The last update to Roboto Italic has a stray point in delta.alt with a
y value of -11299. This causes metrics to give too much vertical space
in some cases.
This is a "hot patch" fix, with the stray point deleted using TTX. We
want to fix the font upstream as well, but this should fix things in
the mean time.
Change-Id: I08edae96f69753b2d46dfc3fd1cc3b9176dcb5ad
This patch updates the Droid Serif fonts to version 20130131. This
fixes bug 6180699 and should expand the range of combining marks
(e.g. Polytonic Greek, African, Canadian aboriginal languages)
significantly.
Change-Id: I31975d0b35f5164b2da9fd7a5df0474aaed3ed5c
This updates Roboto to the latest version. This change should fix a
number of problems, including bug 7306377 (vietnamese o with horn), bug
6679075 (vietnamese u+1ee1), bug 7568194 (cyrillic yeru), and is also on
the critical path of bug 7291977 (t comma accent), bug 8278292 (E
ogonek) and likely others.
Change-Id: Ife3db2aec4784a54cfe0287b61eac8339c198c71
Renames DroidNaskh-Regular-SystemUI to DroidNaskhUI-Regular to be
consistent with naming scheme for Noto fonts.
Change-Id: I3319ee5cd85360c85b0523bd932d461eb05d1b18
This change switches from the older (cibu and lohit) Indic fonts to
the new Noto fonts. All the new fonts come in both normal and bold
weights, and with system ui (more compact vertical metrics, to match
Roboto) variants as well as normal.
The fonts have been tested on a phone (Galaxy Nexus) and seem to work,
but should be more carefully evaluated by native speakers.
Change-Id: Ib0a10ee5a130bab258ff198e22b8304e00141073
This change moves Telugu back before the fallback font. Previously it
had been moved to after it because it contained incorrect mappings of a
number of characters, including U+2212 and U+FFFD. These fonts have now
been fixed. When placed after the Droid Sans Fallback font, the glyph
count for complex script shaping exceeded the 2^16 limit, causing
incorrect rendering.
This patch depends on the fix for the Lohit-Telugu font.
Change-Id: Ib1bc665a246c4ad65a0dea1c261df5696b440344
Fix for bug 6630331.
This is a workaround for a shortcoming in the Thai shaper in Harfbuzz.
De facto (but not documented) shaping rules dictate that a tone mark
followed by a SARA AM should be rendered with the NIKHAHIT below the
tone mark, rather than above it as would be indicated by a
straightforward interpretation of the GSUB and GPOS rules. This patch
adds an additional substitution of the form MAI THO + SARA AM ->
NIKHAHIT + MAI THO + SARA AA. The three tone marks affected are MAI EK,
MAI THO, and MAI CHATTAWA, based on an usage analysis of the Thai
dictioary included in ICU.
When we upgrade to harfbuzz-ng, we can revert this workaround.
Change-Id: I209ed094e7c89ea7f6fd6e983d37e396c6da1f64
Many media files and source code files were marked as executable in Git.
Remove those.
Also a shell script and python script were not marked as executable.
Change-Id: Ieb51bafb46c895a21d2e83696f5a901ba752b2c5
This patch fixes bug 7309211 [ar][fa] Riyal ligature mistakenly on in
Android Arabic fonts
It also fixes a number of other problems in the normal (non-system ui)
versions of the Naskh fonts, similar to the changes applied for the fix
to bug 6872613, and makes those fonts more consistent with the system ui
version.
3. Fixed a wrong GSUB substitution for Urdu Seven
4. Fixed shaping of U+06C1 HEH GOAL
5. Fixed shaping of U+06BA NOON GHUNNA
6. Removed the Rial ligature from the GSUB table
7. Tightened the Alef-Lam-Lam-Heh ligature in the GSUB table which was
getting triggered too frequently
Thanks to Roozbeh for the updated font data.
Change-Id: I29aecba2ea62853a8ed3181c023c62a09e800def
Updated Roboto fonts fixing issue where Tcommaaccent and Tcedilla
characters were encoded improperly (their values were switched). This
patch addresses bug 7198855 and bug 7070912. Original was mistakenly
submitted to master but should be applied to jb-mr1-dev instead.
Also, this commit reverts the addition of RobotoCondensed-Light and its
italic, for which we don't have any build file or XML metadata support.
Change-Id: If1c592474f1f7199e98bb98edf4de6466b43fa51
The primary difference between the AndroidClock and Roboto-Thin fonts is that the new clock font defaults to variable width characters for better alignment on the lock screen.
Bug:7206207
Change-Id: I8509324510ce01d0a73ab1bb6b730b5d646a462c
Latest versions of DroidSansFallback from Monotype.
The DroidSansFallback.ttf file has some additional glyphs and
glyph fixes (including a fix for bug 6723057 and will likely fix
bug 6629748). It continues to cover Korean Hangul but does not
cover CJK Ext A (for space reasons on small system image devices).
