Android Interacting with Other Apps using Messenger or AIDL

在Linux階層上的IPC，Android用了iBinder來包裝之間溝通的管道。iBinder有兩種形式可以傳遞資料。預備知識是使用android的service並且exported，可使其他application能夠喚起該service。


第一種是用Messenger封裝起來，使用Message來傳遞，開啟Service後，建立一個Messenger instance，使用Messenger的getBinder提出傳給service，而client也是透過取得的binder來創建Messenger來溝通。並且可以使用Messenger.replyTo來建立一個收訊息的接口來跟遠端service溝通。溝通的單位是bundle。

第二種是使用Android Interface Definition Language，簡單的說就是iBinder中定好溝通的規則，也就是functions。在server(service端)建立好iInterface的aidl實作的binder，在service中回傳該binder，然後在client端中也是用aidl檔承接轉型該ibinder至iInterface，雙方就透過iInterface中定義好的functions溝通。





一些概念解釋：


iBinder就是android中建立在ipc上的java層，用來使用ipc來溝通的。

Messenger其實就是封裝iBinder，使用Messenger類來透過ibinder來溝通。

AIDL則是在ibinder層先定義好傳輸資料，接著雙方用該接口來互動。該檔案需在sdk中編譯，並且對應相同的pacakge name，比如說你aidl package name是com.handsome.hydra你就得放在src/main/aidl/com/handsome/hydra/的資料夾下放IInterface.aidl，此後sdk編譯好後你才能使用IInterface.stub方法。

若你有使用ipc shared memory, message queue的經驗，基本上java那層就是封裝過的實作層，讓你用他們定義好的資料傳輸方式來使用ipc就是了。

Build nexus player AOSP

我用Virtualbox裝了Ubuntu14.04來安裝，比較保險，I like to do it in an old fashion way.

基本上全部的步驟都在document裡面了
https://source.android.com/source/requirements

我列出過程中需要注意的地方

• adb usb driver跟fastboot usb driver是不同的

基本上用sdk的universal usb driver都可以support. 只是兩個是不同的，都得assign進vm

• fastboot某些權限需要root user. 只能跑在root user裡

build好以後用fastboot燒時指定product directory.

export ANDROID_PRODUCT_OUT="/home/hydrated/Android/out/target/product/fugu"

之後flashall進去以後如果順利 LeanbackLauncher長這樣 (wtf..)

Remote也辦法正常配對，我甚至認為這篇文章可以subtitle: how to pair your nexus remote without native supported.

好險有adb，用盡我對android的了解，終於找到辦法

     1. 首先先用am start AddAccessoryActivity
am start -n com.android.tv.settings/.accessories.AddAccessoryActivity

     2. 正常的remote配對方法，按下back and home，就能配對了。

     3. 找到device後別開心，因為你按不下enter！(wtf again)，好險一樣有adb

shell input keyevent 66

..終於配對成功，只花兩天而已。然後software keyboard也沒支援，打wifi password時長這樣

fugu:/ $ input keyevent 46                                                     

fugu:/ $ input keyevent 43                                                     

fugu:/ $ input keyevent 29                                                     

fugu:/ $ input keyevent 41                                                     

fugu:/ $ input keyevent 51                                                     

fugu:/ $ input keyevent 29                                                     

fugu:/ $ input keyevent 42                                                     

fugu:/ $ input keyevent 32                                                     

fugu:/ $ input keyevent 33                                                     

fugu:/ $ input keyevent 46                                                     

fugu:/ $ input keyevent 66

adb shell input text 'password'

另外一些make指令

mmm -j32 packages/apps/TvSettings/QuickSettings/

其他build framework的啦，build kernel的啦都有，只是我的人生不需要去回想那些就是了...

Android Memory Inspection

• 我弄了一個script可以用來profile process的各種記憶體狀態，並且紀錄hprof file
  [Github Link]
• 
  使用mat工具可以觀看hprof file. 需用hprof-conf轉檔。
• 
  
• google IO 示範memory analyzer的sample
• https://www.youtube.com/watch?v=_CruQY55HOk

中文blog翻譯

http://blog.csdn.net/qian_xiao_lj/article/details/50801925

• 
  
• 一些名詞解釋。
  
  Shallow heap和Retained heap。Shallow heap表示對象本身所占內存大小，一個內存大小100bytes的對象Shallow heap就是100bytes。Retained heap表示通過回收這一個對象總共能回收的內存，比方說一個100bytes的對象還直接或者間接地持有了另外3個100bytes的對象引用，回收這個對象的時候如果另外3個對象沒有其他引用也能被回收掉的時候，Retained heap就是400bytes。
  
  VSS - Virtual Set Size 虛擬耗用內存（包含共享庫占用的內存）
• RSS - Resident Set Size 實際使用物理內存（包含共享庫占用的內存）
• PSS - Proportional Set Size 實際使用的物理內存（比例分配共享庫占用的內存）
• USS - Unique Set Size 進程獨自占用的物理內存（不包含共享庫占用的內存）

一般來說內存占用大小有如下規律：VSS >= RSS >= PSS >= USS

Overview

The aim of this post is to provide information that will assist in interpreting memory reports from various tools so the true memory usage for Linux processes and the system can be determined.

