学习目的
- 在android studio 3.0的 基础上同时生成多个so包，cmake的编写
- 开发安卓程序进行ndk开发 自动生成 增量包 patch
- 生成 合并的so 库进行patch文件的合并
- 解决 引用三方so库 方法报红 的问题
演示
增量更新就是，app自动更新的时候不会把整个新版本的apk包下载下来 进行覆盖安装，而是将一个 新版本和老版本 进行比较 生成一个 patch包，把patch包下载下来 和当前版本进行合并。然后在进行覆盖安装。
为了演示 我们先创建一个项目。将相同包名 的两个不同版本的apk 放到assert 目录下。然后将这两个包拷贝到 指定文件夹下。在生成一个patch 包。
下过如下
就这样 我们把assets目录下的两个apk。拷贝到了文件夹下，并生成了一个patch包。
这个patch包就是 app-new 和app-old 中的区别的部分。
然后我们可以吧这个patch包放到服务器。这样用户在使用app的时候 只用下载patch包，然后和当前app进行合并。就可以覆盖安装了。
下面我们在看一下 合并包的效果。我们还在该手机上。直接安装目录下的app-old.apk,然后点击合成按钮。会在线程里合成一个新的apk。然后进行覆盖安装。
我们看到了。我们的app 进行了覆盖安装。
- 注意：
- 这个叫oldtonew.apk 并不是这个文件夹下的 patch 和 app-old.apk合成的。因为用户使用的时候并不一定手机里会有 老版本的apk安装包。
- 高于7.0 版本 的系统 覆盖安装会有一些坑。可以查看这篇文章。 https://juejin.im/post/5a7163866fb9a01ca3259710
包的拆分
需要的工具
-
bzip2 : http://www.bzip.org/downloads.html 下载放一边
-
bsdiff : https://github.com/mendsley/bsdiff clone 一波
如何生成多个so文件（先了解一下）
如果androidstudio 3.0 上还不会使用ndk 开发 请看这篇文章 https://juejin.im/post/5a72bc53f265da3e364197c0 看完这个基本了解个大概了。androidstudio 已经不用 编写 .mk 文件了。
我们想创建一个项目。生成一个 diff.so 和patch.so 两个包。怎么做呢
这样一目了然了。再看看 这三个CmakeList.txt 怎么写的呢
diff的 CmakeList.txt
cmake_minimum_required(VERSION 3.4.1)
add_library( # Specifies the name of the library.
# 这里是你so的名字。刚才在 MainActivity里面要引用的
diff
# Sets the library as a shared library.
SHARED
# Provides a relative path to your source file(s).
#这里是刚才 创建的c++ 代码的名字
Diff.c
)
find_library( # Sets the name of the path variable.
log-lib
# Specifies the name of the NDK library that
# you want CMake to locate.
log )
target_link_libraries( # Specifies the target library.
diff
# Links the target library to the log library
# included in the NDK.
${log-lib} )
patch 的 CmakeList.txt
cmake_minimum_required(VERSION 3.4.1)
# include_directories(
# "${PROJECT_SOURCE_DIR}/..")
add_library( # Specifies the name of the library.
# 这里是你so的名字。刚才在 MainActivity里面要引用的
patch
# Sets the library as a shared library.
SHARED
# Provides a relative path to your source file(s).
#这里是刚才 创建的c++ 代码的名字
PatchUtils.c
)
find_library( # Sets the name of the path variable.
log-lib
# Specifies the name of the NDK library that
# you want CMake to locate.
log )
target_link_libraries( # Specifies the target library.
patch
# Links the target library to the log library
# included in the NDK.
${log-lib} )
被gradle 引用的 CmakeList.txt
cmake_minimum_required(VERSION 3.4.1)
#添加子目录,将会调用子目录中的CMakeLists.txt
ADD_SUBDIRECTORY(diff)
ADD_SUBDIRECTORY(patch)
复制代码在看一下 我们的gradle文件
可以看到 我们生成了 diff.so 和 patch.so 文件夹。 现在如果我们在创建什么项目 直接使用这两个so文件。就可以进行 拆分 和合并了。
拆分，合并代码编写。
拆分 和 合并 要使用 刚才 下载的 bsdiff 文件。bsdiff 文件又依赖于 bzip2 的库
因为拆分 与 合并 基本类似。就展示 拆分的代码了。我会把所有代码放到github上去。因为这两个库在使用的时候有特别多的坑。
先看一下目录结构吧
在diff目录下新建一些 拆分需要的代码。
注意 如果你不知道 sven_com_apkpatchserver_DiffUtils.h 这个文件是如何生成的，你需要先看这篇文章，不然下面的看不懂了https://juejin.im/post/5a72bc53f265da3e364197c0
新建一个 DiffUtils 的工具类
public class DiffUtils {
static {//导入libhello.so 文件。 这里面只写hello就可以
System.loadLibrary("diff");
}
/**
* native方法 比较路径为oldPath的apk与newPath的apk之间差异，并生成patch包，存储于patchPath
* <p>
* 返回：0，说明操作成功
*
* @param oldApkPath 示例:/sdcard/old.apk
* @param newApkPath 示例:/sdcard/new.apk
* @param patchPath 示例:/sdcard/xx.patch
* @return
*/
public static native String genDiff(String oldApkPath, String newApkPath, String patchPath);
}
复制代码生成 他的 native 头文件 sven_com_apkpatchserver_DiffUtils.h
/* DO NOT EDIT THIS FILE - it is machine generated */
#include <jni.h>
/* Header for class sven_com_apkpatchserver_DiffUtils */
#ifndef _Included_sven_com_apkpatchserver_DiffUtils
#define _Included_sven_com_apkpatchserver_DiffUtils
#ifdef __cplusplus
extern "C" {
#endif
/*
* Class: sven_com_apkpatchserver_DiffUtils
* Method: genDiff
* Signature: (Ljava/lang/String;Ljava/lang/String;Ljava/lang/String;)I
*/
JNIEXPORT jstring JNICALL Java_sven_com_apkpatchserver_DiffUtils_genDiff
(JNIEnv *, jclass, jstring, jstring, jstring);
#ifdef __cplusplus
}
#endif
#endif
复制代码还记得 上文提到的github 中 有一个 bsdiff.c 的文件么 全部拷贝 放到diff文件夹下，这里更名为 diff.c 在这个代码的基础上进行编写 代码如下，特别多。请小心。下面全是c 语言的代码了。如果看不懂可以先从jni学起。
//
// Created by mypro on 2018/1/28.
