Permalink
Comparing changes
Choose two branches to see what’s changed or to start a new pull request.
If you need to, you can also .
Open a pull request
Create a new pull request by comparing changes across two branches. If you need to, you can also .
3
contributors
Commits on Mar 05, 2019
I backed out that code change ... need to address properly!
Commits on Mar 08, 2019
Commits on Mar 12, 2019
Thanks to Xiangdong Zeng
Commits on Mar 21, 2019
Commits on Mar 23, 2019
As discussed on the team list, we need something dedicated for the terminal
We may want to thing about the name: here, you always get something so a (TF) isn't needed, and that might be confusing.
Commits on Mar 26, 2019
Commits on Apr 01, 2019
Commits on Apr 06, 2019
The old version missed out the local ...
Commits on Apr 09, 2019
One file per 'concept', more or less. Should allow some to be marked as stable.
We are very 'close to the metal' here, so the most sensible thing to do is to link to the next level up.
Probably these simply can't work here ... may have to deal with the object return value at some stage.
Currently only for the easy case ...
Commits on Apr 10, 2019
Although we can think about dvisvgm here, basically these are PDF features. Keeps the source file consistent: everything is 'pdf...'.
We will need this for dvips, and it seems to fit OK with almost all existing code. See breakurl and hypdvips for ideas on allowing breaks.
Commits on Apr 11, 2019
The short names look great .. until we want all of the stuff to break dvips hyperlinks. Also, they likely should be more organised.
Also stick to "driver.<thing>" here for naming.
Make debugging a little easier
At present, all the major version stuff is a no-op ...
Thought you get some odd effects in the TL'18 one if you set v2.0 ...
Commits on Apr 12, 2019
This is roughly the same as pdfTeX: there will be edge cases. Using a box for measuring up feels a bit odd, but this is really dvips-specific so a setting seems wrong too. We may need to make the size-setting into a separate function, thus allowing adjustment.
Just in case this is done in a group ...
We are going to want more 'stuff' in the link code, and this should be avoided for annotations.
This will impact on .ps file size ... and makes the trace output a bit clearer.
This will be needed for link breaking.
This puts them around the content, which will be needed for link breaking.
Just the set up at present. We may want to look again at all of this ...
Commits on Apr 13, 2019
These will get more tricky as we add the breaking code.
Commits on Apr 16, 2019
The code is from hypdvips almost unchanged, other than "evenboxes" ideas.
Commits on Apr 17, 2019
By using a box, we can keep the ability of hypdvips to allow breaking of \pdfmark, but avoid the issue with finding the end of the link.
The box needs to be 'transparent'.
Again, this is about boxes.
Commits on Apr 18, 2019
Commits on Apr 19, 2019
One could argue that this belongs higher up, but it is likely we'll need to fiddle with OCGs later: that really does need to be done in the driver. Landscape pages and a few other effects still to look at.
Commits on Apr 21, 2019
Commits on Apr 22, 2019
Perhaps this should just point to the website.
Commits on Apr 23, 2019
Commits on Apr 24, 2019
This needs the path to be 'discarded'.
These are arguably copyright-free: don't really need people to acknowledge test files!
Commits on Apr 25, 2019
Commits on Apr 26, 2019
Commits on Apr 27, 2019
Commits on Apr 28, 2019
Works for pdfTeX/LuaTeX/dvips but not yet (x)dvipdfmx: need an interface for \pdfsavepos.
\pdfcatalog looks rather more tricky!
Commits on Apr 30, 2019
Thanks to AlexG
Commits on May 01, 2019
Commits on May 02, 2019
Thanks to Alexander Grahn: reveals a possible bug in dvipdfmx with rotation
Showing you all comments on commits in this comparison.
This comment has been minimized.
This comment has been minimized.
|
Should it be |
This comment has been minimized.
This comment has been minimized.
|
Should it be |
This comment has been minimized.
This comment has been minimized.
|
Ooops yes |
This comment has been minimized.
This comment has been minimized.
|
Very nice, this will be very useful :-) Just one thing: shouldn't these be |
This comment has been minimized.
This comment has been minimized.
|
Oopps: thought they came from the kernel ... |
This comment has been minimized.
This comment has been minimized.
|
Not fond of \use:c { conditionally@traceoff }, I would prefer
\cs_if_exist_use:N \conditionally@traceoff , with appropriate \exp_not:n
because we are in an x-expanding definition.
|
This comment has been minimized.
This comment has been minimized.
|
Neither fond of that either, really. I think it would be better to move an empty definition to the kernel or just provide an empty definition as part of xparse and just use \conditionall@traceoff unconditionally. Then trace package can change the definition and be done with. Otherwise you add another test to each use of an xparse-defined command and also in tracing situations you see that test unnecessarily. |
This comment has been minimized.
This comment has been minimized.
|
@FrankMittelbach That did occur to me: to be honest, I was surprised when I found that they are not already in the kernel ... |
There are no files selected for viewing
This file was deleted.
This file was deleted.