@@ -19,7 +19,7 @@ msgid ""
19
19
msgstr ""
20
20
"Project-Id-Version :Python 3.8\n "
21
21
"Report-Msgid-Bugs-To :\n "
22
- "POT-Creation-Date :2020-02-09 12:40 +0000\n "
22
+ "POT-Creation-Date :2020-02-22 12:45 +0000\n "
23
23
"PO-Revision-Date :2017-02-16 23:06+0000\n "
24
24
"Last-Translator :tomo, 2019\n "
25
25
"Language-Team :Japanese (https://www.transifex.com/python-doc/teams/5390/ja/)\n "
@@ -2334,15 +2334,51 @@ msgid ""
2334
2334
"A. Yes. In the CPython and PyPy3 implementations, the C/CFFI versions of "
2335
2335
"the decimal module integrate the high speed `libmpdec "
2336
2336
"<https://www.bytereef.org/mpdecimal/doc/libmpdec/index.html>`_ library for "
2337
- "arbitrary precision correctly-rounded decimal floating point arithmetic. "
2338
- "``libmpdec`` uses `Karatsuba multiplication "
2337
+ "arbitrary precision correctly-rounded decimal floating point arithmetic "
2338
+ "[#]_. ``libmpdec`` uses `Karatsuba multiplication "
2339
2339
"<https://en.wikipedia.org/wiki/Karatsuba_algorithm>`_ for medium-sized "
2340
2340
"numbers and the `Number Theoretic Transform "
2341
2341
"<https://en.wikipedia.org/wiki/Discrete_Fourier_transform_(general)#Number-"
2342
- "theoretic_transform>`_ for very large numbers. However, to realize this "
2343
- "performance gain, the context needs to be set for unrounded calculations."
2342
+ "theoretic_transform>`_ for very large numbers."
2343
+ msgstr ""
2344
+
2345
+ #: ../../library/decimal.rst:2131
2346
+ msgid ""
2347
+ "The context must be adapted for exact arbitrary precision arithmetic. "
2348
+ ":attr:`Emin` and :attr:`Emax` should always be set to the maximum values, "
2349
+ ":attr:`clamp` should always be 0 (the default). Setting :attr:`prec` "
2350
+ "requires some care."
2351
+ msgstr ""
2352
+
2353
+ #: ../../library/decimal.rst:2135
2354
+ msgid ""
2355
+ "The easiest approach for trying out bignum arithmetic is to use the maximum "
2356
+ "value for :attr:`prec` as well [#]_::"
2357
+ msgstr ""
2358
+
2359
+ #: ../../library/decimal.rst:2144
2360
+ msgid ""
2361
+ "For inexact results, :attr:`MAX_PREC` is far too large on 64-bit platforms "
2362
+ "and the available memory will be insufficient::"
2363
+ msgstr ""
2364
+
2365
+ #: ../../library/decimal.rst:2152
2366
+ msgid ""
2367
+ "On systems with overallocation (e.g. Linux), a more sophisticated approach "
2368
+ "is to adjust :attr:`prec` to the amount of available RAM. Suppose that you "
2369
+ "have 8GB of RAM and expect 10 simultaneous operands using a maximum of 500MB"
2370
+ " each::"
2371
+ msgstr ""
2372
+
2373
+ #: ../../library/decimal.rst:2176
2374
+ msgid ""
2375
+ "In general (and especially on systems without overallocation), it is "
2376
+ "recommended to estimate even tighter bounds and set the :attr:`Inexact` trap"
2377
+ " if all calculations are expected to be exact."
2378
+ msgstr ""
2379
+
2380
+ #: ../../library/decimal.rst:2185
2381
+ msgid ""
2382
+ "This approach now works for all exact results except for non-integer powers."
2383
+ " Also backported to 3.7 and 3.8."
2344
2384
msgstr ""
2345
- "A. はい。\n"
2346
- "CPython 実装と PyPy3 実装では、 C/CFFI 版の decimal モジュールは、任意精度の正しい丸めを行う 10 進浮動小数点演算のための高速な `libmpdec <https://www.bytereef.org/mpdecimal/doc/libmpdec/index.html>`_ ライブラリを統合しています。\n"
2347
- "``libmpdec`` は `Karatsuba multiplication <https://en.wikipedia.org/wiki/Karatsuba_algorithm>`_ を中程度のサイズの数に対して使い、 `Number Theoretic Transform <https://en.wikipedia.org/wiki/Discrete_Fourier_transform_(general)#Number-theoretic_transform>`_ を非常に大きな数に対して使います。\n"
2348
- "ただし、このパフォーマンス向上を得るためには、計算で丸めが発生しないように context を設定する必要があります。"