| 1020 |
|
| 1021 |
#: ../../tsttcpw.rst:632 |
| 1022 |
msgid "" |
| 1023 |
"The process of applying a `nullness operator <#nullness-operator>`__ " |
| 1024 |
"requires 2 inputs:" |
| 1025 |
msgstr "" |
| 1026 |
|
| 1027 |
#: ../../tsttcpw.rst:635 |
| 1028 |
msgid "the nullness operator to apply" |
| 1029 |
msgstr "" |
| 1030 |
|
| 1031 |
#: ../../tsttcpw.rst:636 |
| 1032 |
msgid "" |
| 1033 |
"the `augmented type <#augmented-type>`__ (which, again, includes a " |
| 1034 |
"`nullness operator <#nullness-operator>`__ for that type) to apply it to" |
| 1035 |
msgstr "" |
| 1036 |
|
| 1037 |
#: ../../tsttcpw.rst:640 |
| 1038 |
msgid "The result of the process is an augmented type." |
| 1039 |
msgstr "" |
| 1040 |
|
| 1041 |
#: ../../tsttcpw.rst:642 |
| 1042 |
msgid "The process is as follows:" |
| 1043 |
msgstr "" |
| 1044 |
|
| 1045 |
#: ../../tsttcpw.rst:644 |
| 1046 |
msgid "" |
| 1047 |
"First, based on the pair of nullness operators (the one to apply and the " |
| 1048 |
"one from the augmented type), compute a “desired nullness operator.” Do " |
| 1049 |
"so by applying the following rules in order. Once one condition is met, " |
| 1050 |
"skip the remaining conditions." |
| 1051 |
msgstr "" |
| 1052 |
|
| 1053 |
#: ../../tsttcpw.rst:649 |
| 1054 |
msgid "" |
| 1055 |
"If the nullness operator to apply is ``MINUS_NULL``, the desired nullness" |
| 1056 |
" operator is ``MINUS_NULL``." |
| 1057 |
msgstr "" |
| 1058 |
|
| 1059 |
#: ../../tsttcpw.rst:651 |
| 1060 |
msgid "" |
| 1061 |
"If either nullness operator is ``UNION_NULL``, the desired nullness " |
| 1062 |
"operator is ``UNION_NULL``." |
| 1063 |
msgstr "" |
| 1064 |
|
| 1065 |
#: ../../tsttcpw.rst:653 |
| 1066 |
msgid "" |
| 1067 |
"If either nullness operator is ``UNSPECIFIED``, the desired nullness " |
| 1068 |
"operator is ``UNSPECIFIED``." |
| 1069 |
msgstr "" |
| 1070 |
|
| 1071 |
#: ../../tsttcpw.rst:655 |
| 1072 |
msgid "The desired nullness operator is ``NO_CHANGE``." |
| 1073 |
msgstr "" |
| 1074 |
|
| 1075 |
#: ../../tsttcpw.rst:657 |
| 1076 |
msgid "" |
| 1077 |
"Then, if the input augmented type is *not* an `intersection type " |
| 1078 |
"<#intersection-types>`__, the output is the same as the input but with " |
| 1079 |
"its nullness operator replaced with the desired nullness operator." |
| 1080 |
msgstr "" |
| 1081 |
|
| 1082 |
#: ../../tsttcpw.rst:661 |
| 1083 |
msgid "" |
| 1084 |
"Otherwise, the output is an intersection type. For every element ``Tᵢ`` " |
| 1085 |
"of the input type, the output type has an element that is the result of " |
| 1086 |
"applying the desired nullness operator to ``Tᵢ``." |
| 1087 |
msgstr "" |
| 1088 |
|
| 1089 |
#: ../../tsttcpw.rst:665 |
| 1090 |
msgid "" |
| 1091 |
"In this case, the desired nullness operator is always equal to the " |
| 1092 |
"nullness operator to apply that was an input to this process. That’s " |
| 1093 |
"because the nullness operator of the intersection type itself is defined " |
| 1094 |
"to always be ``NO_CHANGE``." |
| 1095 |
msgstr "" |
| 1096 |
|
| 1097 |
#: ../../tsttcpw.rst:671 |
| 1098 |
msgid "Capture conversion" |
| 1099 |
msgstr "" |
| 1100 |
|
| 1101 |
#: ../../tsttcpw.rst:673 |
| 1102 |
msgid "" |
| 1103 |
"The Java rules are defined in `JLS 5.1.10 " |
| 1104 |
"<https://docs.oracle.com/javase/specs/jls/se14/html/jls-5.html#jls-5.1.10>`__." |
| 1105 |
" We add to them as follows:" |
| 1106 |
msgstr "" |
| 1107 |
|
| 1108 |
#: ../../tsttcpw.rst:677 |
| 1109 |
msgid "" |
| 1110 |
"The parameterized type that is the output of the conversion has the same " |
| 1111 |
"`nullness operator <#nullness-operator>`__ as the parameterized type that" |
| 1112 |
" is the input type." |
| 1113 |
msgstr "" |
| 1114 |
|
| 1115 |
#: ../../tsttcpw.rst:681 |
| 1116 |
msgid "" |
| 1117 |
"Disregard the JLS rule about ``<?>``. Instead, treat ``?`` like ``? " |
| 1118 |
"extends Object``, where the `nullness operator <#nullness-operator>`__ of" |
| 1119 |
" the ``Object`` bound is specified by `“Bound of an unbounded wildcard.” " |
| 1120 |
"<#unbounded-wildcard>`__" |
| 1121 |
msgstr "" |
| 1122 |
|
| 1123 |
#: ../../tsttcpw.rst:689 |
| 1124 |
msgid "" |
| 1125 |
"When a rule generates a lower bound that is the null type, we specify " |
| 1126 |
"that its nullness operator is ``NO_CHANGE``. (See `“Augmented null " |
| 1127 |
"types.” <#null-types>`__)" |
| 1128 |
msgstr "" |
| 1129 |
|