Closed marcows closed 1 year ago
With MSYS2 the PDF is in letter format whereas with Linux it is in A4. This also applies to the PDFs found in the official releases srecord-1.65.0-win64.zip and srecord-1.65.0-Linux.tar.gz. The PDF from direct website link is the same as from Linux release.
And indeed the paper size has something to do with the wrong destination position.
I changed the paper size via two groff command line options described here: https://www.gnu.org/software/groff/manual/html_node/Paper-Size.html
Here are my test results, both with MSYS2 and Linux: groff options | outline link destination |
---|---|
-dpaper=a4 -P-pa4 | correct, at NAME |
-dpaper=letter -P-pletter | correct, at NAME |
-dpaper=a4 -P-pletter | too high up, near start of page |
-dpaper=letter -P-pa4 | too far down, near DESCRIPTION |
Should I set A4 explicitly? Then the problem with link destination would be fixed and the PDFs in Win64 and Linux release files would also be in the same paper size.
Rebased on master to modify the new doc/man5/srec_os65a.5
Three tasks done with these new five commits. Rebase and cleanup will happen later.
Have to investigate about the wrong placement of srec_wilson(5) PDF outline entry.
Should I set A4 explicitly? Then the problem with link destination would be fixed and the PDFs in Win64 and Linux release files would also be in the same paper size.
If this is still an open question, then yes set to A4. This was an oversight on my part when getting the MSYS2 build up and running.
If this is still an open question, then yes set to A4.
The latest commit "Explicitly set a consistent paper size" addressed this and set it to A4. But thanks for the reply and agreement.
This was an oversight on my part when getting the MSYS2 build up and running.
The A4/letter inconsistency was already present in the 1.64 release from 2014, I have checked right now.
This is looking great Marcus! I'll wait until you hit "ready for review" before I start looking properly (but please don't feel rushed).
Given the number commits, I'm happy to squash but if you'd prefer to rebase into a few select/logical commits, that works too.
I'll definitely rebase into a few commits, I'm a clean-history-fanatic.
It's Markus with k, the c just comes from the "cow" part of my username which I'm not a big fan of anymore :-/
Sounds good and thank you Markus for correcting me on the spelling. I did take my spelling cue from 'cow" so I'm guessing I'm not the first to make that mistake at least.
Finally ready for review, two notes from my side:
* Should "pdfmark" be moved to dictionaries/cmake.txt? I didn't do this because it looked as if only official cmake keywords were included there, but now I don't think that's true anymore. However, I think names.txt is a good place.
I created those dictionaries for "non-fixable" words: cmake.txt
for CMake stuff (they're all from CMake itself), troff.txt
for those I and P prefixed words, and names.txt
for any kind of names of people, tools, formats, ...
"pdfmark" would fit best into names.txt
.
All other words, which I had not analyzed in detail, are excluded in the top-level cspell.config.yaml
. Some words will vanish, when the old release description is reworked.
I'm open for improvements/simplifcations, e.g. perhaps I should rename cmake.txt
to code.txt
and also add the C, troff, ... keywords.
PDF metadata and PDF outline (navigation in the side bar) have been added.
Tasks:
(Partially) implements the second task of #1.