Debian Codegraph @

texmf-dist/doc/pdftex/tests/15-startlink-boxing/test-different-levels.tex

raw
1 % $Id: test-different-levels.tex 831 2020-06-04 18:07:54Z karl $
2 % Public domain.
3
4 \catcode`\{=1 \catcode`\}=2 \pdfoutput=1
5
6 x\hbox{\pdfstartlink
7 attr {/Border[0 0 1]}
8 user{
9 /Subtype/Link%
10 /A<<%
11 /Type/Action%
12 /S/URI%
13 /URI(foo)%
14 >>%
15 } Linktext}
16 \pdfendlink
17 \end
18
19 % https://tug.org/pipermail/tex-live/2018-November/042759.html
20 % Date: Wed, 28 Nov 2018 12:31:25 +0100
21 % From: Ulrike Fischer <[email protected]>
22 % To: [email protected]
23 % Subject: [tex-live] pdfstartlink/pdfendlink and boxing level problems
24 %
25 % 1. Currently the commands give a fatal error if used in different
26 % boxing levels:
27 %
28 % [example program above]
29 %
30 % ! pdfTeX error (ext4): \pdfendlink ended up in different nesting
31 % level than \pdfstartlink.
32 % ! ==> Fatal error occurred, no output PDF file produced!
33 %
34 % This doesn't happen very often in documents, but if the galley is
35 % split and reboxed it is possible that long links end in different
36 % levels. One example involving two column mode and top floats has
37 % been discussed on the latex2e github:
38 % https://github.com/latex3/latex2e/issues/94
39 %
40 % Is the fatal error really sensible here? Couldn't the engines close
41 % the last annotation with some more or less sensible coordinate and
42 % continue with a warning? A pdf with a wrong link area seems better
43 % than no pdf at all (and would release the format and packages from
44 % the need to carefully keep track of boxing levels when splitting and
45 % reassembling stuff).
46 %
47 % [see ../16-nolink-special/*.tex for other request.]