summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* [features] update addfeatures() from upstream font-otc.luaPhilipp Gesang2015-12-181-126/+253
|
* [mkstatus] include git tag in status tablev2.6-fix-2Philipp Gesang2015-12-121-7/+23
| | | | Please ignore, the tag is included to accommodate the CTAN workflow.
* [mkstatus,doc] eliminate references to luaotfload-package.luav2.6-fix-1Philipp Gesang2015-12-123-27/+13
|
* [mkimport,build] integrate with main makefilePhilipp Gesang2015-12-122-8/+3
|
* [package] remove luaotfload-packagePhilipp Gesang2015-12-121-99/+0
| | | | Not required since we now package the loader ourselves. Good riddance.
* [mkimport] include MD5 hash with file scope infoPhilipp Gesang2015-12-121-6/+7
|
* [mkimport] directly package fontloaderPhilipp Gesang2015-12-121-54/+193
| | | | | | | | | | Fix issue #305 No round-trip through mtx-package is required anymore. This assumes “util-mrg.lua” which comes with Context is present in the path. As as side-effect we can inject custom annotations into the file that reduce possible confusion for users.
* [build] include package, import, and test scripts in treev2.6Philipp Gesang2015-12-091-0/+3
|
* [*] bump version and datePhilipp Gesang2015-12-0911-27/+27
| | | | | We still had some 2014 values lingering around dark corners. In theory this is all meaningless wrt. the Git repo.
* [init] suppress one less useful messagePhilipp Gesang2015-12-091-1/+1
|
* [build] keep fontloader source in tex subtree to keep kpse happyPhilipp Gesang2015-12-031-1/+1
| | | | | | kpathsea doesn’t return paths to files in the source subtree so loading the unpackaged fontloader files is broken. Fix this by moving the relevant files in the tex tree.
* [tool] fix path displayed in help messagePhilipp Gesang2015-12-021-1/+1
| | | | | | This extends commit 78a1273715.. to the tool. Also fixes the wrong path being issued to the formatter. Luckily, this is not a functional defect since the path is used for display purposes only.
* [doc] load luatexbase for the catcodetable stuffPhilipp Gesang2015-12-022-1/+2
|
* [build] include fontloader source in packagesPhilipp Gesang2015-12-021-4/+9
|
* [init] typoDohyun Kim2015-11-301-1/+1
|
* [*] tidy up headers, attributionsv2.6-rc3Philipp Gesang2015-11-297-7/+9
|
* [main] refine very early loggingPhilipp Gesang2015-11-291-1/+7
| | | | | This kludge dumps the input arguments on one line each -- fix this by defining a dumb imitation of our real logger that is installed later on.
* [conf] ensure correct fontloader is loaded without config filePhilipp Gesang2015-11-291-2/+6
|
* [loaders] customize define_font resetPhilipp Gesang2015-11-291-8/+17
| | | | | | This function needn’t be as general for this kind of callback. The new implementation is simpler but also complains about irregularities like multiple registered functions.
* [aux] fix typo in function namePhilipp Gesang2015-11-291-1/+1
| | | | | Doesn’t seem like anyone is using these since this would’ve exploded immediately …
* [main] remove locals for provides_module() return valuesPhilipp Gesang2015-11-281-2/+1
| | | | As of ltluatex, provides_module() has no return value any longer.
* [tool,db] make cache directives work againPhilipp Gesang2015-11-282-8/+11
| | | | | These were rendered unusable due to the suspended initialization. Creating the directives mapping on the fly is just as good.
* [diagnose] adapt invocation of getwritablepath()Philipp Gesang2015-11-281-5/+8
| | | | | | | | | Some upstream API change (possibly in file.join()) caused that function to emit relative paths. file.join("/foo", "bar") --> good file.join("/foo", "" ) --> good file.join("/foo" ) --> wtf‽
* [doc] clarify fontloader options in luaotfload.conf(5)Philipp Gesang2015-11-281-21/+23
|
* [colors] support pgf transparency in luatex v0.85Dohyun Kim2015-11-281-16/+16
| | | | | | With tex.scantoks provided by luatex v0.85, now we can support color transparency in harmony with beamer/pgf/tikz. No hacking needed at tex macro level, all done in lua code.
