From: Xidorn Quan <quanxunzhen@gmail.com>

Date: Tue, 25 Feb 2014 09:28:58 +1100

Message-ID: <CAMdq69_sq2R-OSiM3ihNbO0FORZecYLjEtWEqyFJM2+jtqS+Hg@mail.gmail.com>

To: "Tab Atkins Jr." <jackalmage@gmail.com>

Cc: www-style list <www-style@w3.org>

Date: Tue, 25 Feb 2014 09:28:58 +1100

Message-ID: <CAMdq69_sq2R-OSiM3ihNbO0FORZecYLjEtWEqyFJM2+jtqS+Hg@mail.gmail.com>

To: "Tab Atkins Jr." <jackalmage@gmail.com>

Cc: www-style list <www-style@w3.org>

On Tue, Feb 25, 2014 at 8:44 AM, Tab Atkins Jr. <jackalmage@gmail.com> wrote: > On Sun, Feb 23, 2014 at 12:52 AM, Xidorn Quan <quanxunzhen@gmail.com> wrote: >> Hi, >> >> In the current draft, the algorithm of system 'symbolic' seems to be >> incorrect. We use the style "footnote" defined in the spec for >> example: >> >> Let counter value = 1, N = 4, then >> >> 1. chosen symbol = symbol(1 mod 4) = symbol(1) = ⁑ (SHOULD be '*') >> 2. representation length = floor( (1 - 1) / 4 ) = 0 (SHOULD be 1) >> 3. S = representation length * chosen symbol = (empty sequence) >> >> The correct algorithm should be: >> >> Let N be the length of the list of counter symbols, value be the >> counter value, S initially be the empty string, and symbol(n) be the >> nth counter symbol in the list of counter symbols (0-indexed). >> >> 1. Let the chosen symbol be symbol( (value - 1) mod N). >> 2. Let the representation length be ceil(value / N). >> 3. Append the chosen symbol to S a number of times equal to the >> representation length. >> >> Finally, return S. > > Oh jeez, how did I get this so wrong? That's embarrassing. ;_; > > Fixed, thanks. I found that you just fixed one error, however there were two. See the second step, which should be `ceil(value / N)` instead of `floor((value - 1) / N)`. - XidornReceived on Monday, 24 February 2014 22:30:05 UTC

*
This archive was generated by hypermail 2.3.1
: Monday, 2 May 2016 14:39:19 UTC
*