A set of new tags for themes that allow them to display a quickscreen item's name or value like what is displayed on the default quickscreen.
There are 8 tags in total, 2 for each direction or "item".
One type of tag displays the setting name, while the other displays the setting's value.
All tags output an "ERR" string if no valid setting is found for that item.
Quickscreen Item name tags: %QT, %QR, %QB and %QL.
Quickscreen Item value tags: %Qt, %Qr, %Qb and %Ql.
Change-Id: Ia08ba5940e38065e051a0aefa2cff142c9e58684
reading the disk works fine for on disk playlist but
trying to read from the disk with the current playlist
becomes unbearably slow
removes the static playlist_track_info prefering the unused one
already on the stack from search_playlist()
Change-Id: I01b836b4fe46bb51ef6a28d5db6b3f9cdc7d1e51
the extra title pointer and alignment adds around 1k to the bss area
since we already have a pointer to track->name we can just save an offset
for the title data
Change-Id: I3a19857631d70276134bcc97940824a3e2f80e4a
Not sure this is a great idea from disk and battery standpoint
but there is no reason you can't..
using the name buffer to fill title data
prevent hitting the disk for each screen scroll
add get_metadata_ex to allow flags
METADATA_EXCLUDE_ID3_PATH
prevent copying the filename to the ID3 struct
METADATA_CLOSE_FD_ON_EXIT
instead of seeking to the beginning the file is closed before
get_metadata returns
add logic to allow a invalid fd to signal that get_metadata
should open and close the file within its call
bugfix per Chris_s don't use the tagcache for the trackinfo
Change-Id: Ic7a595b39a8d7a57f975312bc9c8bb4111f22a88
clean-up a bit more
add/correct some comments
fix some error passing
guard delete path on PATH_TOO_LONG
add some cpu_boost
Change-Id: Icf179dd727271bdc61ab78400e10847222b9f858
- offscreen ball check;
- "suicide button", with its documentation in the help (defaults to Down+Fire+Quit on targets that have DOWN defined, Fire+Quit held for some time for the others) ;
- help screen made a little more clear (IMO) by underlining the key names;
- help screen *code* made a little easier to handle, adding the last element number in the words array (this eases the formatting specified in the other array);
- now the game does not delete the savefile after it loads it, but only by request (to do it a new menu entry was added); so, now "Quit without saving" will actually leave you to the state of the last savefile instead of resetting the game;
Change-Id: If3eeabada1566dcd4a9473d3d1e361d4921e4005
* Correct broken CREDITS file causing HTML manual build failures
* Fix warning on iAudio M3 and MPIO HD200
* Fix XRGB888 pixelformat builds, hopefully correctly.
This should bring the board back to green.
Change-Id: Ia7d54641d37db19d4cee3b5d9f5d8ab1567ac30b
* Piles of warnings in miniz when built with modern toolchain
* Pointer arithematic error in PRNG
* Casting between int and void * for file descriptors
* Warning on non-color targets
Remaining:
* Failure on XRGB888 LCDs due to issue with core LCD macros
* Failure on interleaved greyscale LCDs (?)
* HTML manual build failure
Change-Id: Ibf6d2c001ec8daf583731d5da15b86b5352773e7
original xrick code by 'BigOrno' at:
http://www.bigorno.net/xrick/
Rockbox port, plus bugfixes at:
https://github.com/pierluigi-vicinanza/xrick
Further changes:
* Additonal fixes from g#3026
* Port to modern plugin API
* Add Pluginlib keymap fallback
* Support all >1bpp screens
* Fix build warnings in miniz
* Better error message when resources are missing
Change-Id: Id83928bc2539901b0221692f65cbca41389c58e7
first some clean-up of onplay.c
extend/move fileobject copy, move, delete routines in prep for other users
add error checking, better error codes
pre scan to make sure the operation doesn't exceed system resources
show progress for file and directory copies
Change-Id: Ife2e62df554892dab651bab40433bf70b27e73ff
The fft plugin uses pluginlib osd stuff _and_ greylib together,
which results in two conflicting declarations of the _grey_info struct.