* [build] kill off install targetPhilipp Gesang2015-11-281-4/+8
| | | | | | | | | A while back we had one confused user who apparently had a borked install due to this. Since it hasn’t been tested, we replace the original instructions with a help message. The target may be reintroduced at some point (build TDS, then extract to local tree), but for now it’s not worth the hassle.
* [doc] explain usual packaging commandsPhilipp Gesang2015-11-281-5/+17
|
* [build] include import and package targets in build infoPhilipp Gesang2015-11-281-0/+3
|
* [doc,init] clarify default/reference distinctionPhilipp Gesang2015-11-283-10/+30
| | | | | * default: our package (without lualibs) * reference: the package from upstream
* [mkstatus] update file listPhilipp Gesang2015-11-271-28/+88
|
* @ may not be catcode 11 in plainv2.6-rc2David Carlisle2015-11-261-1/+1
|
* [*] update news and readmePhilipp Gesang2015-11-262-0/+2
|
* [letterspace] fix handling of interword spacePhilipp Gesang2015-11-261-23/+71
| | | | | | | | | Fix https://github.com/lualatex/luaotfload/issues/297 Interword glue hasn’t been considered yet. This again adapts the relevant logic from Context to our letterspacing method. The code is deliberately simplistic and will most likely not address all constellations of a glue preceding a character.
* [colors] remove workaround for deprecated interfacePhilipp Gesang2015-11-261-13/+10
| | | | | | | The Luatexbase API was consolidated. Since the priority_in_callback functionality wasn’t used for anything but determining whether we need to convert to an hlist first, there’s not really any gain in keeping it around in the generic form.
* define priority_in_callback (from luatexbase)David Carlisle2015-11-261-1/+10
|
* [doc] document mkimport usagePhilipp Gesang2015-11-241-21/+54
|
* [mkimport] allow output in pwdPhilipp Gesang2015-11-241-1/+1
|
* [mkimport] remove code obstructing user-specified import pathPhilipp Gesang2015-11-241-9/+0
|
* [doc] fix several listingsPhilipp Gesang2015-11-241-5/+7
|
* [doc] document setting the Context pathPhilipp Gesang2015-11-241-3/+15
|
* [db] fix search path construction on Win hostsPhilipp Gesang2015-11-221-1/+1
| | | | | | | | | | This seems to help with the examples from #165 that rely on setting ``$OPENTYPEFONTS`` via the environment. Referencing a non-existing value caused a branch to always pick Unix path separators. The affair was pretty insidious since the paths we received from kpse were sane so the behavior was only triggered with manual overrides.
* [db] emit message when skipping PWDPhilipp Gesang2015-11-221-1/+5
|
* Merge branch 'canonical-master'Philipp Gesang2015-11-223-11/+20
|\
| * Merge pull request #287 from davidcarlisle/masterPhilipp Gesang2015-11-223-11/+20
| |\ | | | | | | Patch to just load ltluatex, not luatexbase
| | * local reset_callback definition for ltluatexDavid Carlisle2015-11-211-1/+11
| | |
| | * for 2.6David Carlisle2015-11-2150-3729/+6532
| | |\ | | |/ | |/|
| * | Merge pull request #293 from phi-gamma/masterPhilipp Gesang2015-11-218-175/+123
| |\ \ | | | | | | | | update fontloader
| * \ \ Merge pull request #291 from phi-gamma/masterPhilipp Gesang2015-11-2048-3588/+6446
| |\ \ \ | | | | | | | | | | pluggable fontloaders
| | | | * for 2.6David Carlisle2015-11-211-3/+5
| | | | |
| | | | * just use ltluatex not luatexbaseDavid Carlisle2015-10-031-17/+7
| | | | |