!NuxCISwYQJuyWwNsEI:matrix.org

PGF/TikZ

54 Members
https://github.com/pgf-tikz11 Servers

Load older messages


SenderMessageTime
10 Nov 2021
@hmenke:matrix.orghmenke

I was thinking about using \pgf@marshal but that is actually a lot more tokens.

--- a/tex/generic/pgf/basiclayer/pgfcoretransformations.code.tex
+++ b/tex/generic/pgf/basiclayer/pgfcoretransformations.code.tex
@@ -466,6 +466,7 @@
 % }
 %
 \def\pgftransformlineattime#1#2#3{%
+  \pgfgettransformentries\pgf@tempaa\pgf@tempab\pgf@tempba\pgf@tempbb\pgf@notimportantx\pgf@notimportanty
   \pgf@process{#2}%
   \pgf@xb=\pgf@x% xb/yb = start point
   \pgf@yb=\pgf@y%
@@ -479,6 +480,20 @@
   \ifpgfslopedattime%
     \advance\pgf@xc by-\pgf@xb%
     \advance\pgf@yc by-\pgf@yb%
+    % Now that we get the tangent vector without transformation
+    % It suffices to apply the non-shift part of the transformation
+    \ifpgfresetnontranslationattime
+      \begingroup
+        \pgfsettransformentries\pgf@tempaa\pgf@tempab\pgf@tempba\pgf@tempbb{0pt}{0pt}%
+        \pgf@pos@transform{\pgf@xc}{\pgf@yc}%
+        \edef\pgf@marshal{%
+          \pgf@xc=\the\pgf@xc
+          \pgf@yc=\the\pgf@yc
+        }%
+      \expandafter\endgroup\pgf@marshal
+    \fi
+    % OK, now xc and yc are correctly tangent to the line
+    % Continue the usual routine
     \ifpgfallowupsidedownattime%
     \else%
       \ifdim\pgf@xc<0pt%
08:26:48
@muzzi:matrix.orgmuzimuzhi

+1 to the use of \pgf@marshal. I was also thinking about providing a pair of \pgf-get-transform-@non-shifting and \pgf-set-transform-@non-shifting, so we could use

\pgfgettransform@nonshifting\pgf@temp
[...]
\expandafter\pgfsettransformentries\pgf@temp{0pt}{0pt}
08:50:56
@muzzi:matrix.orgmuzimuzhi

with definition

\pgfgettransform@nonshifting#1{%
    \edef#1{{\pgf@pt@aa}{\pgf@pt@ab}{\pgf@pt@ba}{\pgf@pt@bb}}%
}
08:53:04
@muzzi:matrix.orgmuzimuzhi or named as \pgf-transform-get-non-translations, following the existing \pgf-transform-reset-non-translations 08:59:26
@muzzi:matrix.orgmuzimuzhi * or named as \pgf-transform-get-non-translations, following the existing \pgf-transform-reset-non-translations
(what a long name)
08:59:44
@muzzi:matrix.orgmuzimuzhi *

+1 to the use of \pgf@marshal. I was also thinking about providing a pair of \pgf-(get|set)-transform-@non-shifting, so we could use

\pgfgettransform@nonshifting\pgf@temp
[...]
\expandafter\pgfsettransformentries\pgf@temp{0pt}{0pt}
13:23:59
12 Nov 2021
@muzzi:matrix.orgmuzimuzhioops, i added changes prepared for a PR to a directly pushed commit https://github.com/pgf-tikz/pgf/commit/d835e0312028e7e398626bdd77e5effeaab280f004:51:09
@muzzi:matrix.orgmuzimuzhihow do we proceed?04:51:32
@muzzi:matrix.orgmuzimuzhi Specifically, that innocent change-comment-only commit wrongly contains fix for https://github.com/pgf-tikz/pgf/issues/1075 (caused by git -a hmmm). hmenke 05:01:17
14 Nov 2021
@mgttlinger:matrix.orgMerlin Göttlinger It seems that putting a tikzpicture inside the content of a callout node breaks the callout pointer (i.e. the set callout absolute position is ignored). 12:39:41
@mgttlinger:matrix.orgMerlin Göttlinger