Android has a tool called procrank (/system/xbin/procrank), which lists out the memory usage of Linux processes in order from highest to lowest usage. The sizes reported per process are VSS, RSS, PSS, and USS.

For the sake of simplicity in this description, memory will be expressed in terms of pages, rather than bytes. Linux systems like ours manage memory in 4096 byte pages at the lowest level.

VSS (reported as VSZ from ps) is . This size also includes memory that may not be resident in RAM like mallocs that have been allocated but not written to. VSS is of very little use for determing real memory usage of a process.

RSS is the. RSS can be misleading, because it reports the total all of the shared libraries that the process uses, even though a shared library is only loaded into memory once regardless of how many processes use it. RSS is not an accurate representation of the memory usage for a single process.

PSS , i.e. if three processes all use a shared library that has 30 pages, that library will only contribute 10 pages to the PSS that is reported for each of the three processes. PSS is a very useful number because when the PSS for all processes in the system are summed together, that is a good representation for the total memory usage in the system. When a process is killed, the shared libraries that contributed to its PSS will be proportionally distributed to the PSS totals for the remaining processes still using that library. In this way PSS can be slightly misleading, because when a process is killed, PSS does not accurately represent the memory returned to the overall system.

USS is . USS is an extremely useful number because it indicates the true incremental cost of running a particular process. When a process is killed, the USS is the total memory that is actually returned to the system. USS is the best number to watch when initially suspicious of memory leaks in a process.

For systems that have Python available, there is also a nice tool called smem that will report memory statistics including all of these categories.

swift 語言一些心得

這其實是1.0的心得，現在已經出到3.0了 其實這份也不用看細節了，直接看一些特別的語法特性就好了。

https://drive.google.com/file/d/0B6BL3rjr8LORb3ptY2g0a3ZpQVE/view?usp=sharing

CAMediaTimingFunction

Basically I want to make a dots animation view like this.


So I look into the CAMediaTimingFunction and it's a bezier function.

https://en.wikipedia.org/wiki/B%C3%A9zier_curve

There is a website that you can view your bezier curve in CAMediaTimingFunction

http://netcetera.org/camtf-playground.html

So I just create 3 dots with different CAMediaTimingFunction that changes alpha like this.




But eventually it don't really have to use CAMediaTimingFunction to approach this. Just create animation groups will be more precise.

iCloud Key-value storage development

The goal is store your app preference to iCloud for backup. That only involved to the key-value storage of iCould.

To use iCloud in an app. You need to configure the in the apple develop like this

Where you can store different type of data like .

Which only key-value storage don't need a container, others are needed a container to store the data in. Look into the apple document would be more helpful.

iCloudDesignGuide

Once you setup the iCould, you could be able to access NSUbiquitous store

    [NSUbiquitousKeyValueStore defaultStore]

and store your key-value data like UserDefault to the ubiquitouskeyvaluestore

  

    [[NSUbiquitousKeyValueStore defaultStore] setObject:obj forKey:key];

    [[NSUbiquitousKeyValueStore defaultStore] synchronize];

and you register a notification of the key

NSUbiquitousKeyValueStoreDidChangeExternallyNotification

And analysis the notification you will get the key-value backup from your iCould account.

    NSDictionary *userInfo = [notificationObject userInfo];

    NSInteger reasonForChangeValue = [[userInfo objectForKey:NSUbiquitousKeyValueStoreChangeReasonKey] integerValue];

        

     if (reasonForChangeValue == NSUbiquitousKeyValueStoreInitialSyncChange)

    {

          // your merge code here

    }

That's it.
You can take look into the
https://github.com/hydrated/iOS-Utilities/tree/master/iCouldSync 


Furthermore you want to make more functionality,
NSUbiquitousKeyValueStoreChangeReasonKey has 4 values can be listening notification. like account change , violate limitation storage.

UITableView and ListView in Polymorphism

UITableView and ListView都有dequeue cell and item的作用，一些心得，

可使用autolayout

_mainTableView.estimatedRowHeight =  44;

_mainTableView.rowHeight = UITableViewAutomaticDimension;

一個tableview的cell會保持自己的instance，你無法取代新的instance該裡面的cell，尤其UIWebView塞進cell更是難以計算高度，用之前千萬三思。


Android ListView更為複雜，一個getView會call三次左右才決定一個item，我設計了一個polymorphism 的架構來控制items


interface {

    public abstract int getType();

    public abstract void setViewHolder(View convertView);

    public abstract void inflateViewHolder(View convertView);

    public abstract View getTemplateView(ViewGroup parent);

}

@Override

public int getItemViewType(int position) {

    return ((interface) getItem(position)).getType();

}

@Override

public View getView(int position, View convertView, ViewGroup parent) {

    final int type = getItemViewType(position);

    if (convertView != null) {

        mList.get(position).inflateViewHolder(convertView);

    } else {

        convertView = mList.get(position).getTemplateView(parent);

        mList.get(position).setViewHolder(convertView);

    }

return convertView;

}

mList必須保持所有子item的data instance，注意只有data不存view，那麼getItemViewType內的gettype決定你item type，之後需傳入convertView給該子item作render。以及null的情況下new一個新的instance view給convertView。

目前使用起來沒出問題，listview item開兩萬個也無礙，我想應該是可行的。