Normally we treat the one in pluginlib as __weak but we disable that on
windows simulator builds, because the osd code ends up requiring it.
due to not using-function-sections and -fdata-sections on simulator builds.
Simply disabling all of that crap doesn't seem to help.
So instead, for simulator builds, mark the fft's struct as extern, only
using the one in pluginlib.
Change-Id: I1adf9acc265a60f91b8ac0fcad9d792e3240a9d9
10 is too short for some localizations (for instance russian 103.23МиБ requires 13 chars due to utf8 encoding). Use 20 instead
Change-Id: I4c7d1233358c8322545fce45dc798691410dd21e
Improves readability on small screens.
See WPS -> Context Menu -> Current Playlist -> Playing time
Also, Russian translation is adjusted accordingly.
Change-Id: I23662ccb324f07a742179ff274453416bb8eff08
Unfortunately, this brings us back to the "Default" behavior of simply
not voicing talk clips at all if there's no voice file.
Change-Id: I457ec4e3fcefaaffcbf6222fe01617bf0903518c
* Voice generation script will create standalone .talk clips
* These talk clips will be included in the rockbox .zip file
* All .voice files will be included in the rockbox .zip file
* Added LANGUAGE_NAME for all languages in the nightly builds
This way, any voice pack installed will give you a the langauge
voiced in the browser, and if the voice file fails to load you
will get a natively translated error message.
Change-Id: I6b627a51746cd088d6e200666dd326ea2745f55f
While playing, the user can switch between
status/samples/instruments/comments screens. The instrument names are
blindly passed to printf, which will lead to a null pointer dereference if
the instrument name is NULL.
...Which happens with some .IT modules that I have.
This should arguably be fixed in the printf implementation, but we still
shouldn't be passing in NULLs as string arguments!
Change-Id: I2899e2f7e10565c251e495752b3ef1bbccea82c7
If disk_mount_all() fails at startup, the device should enter USB mode,
so the storage can be repartitioned/reformatted. After
unmounting/ejecting, the device is rebooted. Unfortunately if the
storage is a RAM disk, the data won't persist after a reboot, so this
patch tries to mount the storage again, instead of just rebooting.
Change-Id: I421a9fd8ae536bee07d292f27d1da0615456df62
If no split tokens are found the string just goes off into space
instead shoten the string till it fits and continue on with the
remaining string
Change-Id: I7fa3619fe8c75ec6c849996d4c3518409938152b
* Only codecs and plugins are enabled
* Only native builds (so far)
* Only tested on xDuoo X3 (MIPS, monochrome)
* opus & speex generate some warnings
* Significant compile time impact
Change-Id: I519b0d179631a54b2103cd67225bd5ec6ad3bd2f
path_append throws away the basepath if currdir starts with a slash
I thought I had this fixed but it was just getting covered up
instead step past all the slashes and use that for our append dir
Change-Id: I170a2af99455cea09e6e695dfb89fdd42733bd4b
(This is a different implementation of the fix in g#5726, by Vencislav Atanasov)
The core problem is that the user is prompted to insert a USB cable
to fix the partitioning etc but the code that monitors for USB insertion
hasn't been started yet. Correct this.
If no USB support is present, reboot after 5 seconds if it's not a debug
build.
If USB support is present, want for insertion first, then do the
reboot-if-not-debug behavior.
Change-Id: I87827e7fe2fe9a02298918c6ebc4d8a9fb33d624
* Converted over all previously-translated strings
* Lots of comments added to mark the next phase
* Voice filename of the file being loaded if possible
Change-Id: Ia3219bbfa4505ad29a25bcf1d6eacc94c59a2a83
The strings in the struct configdata[] are only used in the settings
file, and not displayed. Therefore there is no need to translate them.
Change-Id: Id9e03c4b0454412b319ad678e6f57a46e1da0f6e