Example:

\begin{tikzpicture}[scale=1.35, transform shape]
      \node [shape=person, monitor, minimum size=5em] (h) {};
      \node [ellipse callout, callout absolute pointer = {(0,0)}, draw, above right = 1em and 0em of h, align=center] (co) {\raisebox{.2\height}{
        \begin{tikzpicture}
          \node [draw, alice] (a) {Alice};
          \node [draw, bob, right = of a] (b) {Bob};
          \node [draw, charlie, right = of b] (c) {Charlie};
          \node [draw, overlay, cloud callout, callout absolute pointer = {([yshift=.5em]a.north east)}, above = 1em of b] (x1) {x};
          \node [draw, overlay,cloud callout, callout absolute pointer = {([yshift=.5em]b.north)}, callout pointer segments = 1, above = 1em of b] (x2) {x};
          \node [draw, overlay, cloud callout, callout absolute pointer = {([yshift=.5em]c.north west)}, above = 1em of b] (x3) {x};
          \node [above = -2em of x1] (x1) {};
        \end{tikzpicture}}
      };
      \node [rectangle callout, callout absolute pointer = {([xshift=2.5em,yshift=-.5em]h.east)}, draw, above right = -2em and 4em of h] (f) {\(K_{Alice}x \land K_{Bob} x \land K_{Charlie} x\)};
      \node [rectangle callout, callout absolute pointer = {([xshift=2.5em,yshift=-.6em]h.east)}, draw, below = 1em of f, align=center] (p) {\(E_{\{Alice, Bob, Charlie\}} x\)};
      \node [rectangle callout, callout absolute pointer = {([xshift=2.5em,yshift=-.7em]h.east)}, draw, below = 1em of p, align=center] (p2) {\(C_{\{Alice, Bob, Charlie\}} x\)?};
    \end{tikzpicture}
12:48:07
@hmenke:matrix.orghmenke Merlin Göttlinger: Nested tikzpictures just don't work. Use a lrbox or so outside of the picture. 13:39:05
@hmenke:matrix.orghmenkeYour code is also not a MWE, so I can't post a corrected version.13:40:39
@hmenke:matrix.orghmenke muzimuzhi: Do I understand this correctly, that you already committed a fix for #1075? In this case, just close it with label closed-fixed. Maybe add a comment to the issue containing the hash of the commit. Don't worry about the commit message, accidents happen and if you look at the old commit imported from CVS you'll find lots of *** empty log message *** with thousands of lines of changes. 13:42:56
@mgttlinger:matrix.orgMerlin Göttlingerthat works, thanks13:43:00
@muzzi:matrix.orgmuzimuzhi
In reply to @hmenke:matrix.org
muzimuzhi: Do I understand this correctly, that you already committed a fix for #1075? In this case, just close it with label closed-fixed. Maybe add a comment to the issue containing the hash of the commit. Don't worry about the commit message, accidents happen and if you look at the old commit imported from CVS you'll find lots of *** empty log message *** with thousands of lines of changes.
done
15:23:49
@marcel:math.hamburgMarcel Krüger left the room.19:36:04
15 Nov 2021
@marcel:2krueger.deMarcel Krüger joined the room.13:36:00
16 Nov 2021
@tbaudoin:matrix.orgThierry Baudoin joined the room.15:16:03
23 Nov 2021
@fabian.drinck:matrix.orgfabric-and-ink joined the room.10:42:42
1 Dec 2021
@freetux:tetaneutral.netfreetux changed their profile picture.07:39:23
3 Dec 2021
@hmenke:matrix.orghmenke invited @_neb_github:matrix.orgGithub.14:33:37
@_neb_github:matrix.orgGithub joined the room.14:33:39
@hmenke:matrix.orghmenke invited @_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org].14:33:54
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] joined the room.14:33:55
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] [pgf-tikz/pgf] gabor-braun synchronize pull request #1081: Support & in matrices without catcode change [open] - https://github.com/pgf-tikz/pgf/pull/1081 15:28:57
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] [pgf-tikz/pgf] muzimuzhi opened issue #1082: `\pgfkeyscurrentname` shown in `unknown key` error messages [open] - https://github.com/pgf-tikz/pgf/issues/1082 22:00:47
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] [pgf-tikz/pgf] muzimuzhi review_requested pull request #1083: fix(decorations): expand `pgfkeys` internals in error msg [open] - https://github.com/pgf-tikz/pgf/pull/1083 22:20:23
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] [pgf-tikz/pgf] muzimuzhi opened pull request #1083: fix(decorations): expand `pgfkeys` internals in error msg [open] - https://github.com/pgf-tikz/pgf/pull/1083 22:20:23
@_neb_github_=40hmenke=3amatrix.org:matrix.orgGithub [@hmenke:matrix.org] [pgf-tikz/pgf] Symbol1 opened issue #1084: /- associativity issue in pgfmath [open] - https://github.com/pgf-tikz/pgf/issues/1084 22:32:36

There are no newer messages yet.


Back to Room ListRoom Version: 6