diff -r 6969d0ffc576 -r 7ff71bdcf731 src/HOL/Analysis/Complex_Transcendental.thy --- a/src/HOL/Analysis/Complex_Transcendental.thy Tue Dec 20 22:24:36 2022 +0000 +++ b/src/HOL/Analysis/Complex_Transcendental.thy Wed Dec 21 12:30:48 2022 +0000 @@ -2049,6 +2049,48 @@ by simp qed +text\Prop 17.6 of Bak and Newman, Complex Analysis, p. 243. + Only this version is for the reals. Can the two proofs be consolidated?\ +lemma uniformly_convergent_on_prod_real: + fixes u :: "nat \ real \ real" + assumes contu: "\k. continuous_on K (u k)" + and uconv: "uniformly_convergent_on K (\n x. \ku k x\)" + and K: "compact K" + shows "uniformly_convergent_on K (\n x. \k \k. complex_of_real \ u k \ Re" + define L where "L \ complex_of_real ` K" + have "compact L" + by (simp add: \compact K\ L_def compact_continuous_image) + have "Re ` complex_of_real ` X = X" for X + by (auto simp: image_iff) + with contu have contf: "\k. continuous_on L (f k)" + unfolding f_def L_def by (intro continuous_intros) auto + obtain S where S: "\\. \>0 \ \\<^sub>F n in sequentially. \x\K. dist (\ku k x\) (S x) < \" + using uconv unfolding uniformly_convergent_on_def uniform_limit_iff by presburger + have "\\<^sub>F n in sequentially. \z\L. dist (\k S \ Re) z) < \" + if "\>0" for \ + using S [OF that] by eventually_elim (simp add: L_def f_def) + then have uconvf: "uniformly_convergent_on L (\n z. \k\. \>0 \ \\<^sub>F n in sequentially. \z\L. dist (\k" + using uniformly_convergent_on_prod [OF contf \compact L\ uconvf] + unfolding uniformly_convergent_on_def uniform_limit_iff by blast + have \
: "\(\k \ cmod ((\kk\N. of_real (1 + u k x)) = (\k\N. 1 + of_real (u k x))" for N + by force + then show ?thesis + by (metis Re_complex_of_real abs_Re_le_cmod minus_complex.sel(1) of_real_prod) + qed + have "\\<^sub>F n in sequentially. \x\K. dist (\k P \ of_real) x) < \" + if "\>0" for \ + using P [OF that] by eventually_elim (simp add: L_def f_def dist_norm le_less_trans [OF \
]) + then show ?thesis + unfolding uniformly_convergent_on_def uniform_limit_iff by blast +qed + + subsection\The Argument of a Complex Number\ text\Unlike in HOL Light, it's defined for the same interval as the complex logarithm: \(-\,\]\.\