Source Activity

Syndicate content
Haiku's main repository
Updated: 27 min 27 sec ago

BSocketMessenger: Fix size mismatch.

2 hours 48 min ago
- The header size was accidentally being included in the flat buffer size, leading to a mismatch error when reading the incoming message. - The header size was accidentally being included in the flat buffer size, leading to a mismatch error when reading the incoming message.
Categories: Development

BSocketMessenger: Style fixes.

7 hours 19 min ago
Categories: Development

Added package for Einsteinium

13 hours 24 min ago
Einsteinium provides smarter monitoring of applications and system services for Haiku. It will restart applications and system services that quit or crash, gather statistics on application usage and provide customizable ranked lists of applications. Einsteinium provides smarter monitoring of applications and system services for Haiku. It will restart applications and system services that quit or crash, gather statistics on application usage and provide customizable ranked lists of applications.
Categories: Development

libbnetapi: Add socket messenger class.

Sat, 2016-04-30 22:29
- Introduces new network API class BSocketMessenger, allowing one to send and receive BMessages across a network socket in a BMessenger-like fashion. Still very much WIP, hence currently not exposed via public headers. Based partly on previous work by Axel. - Introduces new network API class BSocketMessenger, allowing one to send and receive BMessages across a network socket in a BMessenger-like fashion. Still very much WIP, hence currently not exposed via public headers. Based partly on previous work by Axel.
Categories: Development

Update bash package on x86_64.

Sat, 2016-04-30 16:14
Categories: Development

Update translations from Pootle

Sat, 2016-04-30 04:57
Categories: Development

LaunchRoster docs: Add warning not to use class

Fri, 2016-04-29 22:51
...as it is not yet finalized. ...as it is not yet finalized.
Categories: Development

Work around gcc 2 member template issue

Fri, 2016-04-29 21:02
Fixes the gcc 2 build. Fixes the gcc 2 build.
Categories: Development

ControlLook: Rename border param to borderStyle

Fri, 2016-04-29 20:57
border vs. borders is confusing. BTabView: Rename border param to borderStyle BTabView docs: rename border param to borderStyle enumerate border styles in docs border vs. borders is confusing. BTabView: Rename border param to borderStyle BTabView docs: rename border param to borderStyle enumerate border styles in docs
Categories: Development

BTabView: Draw bottom tab border in DrawTabs()

Fri, 2016-04-29 20:57
...instead of DrawBox(). Also use ceilf when calculating tab height to prevent non-integral height. Fixes #12683 More Todo: You have to understand way too much about how this class draws if you want to have any hope of overriding one of its Draw... methods and have it do what you expect. e.g. The BeBook implies that the tabs are drawn first, then the box, but, we draw them in the opposite order. Probably better this way but not intuitive. There are a number of remaining questions: 1. Why don't we need to draw the bottom of tabs if B_FANCY_BORDER? 2. Why do we need to expand tab border horizontally if B_PLAIN_BORDER? 3. Why is the bottom border color (152, 152, 152) instead of (151, 151, 151)? Add a bunch of TODOs for these questions and more. There can be some unitiutive gaps between the box border and view depending on if you choose B_FANCY_BORDER or B_PLAIN_BORDER. You don't notice the gaps unless the view draws right on it's edge. Some views, including in Devices and Media Prefs do this though. Media Prefs further complicates matters by overriding BTabView. ...instead of DrawBox(). Also use ceilf when calculating tab height to prevent non-integral height. Fixes #12683 More Todo: You have to understand way too much about how this class draws if you want to have any hope of overriding one of its Draw... methods and have it do what you expect. e.g. The BeBook implies that the tabs are drawn first, then the box, but, we draw them in the opposite order. Probably better this way but not intuitive. There are a number of remaining questions: 1. Why don't we need to draw the bottom of tabs if B_FANCY_BORDER? 2. Why do we need to expand tab border horizontally if B_PLAIN_BORDER? 3. Why is the bottom border color (152, 152, 152) instead of (151, 151, 151)? Add a bunch of TODOs for these questions and more. There can be some unitiutive gaps between the box border and view depending on if you choose B_FANCY_BORDER or B_PLAIN_BORDER. You don't notice the gaps unless the view draws right on it's edge. Some views, including in Devices and Media Prefs do this though. Media Prefs further complicates matters by overriding BTabView.
Categories: Development

Debugger: Support for reading symbols from core file images

Fri, 2016-04-29 20:31
* New class ElfSymbolLookup. Given a data source, the address of a symbol table, symbol hash table, string table, and a few more details it provides access to the symbol information. * CoreFile: Add CreateSymbolLookup(). Creates an ElfSymbolLookup for a specified image. * CoreFileDebuggerInterface: Implement GetSymbolInfos() for regular images (no commpage yet) via the new functionality. * New class ElfSymbolLookup. Given a data source, the address of a symbol table, symbol hash table, string table, and a few more details it provides access to the symbol information. * CoreFile: Add CreateSymbolLookup(). Creates an ElfSymbolLookup for a specified image. * CoreFileDebuggerInterface: Implement GetSymbolInfos() for regular images (no commpage yet) via the new functionality.
Categories: Development

Core file images note: Add text delta field

Fri, 2016-04-29 20:18
Categories: Development

extended_image_info: Add text_delta field

Fri, 2016-04-29 20:16
Needed to offset a symbol address read from the symbol table to the actual load address. Needed to offset a symbol address read from the symbol table to the actual load address.
Categories: Development

net_socket: check underflow condition happening on x86_64.

Fri, 2016-04-29 15:18
* move the test for the availability of the add_anillary_data function in the protocol out of the loop, as suggested by Axel. * fix #12737. * move the test for the availability of the add_anillary_data function in the protocol out of the loop, as suggested by Axel. * fix #12737.
Categories: Development

unix: fix x86_64 build with trace.

Thu, 2016-04-28 20:48
Categories: Development

intel_extreme: Improve LVDS panel detection robustness

Thu, 2016-04-28 15:45
* If older generation, check for mobile. If mobile GPU is found, make an assumption that a LVDS panel exists and attempt to leverage the vbios or VESA EDID. * If older generation, check for mobile. If mobile GPU is found, make an assumption that a LVDS panel exists and attempt to leverage the vbios or VESA EDID.
Categories: Development