summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Fix import order in Build.pyJames Godfrey-Kittle2015-11-20
|
* Merge pull request #135 from jamesgk/italicsJames Godfrey-Kittle2015-11-20
|\ | | | | Italics stuff
| * [italics] Add some commentsJames Godfrey-Kittle2015-11-19
| | | | | | | | It's not much, but maybe a start....
| * [italics] Work around RoboFab-related issuesJames Godfrey-Kittle2015-11-19
| | | | | | | | | | | | | | | | | | | | | | | | This is (hopefully) a temporary hack until a more robust solution to the issue is found. Basically the problem is that RoboFab pens do float-to-float comparisons to decide whether to remove duplicate points from contours, and breaks with BooleanOperations output (which can include point coordinates with very small fractional components). The italicizing code assumes certain behavior from RoboFab pens and can't handle duplicate points being kept.
| * [italics] Fix import order, remove unused importsJames Godfrey-Kittle2015-11-13
| |
| * [italics] Fix whitespaceJames Godfrey-Kittle2015-11-13
|/
* Correct dates in sourcesJames Godfrey-Kittle2015-11-03
| | | | | These dates seem to be incorrectly exported by vfb2ufo. The new dates are taken from the currently available webfont binaries.
* Merge pull request #129 from anthrotype/readme-bool-opsJames Godfrey-Kittle2015-10-23
|\ | | | | [README.md] modify instructions for installing booleanOperations
| * [README.md] modify instructions for installing booleanOperationsCosimo Lupo2015-10-23
|/ | | | | | | The current master/HEAD no longer requires to run two different setup.py scripts and manually copy the *.so module. https://github.com/typemytype/booleanOperations/pull/6
* Merge pull request #119 from jamesgk/testingJames Godfrey-Kittle2015-10-12
|\ | | | | Use common tests from nototools
| * Use common tests from nototoolsJames Godfrey-Kittle2015-10-06
| |
* | Merge pull request #118 from jamesgk/testingJames Godfrey-Kittle2015-10-06
|\ \ | |/ | | Further generalize common_tests.py
| * Test version string, revision # as separate valuesJames Godfrey-Kittle2015-10-06
| |
| * Add example of expected_version definitionJames Godfrey-Kittle2015-10-06
| |
| * Further generalize common_tests.pyJames Godfrey-Kittle2015-10-06
|/
* Merge pull request #116 from jamesgk/testingJames Godfrey-Kittle2015-10-06
|\ | | | | Test consolidation
| * Move exhaustive tests to common_testsJames Godfrey-Kittle2015-09-29
| |
| * Use common digit width test for androidJames Godfrey-Kittle2015-09-29
| |
| * Move TestHints to common_testsJames Godfrey-Kittle2015-09-29
| |
| * Generalize and move vertical metrics testsJames Godfrey-Kittle2015-09-29
| |
| * Generalize coverage tests in common_testsJames Godfrey-Kittle2015-09-29
| |
| * Move all name tests to common_testsJames Godfrey-Kittle2015-09-29
| |
* | Merge pull request #114 from jamesgk/meta-fixJames Godfrey-Kittle2015-10-06
|\ \ | |/ |/| Provide a hotfix for webfont names
| * Provide a hotfix for webfont names.James Godfrey-Kittle2015-09-29
|/ | | | | | | | | | This ensures that we'll only have four fonts with family name "Roboto" (regular, bold, italic, and bold italic), which is necessary for Windows compatibility. Other styles have their weight appended to the family name. Also ensures that the macStyle bold bit is set for weight Black, to be consistent with older Roboto binaries.
* Merge pull request #113 from jamesgk/add-testsJames Godfrey-Kittle2015-09-29
|\ | | | | Temporarily disable superscript digit test for web
| * Tmp disable superscript digit test for web.James Godfrey-Kittle2015-09-29
|/ | | | | | | | See added comment for motivation. Probably the *.frac and *superior glyphs should be merged. More info: https://github.com/google/roboto/issues/22 https://github.com/google/roboto/issues/108
* Merge pull request #111 from jamesgk/add-testsJames Godfrey-Kittle2015-09-29
|\ | | | | Add family name tests
| * Add option to check heavier fonts as marked bold.James Godfrey-Kittle2015-09-28
| | | | | | | | | | | | Not sure if there's any rule for whether fonts heavier than Bold (e.g. Black) should be marked bold in the macStyle and subfamily name, so this adds an option to test either way.
