From ef0e0539aa97a9f2e470f8477b2e9e4690be4cdf Mon Sep 17 00:00:00 2001 From: Carisa-Li Date: Mon, 21 Oct 2024 23:55:59 +0800 Subject: [PATCH] rst: 87-102, 239-374 --- library/math.po | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/library/math.po b/library/math.po index 2e39de6f66..56989493f1 100644 --- a/library/math.po +++ b/library/math.po @@ -8,7 +8,7 @@ msgstr "" "Project-Id-Version: Python 3.13\n" "Report-Msgid-Bugs-To: \n" "POT-Creation-Date: 2024-03-14 00:03+0000\n" -"PO-Revision-Date: 2024-10-06 23:31+0800\n" +"PO-Revision-Date: 2024-10-21 23:50+0800\n" "Last-Translator: Adrian Liaw \n" "Language-Team: Chinese - TAIWAN (https://github.com/python/python-docs-zh-" "tw)\n" @@ -114,7 +114,6 @@ msgstr "" "以整數回傳 *n* 的階乘。若 *n* 非整數型別或其值為負會引發 :exc:`ValueError`。" #: ../../library/math.rst:74 -#, fuzzy msgid "Floats with integral values (like ``5.0``) are no longer accepted." msgstr "允許傳入其值為整數的浮點數(如:``5.0``)已被棄用。" @@ -135,6 +134,8 @@ msgid "" "``float`` format. This operation often provides better accuracy than the " "direct expression ``(x * y) + z``." msgstr "" +"融合乘加運算。回傳 ``(x * y) + z``,用近似於無限精度及範圍的方式計算,而後一" +"次轉換為浮點數格式。此操作通常能提供比運算式 ``(x * y) + z`` 更高的精度。" #: ../../library/math.rst:92 msgid "" @@ -144,6 +145,9 @@ msgid "" "``fma(inf, 0, nan)``. In these cases, ``math.fma`` returns a NaN, and does " "not raise any exception." msgstr "" +"此函式遵循 IEEE-754 標準中規範的融合乘加(fusedMultiplyAdd)運算。該標準保留" +"一種由實作環境決定的案例,即 ``fma(0, inf, nan)`` 和 ``fma(inf, 0, nan)`` 的" +"結果;在此案例中,``math.fma`` 回傳 ``NaN`` 且不會引發例外。" #: ../../library/math.rst:103 msgid "" @@ -442,9 +446,9 @@ msgid "" "remainder operation is zero, that zero will have the same sign as *x*." msgstr "" "特殊情況遵循 IEEE 754:特別是,對任何有限數 *x*,``remainder(x, math.inf)`` " -"值為 *x*;對任何非 NaN 值 *x*,``remainder(x, 0)`` 及 ``remainder(math.inf, " -"x)`` 會引發 :exc:`ValueError`。若取餘數操作的結果為零,則該零值會與 *x* 同" -"號。" +"值為 *x*;對任何非 ``NaN`` 值 *x*,``remainder(x, 0)`` 及 ``remainder(math." +"inf, x)`` 會引發 :exc:`ValueError`。若取餘數操作的結果為零,則該零值會與 *x* " +"同號。" #: ../../library/math.rst:307 msgid ""