Change-Id: I002470c2888d6dec876af18c2c016cc789eb399e

We need to pass the location of the icu build root for the BUILD platform with
the --with-cross-build option when configuring icu for cross-compilation (to
the HOST platform). I.e. we always need to build it for the BUILD
platform. Surely this could be done in a more elegant way, but I got lost in
the mess, and just hacked it in the cross-toolset rule in Makefile.in.

Change-Id: Ia5e9d13d01067c6dded1fdc86c56f8f39aae6bc8

Change-Id: I874957a399c306bfdb057322edb09c0b56373f10

Change-Id: I2f0ae1eff5bd009a7cb467b38c1743eff51dc563

Change-Id: Id829cd4f12cb4d3fc8cb08dc5b297ee3cfc7591b

This should help a lot with Windows release build times (i.e., move it
from "glacially slow" category to just "slow").

Change-Id: I6d397eb5d1ec1779616c3eef8e51f9cf29d5acec

uiex differs from the other *ex tools in that translation for every
language must be in a standalone file, named after the language code. So
uiex should take an output _directory_ instead of a file.

Change-Id: If3ed966147c6d11d1fe85c484463f1bca4eec172

Change-Id: If62a89dc2f442f94cd0e36f4447ebc4899ee0ab3

Change-Id: I65c30e751fc01b63a54aa9c31b4b2a792b795a79

Change-Id: Idfea805082acaf4afa3fc198226379345ceda726

Change-Id: Id7d8bc05b1393cc2bae4a531c8a47f62df24b1d6
Reviewed-on: https://gerrit.libreoffice.org/1488Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

Change-Id: I3ce6d7c9cfba9f1391e9cb26df2573ea969d2f94

Change-Id: Ie5070d514c1f4dd9f9708253eb44134baec6524a

Change-Id: I085f14f40fb83cba171b35b14cf7c0d99091fc00

We still have checks for $(COM) being GCC or MSC in makefiles, of
course. But there is no reason to pass -D$(COM) to compilations.
Checking the built-in compiler-specific macros __GNUC__, _MSC_VER
(etc) is the right thing to do.

Change-Id: Ia961a29ba74e2c4977e5300a92318f38104c6592

Change-Id: Ibaf2501e6b655913963a53c1cd63668f1c4cccb2

Change-Id: I43464dd702c6b4a63df39b6a4b35bec63712d5e6
Signed-off-by: Marcos Paulo de Souza <marcos.souza.org@gmail.com>
Reviewed-on: https://gerrit.libreoffice.org/1486Reviewed-by: Tor Lillqvist <tml@iki.fi>
Tested-by: Tor Lillqvist <tml@iki.fi>

Change-Id: I57b1c99d1e4a92ca7d61f6c54683016945ffa14c

Change-Id: Ied7eef51ad46e79f8e3d55f8bdcf38d15fa68c09

Change-Id: I788473f88d78cd3818dfed20112fa36cea81de0d

Change-Id: I344030d48c3f872cc63b5c1f444d18075fd160f5

Change-Id: Ib7225307f51c8764c4b2b0e4ce45702d7919a017

Change-Id: I77f714bca9ddebffd6241d901661471aa27f03d5

Change-Id: I24c485d31ea3f0540164ff65074b392be1e557e6

When compiling using a PCH, includes included using the PCH usually don't
show up in compiler dependencies output, so the only the PCH has a dependency
on the include, and the source has only a transitive dependency. Switching
to non-PCH (e.g. ENABLE_PCH= on cmdline) means the dependency is lost.
Trying with the PCH timestamp was a nice try, but it can sometimes still
miss something. So whoever wants to switch needs cleaning first or
just know what they're doing.

Change-Id: I3bdf909d169696b0ce79ee9830aa99ae880636db

Not that I have any clue about dmake, but without this it seems
like incremental make doesn't remake external if necessary, and
with this copied and adjusted from other non-tail_build module
it seems to work *shrug*.

Change-Id: If6818860d12ed752814574028895870a60d6a7d1

I'm not sure if this is clang being more strict or buggy, probably the latter.

Change-Id: I900ba2cd2b1d4c23e41e4787be4263954ce8b4c0