summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* | Merge pull request #379 from phi-gamma/masterPhilipp Gesang2016-09-2015-137/+983
|\| | | | | latest loader and changes for a pre-release
| * [fontloader] HACK hot-patch loader to fix crashv2.8-rc1Philipp Gesang2016-09-201-4/+6
| | | | | | | | Patch has been upstreamed for a while but no response yet.
| * [db] install required functions from font-synPhilipp Gesang2016-09-201-0/+44
| | | | | | | | | | | | The loader makes some assumptions about available lookup functions early on. Since fonts-syn.lua only installed dummies for most of these, we might as well do that too.
| * [mkimport] drop fonts-syn.lua from mergePhilipp Gesang2016-09-191-1/+0
| | | | | | | | | | No idea how this ended up being merged in the first place. Luaotfload does its own name handling.
| * [fontloader] sync with Context as of 2016-09-15Philipp Gesang2016-09-1512-132/+933
| |
* | Merge pull request #378 from phi-gamma/masterPhilipp Gesang2016-08-2914-909/+1145
|\| | | | | [fontloader] sync with Context as of 2016-08-29
| * [fontloader] sync with Context as of 2016-08-29Philipp Gesang2016-08-2914-909/+1145
| |
* | Merge pull request #376 from phi-gamma/masterPhilipp Gesang2016-08-157-788/+1748
|\| | | | | font-otc now part of the loader
| * [features] drop code from font-otcPhilipp Gesang2016-08-151-778/+8
| | | | | | | | | | | | | | | | | | addfeature() and the likes -- with less extensive logging -- is now provided directly by means of font-otc.lua so we don’t need to keep these parts around any longer. The loader installs anum features identical to ours and a slightly different “tlig” one which we override. Clean up locals and misleading comments while we’re at it.
| * [scripts, init] include font-otc at the appropriate placesPhilipp Gesang2016-08-153-9/+14
| |
| * [fontloader] sync with Context as of 2016-08-15Philipp Gesang2016-08-153-1/+1726
| | | | | | | | | | CAVEAT LECTOR This adds font-otc.lua part of which has already been included in our feature handler.
* | Merge pull request #374 from phi-gamma/masterPhilipp Gesang2016-07-307-707/+829
|\| | | | | lastest feature code
| * [features] use the new feature registry for TFMPhilipp Gesang2016-07-301-30/+30
| | | | | | | | No need for patching font-tfm.lua any longer.
| * [fontloader] sync with Context as of 2016-07-30Philipp Gesang2016-07-306-677/+799
| | | | | | | | Hans appears to have heard our pleas =)
* | Merge pull request #373 from phi-gamma/masterPhilipp Gesang2016-07-2817-448/+248
|\| | | | | fix adding of custom features
| * [features,fontloader] fix adding featuresPhilipp Gesang2016-07-273-11/+55
| | | | | | | | | | | | | | | | | | | | This makes our own features work with otf.addfeature(). However, for TFM it’s still not functional out of the box since the enhancer is installed at a time when it has not been defined yet. An answer from Hans is pending. In the meantime, we put a crude hack into our font-tfm.lua to allow injecting the enhancer retroactively on Luaotfload init.
| * [loaders] take tfm path directly where appropriatePhilipp Gesang2016-07-271-2/+3
| | | | | | | | | | The AFM one is still preferable when passed a PFB because of the extra checking it does.
| * [features] remove check preventing application of features to TFM filesPhilipp Gesang2016-07-271-3/+4
| |
| * [aux] unrestrict fontname rewritingPhilipp Gesang2016-07-271-7/+5
| | | | | | | | | | Since TFM fonts now take the same path as OTF and suchlike, not passing on the spec literally may actually be harmful.
| * [fontloader,import,status] drop luatex-plain-tfm.luaPhilipp Gesang2016-07-273-123/+0
| |
| * [fontloader,import] drop font-otd.luaPhilipp Gesang2016-07-271-268/+0
| |
| * [fontloader] sync with Context as of 2016-07-26Philipp Gesang2016-07-268-34/+181
| |
* | Merge pull request #370 from phi-gamma/masterPhilipp Gesang2016-07-152-20/+6
|\| | | | | fix TFM loading
| * [loaders] hook the generalized afm loader directly into tfm, pfb, and afm ↵Philipp Gesang2016-07-151-20/+4
| | | | | | | | | | | | | | formats Latest updates to the loaders seem to have obsoleted a number of our precautions. Neat.
| * [features] correctly unset subfontPhilipp Gesang2016-07-151-0/+2
| | | | | | | | | | Even though it works fine as it was, the loader actually expects “false” as the value to indicate that no subfont was requested.