| * Re-add check for bold bit in Black weight.James Godfrey-Kittle2015-09-28
| | | | | | | | | | | | | | | | Check removed in ca851298519eb19c3c89cc2b3a6c0770b04257a4. It may or may not be standard practice, but Roboto has previously set the bold bit for its Black weight so we'll keep it this way to avoid regression.
| * Remove unused imports.James Godfrey-Kittle2015-09-28
| |
| * Test that name and style are generally compatible.James Godfrey-Kittle2015-09-28
| | | | | | | | | | | | Changes to the TestNames unit tests reflect the issues raised in the link added to common_tests.py. Also generally account for code changed in nototools.
* | Fix typo in readme (availabe -> available)James Godfrey-Kittle2015-09-29
|/
* Merge pull request #110 from jamesgk/shape-fixJames Godfrey-Kittle2015-09-21
|\ | | | | Decompose nested components.
| * Rename decomposeGlyph's recursive helper function.James Godfrey-Kittle2015-09-21
| |
| * Reduce code outside of decomposeGlyph's recursion.James Godfrey-Kittle2015-09-18
| |
| * Decompose nested components.James Godfrey-Kittle2015-09-17
|/ | | | | If components of components aren't also decomposed, we may miss some contours in the resulting glyphs.
* Correct formula for Ncommaaccent.unic.James Godfrey-Kittle2015-09-17
| | | | | | | This seems to have been obviously incorrect, looking at the inconsistency with other Ncommaaccent formulas (commaaccent should always go below the N). And sure enough, the accent was not being added to the composite glyph.
* Merge pull request #107 from jamesgk/shape-fixJames Godfrey-Kittle2015-09-09
|\ | | | | Fix up *num features
| * Fix up *num features.James Godfrey-Kittle2015-09-04
| | | | | | | | | | | | | | | | | | | | A few things happening here: - References to *.lnum glyphs changed to *.pnum (reflecting the renames in the last commit). - Feature definitions expanded to handle feature combinations, not-working definitions removed from the resource directory. - Glyph classes added to make feature definitions clearer. - VFBs updated via vfb2ufo, based on updates to UFOs.
| * Rename *.lnum files to .pnum.James Godfrey-Kittle2015-09-04
|/ | | | These are proportional numbers, not lining numbers.
* Merge pull request #105 from jamesgk/shape-fixJames Godfrey-Kittle2015-09-02
|\ | | | | Don't use RGlyph.getParent.
| * Don't use RGlyph.getParent.James Godfrey-Kittle2015-09-02
|/ | | | | | Apparently this method returns a different font than we expect in decomposeGlyph, probably due to the glyph copying we do when mixing masters. The easiest thing to do is just not trust or use the method.
* Merge pull request #103 from jamesgk/add-testsJames Godfrey-Kittle2015-09-01
|\ | | | | Add test to compare glyph areas between weights
| * Add test to compare glyph sizes between fonts.James Godfrey-Kittle2015-08-31
| | | | | | | | | | | | For now, we just check which glyphs are unchanged between the Thin and Bold masters, and make sure only those glyphs are unchanged between Thin, Regular, and Bold output (ignoring empty glyphs).
| * Allow source and output UFOs to be tested.James Godfrey-Kittle2015-08-31
|/
* Merge pull request #100 from jamesgk/shape-fixJames Godfrey-Kittle2015-08-31
|\ | | | | Always deal with floats when mixing font masters.
| * Always deal with floats when mixing font masters.James Godfrey-Kittle2015-08-21
|/ | | | | | | | | Without this change, some of the master data contains int values which are not interpolated correctly due to numpy array behavior. For example adding float values to a numpy array of ints will always result in an array of ints, so a scale value which should be a float between 0 and 1 might instead be rounded.
* Merge pull request #97 from google/delivery-updatesdougfelt2015-08-20
|\ | | | | Delivery updates
| * Hinted TTF updates 15/08/20.James Godfrey-Kittle2015-08-20
| |
| * Unhinted output updates 15/08/20.James Godfrey-Kittle2015-08-20
| | | | | | | | Note that these binaries are also missing mark/mkmk features.