The DroidSansFallbackFull.ttf file has the bug fixes listed and
also removes the Korean Hangul because we are now going to use
NanumGothic for Korean (NanumGothic.ttf is added in a separate
CL in the external/naver-fonts directory).
The falback_fonts.xml file has been modified to add NanumGothic.ttf
before DroidSansFallback.
Bug: 4531601
Bug: 6723057
Bug: 6629748
Change-Id: I670d33078b4a97c4eda00fc2323be187696e927a
This change adds the Roboto Thin and Thin Italic font variants, which
are accessible through, for example, Typeface.create("sans-serif-thin",
Typeface.NORMAL).
The version checked in is Roboto_1.100150 as obtained from Christian.
The other fonts in the family should probably be rev'ed as well, but
that would be a separate change.
I verified that the fonts work, that they're thin, that kerning and
ligatures work, and that the "kern" table (redundant and bulky) is
missing.
This should close bug 7173069: Integrate roboto thin to support jb mr1
time displays
Change-Id: I5c950a1843675668b83c005234b44abbe924d9fe
Roozbeh provided an updated Arabic system UI font which does positioning
properly (using the GPOS mechanism, which is now fully supported in text
rendering), as opposed to the old workaround of using GSUB combinations.
It contains the following additional fixes:
1. Removed the new glyphs and relevant table fields created by GPOS2GSUB
2. [Kept empty glyphs for ZWSP, ZWNJ, and ZWJ]
3. Fixed a wrong GSUB substitution for Urdu Seven
4. Fixed shaping of U+06C1 HEH GOAL
5. Fixed shaping of U+06BA NOON GHUNNA
6. Removed the Rial ligature from the GSUB table
7. Tightened the Alef-Lam-Lam-Heh ligature in the GSUB table which was
getting triggered too frequently
This change just checks in the above work. I did a sanity-check
verification that Arabic text still renders reasonably in the system UI,
but did not check the above cases carefully.
Change-Id: Ie5fb9fd78d21bbfcd62b86deb327a5a2ae9321c6
Cherry-pick Ib5dd86950156c5a438f25c289acb839206bb455a from master.
Data: label MTLmr3m with "ja" locale attribute, fallback_fonts-ja.xml removed,
as we only need a single fallback font file
Code: Add locale and variant to TextLayoutCache. Paint.java sets textLocale as
the language (for example, "ja") rather than the language/locale concatenated
(for example "ja_JP")
This checkin, along with Change-Id: Id8c91ae0be6cad8a7ef77a0cd5803676290986c1,
allows text view objects to set their locale dynamically and skia will use the
correct font for the locale.
Change-Id: Ieb60b0d7a39fcfef4f8ce90cd4f6065d33673710
Lohit-Telugu.ttf provides incorrect glyphs for U+FFBC-10007, masking
fonts later in the fallback chain that provide correct glyphs for
that range. I've temporarily demoted Lohit-Telugu in the fallback
order so that this does not happen, and appended a comment that
explains why. In the future, when we have our own Telugu font, we
can place that with the rest of the Indic fonts in the fallback list.
Bug: 6723057
Change-Id: I15fb06dcb7dd13289e226de01c36058134a1b7cb
Font files are marked with elegant or compact in attributes.
This is the data check for the code checkin with a similar title
Bug: 6649136
Change-Id: Ie246ae0cff55c1e76c0f32012a740547274e8221
This patch changes the font data so that all anchor points are in
absolute coordinates, rather than relative to outline points. With
the workaround, we don't trigger the bug in the HarfBuzz/Skia
integration that causes incorrect outline point data to be read.
The font was prepared with FontForge. There's some risk of other
problems, but I tested the font fairly carefully and it looks fine.
(With a version prepared in TTX, the substitution of the different
width forms of the ivowelsigndeva got broken, likely due to a TTX
bug, as this seems to happen with roundtrip of unedited original).
Change-Id: I3f6fffce91c3ea0695e6005b051bcf4e9b6f6361
This adds the Roboto Condensed family (with the name
sans-serif-condensed, for consistency with existing family names).
Change-Id: Ied2c509ebbb2b3b8637c659e62c9a821b6941eaa
These fonts are replacements for the Lohit Devanagari/Tamil fonts.
We need to fit Devanagari onto all builds, so we only provide a single,
regular-weight font there, but Tamil is omitted for SMALLER_FONT_FOOTPRINT
builds and gets UI/non-UI and Regular/Bold weight versions. The UI versions
of the fonts are used for UI/system display, and the non-UI versions are
used only by WebView.
Bug: 6318791
Change-Id: I50ff6ec4bb428c0ac30049273f03a94de05b0c4f
- bug #5987379 Need an Arabic font with metrics "compatible" with Roboto
- use the Alt version with GSUB optimizations
Change-Id: I4d8c62cab37a7b010abab602c39899084d347fdc
This is the latest drop (ver 1.00) of DroidNaskhShifAlt.ttf
from Monotype. Checking it in so we can see if there is any
clipping.
Bug: 5987379
Change-Id: I25697e64c1643dc4d36fe4ff6a407549c1165d60