* | Merge pull request #369 from phi-gamma/masterPhilipp Gesang2016-07-1432-768/+2113
|\| | | | | New TFM code
| * [loaders] make TFM take the AFM pathPhilipp Gesang2016-07-141-1/+2
| |
| * [fontloader] sync with Context as of 2016-07-14Philipp Gesang2016-07-144-135/+252
| |
| * [features] include feature spec validationPhilipp Gesang2016-07-081-0/+36
| |
| * [tool] update load sequence for font-tfmPhilipp Gesang2016-07-071-2/+4
| | | | | | | | | | | | Also set a dummy ``pdf`` table that is referenced in the file. Whether we actually need font-tfm in luaotfload-tool should be reevaluated at some point before 2.8.
| * [status,import,fontloader] add luatex-plain-tfm.lua and update scriptsPhilipp Gesang2016-07-073-3/+124
| |
| * [features] update otf.enhancers.addfeature()Philipp Gesang2016-07-071-15/+75
| |
| * [init] postpone loading of font-tfmPhilipp Gesang2016-07-071-2/+2
| |
| * [fontloader] sync with Context as of 2016-07-05Philipp Gesang2016-07-0522-611/+1619
| |
* | Merge pull request #366 from phi-gamma/masterPhilipp Gesang2016-06-1611-411/+447
|\| | | | | [fontloader] sync with Context as of 2016-06-16
| * [*] update datev2.7-fix-3Philipp Gesang2016-06-165-6/+6
| |
| * [fontloader] sync with Context as of 2016-06-16Philipp Gesang2016-06-166-405/+441
| |
* | Merge pull request #365 from phi-gamma/masterPhilipp Gesang2016-06-152-6/+6
|\| | | | | fix dynamic db initialization
| * [doc] clarify behavior of update-live in config manpagePhilipp Gesang2016-06-151-1/+2
| |
| * [db] fix initialization behavior to prevent crashPhilipp Gesang2016-06-151-5/+4
| | | | | | | | | | | | | | | | | | | | | | Reported by @dohyunkim: https://github.com/lualatex/luaotfload/pull/364#issuecomment-226059150 Under certain circumstances, update_names() was invoked with an empty table instead of a correctly initialized one, breaking the assumptions of the db populating code. This commit also guards more strongly against this kind of oversight and tidies up the db constructor.
* | Merge pull request #364 from phi-gamma/masterPhilipp Gesang2016-06-1535-2685/+4790
|\| | | | | fixes, 3rd edition
| * [fontloader] sync with Context as of 2016-06-14Philipp Gesang2016-06-143-21/+27
| |
| * [db] adapt error handling to new fontloader conventionsPhilipp Gesang2016-06-121-14/+21
| |
| * [db] fix crash with missing name dataPhilipp Gesang2016-06-121-2/+2
| |
| * [fontloader] import font-oclPhilipp Gesang2016-06-126-8/+315
| | | | | | | | | | Some of the more absurd aspects of fonts ;) This will fork inkscape for the actual SVG processing.
| * [fontloader] sync with Context as of 2016-06-12Philipp Gesang2016-06-1218-86/+1464
| |
| * [features] fix missing localPhilipp Gesang2016-06-111-2/+3
| | | | | | | | | | | | | | The new addfeature() as of 499160de78a1 accesses utf8.len() which wasn’t imported at this point. Another import, utf8.char(), became irrelevant. Spotted by @doyunkim
| * [init] adapt Context base path for file loaderPhilipp Gesang2016-06-101-10/+31
| | | | | | | | | | | | | | | | | | Fix issue #344 The paths changed a while ago making a change necessary. We can’t just replace the path because older versions of the tree would fail, rendering bisection unusable. We compensate by testing the candidate directories beforehand.
| * [fontloader] sync with Context as of 2016-05-31Philipp Gesang2016-06-105-78/+128
| | | | | | | | | | | | This corresponds to commit a274872832cd on the mirror repo. The changes since then were numerous but seem to introduce a couple new files we didn’t ask for =) Later.
| * [db] resolve symlinks when traversing the file systemPhilipp Gesang2016-06-081-31/+41
| | | | | | | | | | | | | | | | | | Addresses #359 and #325 To avoid duplicate entries, paths have to be resolved before collecting them. This necessitates loop detection of some sort, currently implemented naively as a flat table containing the directories already traversed.