//
/*-
* Copyright 2003-2005 Colin Percival
* All rights reserved
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted providing that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
* IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "sven_com_apkpatchserver_DiffUtils.h"
#include "../bzip2/bzlib.c"
#include "../bzip2/crctable.c"
#include "../bzip2/compress.c"
#include "../bzip2/decompress.c"
#include "../bzip2/randtable.c"
#include "../bzip2/blocksort.c"
#include "../bzip2/huffman.c"
#include <sys/types.h>
#include <err.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#define MIN(x, y) (((x)<(y)) ? (x) : (y))
static void split(off_t *I, off_t *V, off_t start, off_t len, off_t h) {
off_t i, j, k, x, tmp, jj, kk;
if (len < 16) {
for (k = start; k < start + len; k += j) {
j = 1;
x = V[I[k] + h];
for (i = 1; k + i < start + len; i++) {
if (V[I[k + i] + h] < x) {
x = V[I[k + i] + h];
j = 0;
};
if (V[I[k + i] + h] == x) {
tmp = I[k + j];
I[k + j] = I[k + i];
I[k + i] = tmp;
j++;
};
};
for (i = 0; i < j; i++) V[I[k + i]] = k + j - 1;
if (j == 1) I[k] = -1;
};
return;
};
x = V[I[start + len / 2] + h];
jj = 0;
kk = 0;
for (i = start; i < start + len; i++) {
if (V[I[i] + h] < x) jj++;
if (V[I[i] + h] == x) kk++;
};
jj += start;
kk += jj;
i = start;
j = 0;
k = 0;
while (i < jj) {
if (V[I[i] + h] < x) {
i++;
} else if (V[I[i] + h] == x) {
tmp = I[i];
I[i] = I[jj + j];
I[jj + j] = tmp;
j++;
} else {
tmp = I[i];
I[i] = I[kk + k];
I[kk + k] = tmp;
k++;
};
};
while (jj + j < kk) {
if (V[I[jj + j] + h] == x) {
j++;
} else {
tmp = I[jj + j];
I[jj + j] = I[kk + k];
I[kk + k] = tmp;
k++;
};
};
if (jj > start) split(I, V, start, jj - start, h);
for (i = 0; i < kk - jj; i++) V[I[jj + i]] = kk - 1;
if (jj == kk - 1) I[jj] = -1;
if (start + len > kk) split(I, V, kk, start + len - kk, h);
}
static void qsufsort(off_t *I, off_t *V, u_char *old, off_t oldsize) {
off_t buckets[256];
off_t i, h, len;
for (i = 0; i < 256; i++) buckets[i] = 0;
for (i = 0; i < oldsize; i++) buckets[old[i]]++;
for (i = 1; i < 256; i++) buckets[i] += buckets[i - 1];
for (i = 255; i > 0; i--) buckets[i] = buckets[i - 1];
buckets[0] = 0;
for (i = 0; i < oldsize; i++) I[++buckets[old[i]]] = i;
I[0] = oldsize;
for (i = 0; i < oldsize; i++) V[i] = buckets[old[i]];
V[oldsize] = 0;
for (i = 1; i < 256; i++) if (buckets[i] == buckets[i - 1] + 1) I[buckets[i]] = -1;
I[0] = -1;
for (h = 1; I[0] != -(oldsize + 1); h += h) {
len = 0;
for (i = 0; i < oldsize + 1;) {
if (I[i] < 0) {
len -= I[i];
i -= I[i];
} else {
if (len) I[i - len] = -len;
len = V[I[i]] + 1 - i;
split(I, V, i, len, h);
i += len;
len = 0;
};
};
if (len) I[i - len] = -len;
};
for (i = 0; i < oldsize + 1; i++) I[V[i]] = i;
}
static off_t matchlen(u_char *old, off_t oldsize, u_char *new, off_t newsize) {
off_t i;
for (i = 0; (i < oldsize) && (i < newsize); i++)
if (old[i] != new[i]) break;
return i;
}
static off_t search(off_t *I, u_char *old, off_t oldsize,
u_char *new, off_t newsize, off_t st, off_t en, off_t *pos) {
off_t x, y;
if (en - st < 2) {
x = matchlen(old + I[st], oldsize - I[st], new, newsize);
y = matchlen(old + I[en], oldsize - I[en], new, newsize);
if (x > y) {
*pos = I[st];
return x;
} else {
*pos = I[en];
return y;
}
};
x = st + (en - st) / 2;
if (memcmp(old + I[x], new, MIN(oldsize - I[x], newsize)) < 0) {
return search(I, old, oldsize, new, newsize, x, en, pos);
} else {
return search(I, old, oldsize, new, newsize, st, x, pos);
};
}
static void offtout(off_t x, u_char *buf) {
off_t y;
if (x < 0) y = -x; else y = x;
buf[0] = y % 256;
y -= buf[0];
y = y / 256;
buf[1] = y % 256;
y -= buf[1];
y = y / 256;
buf[2] = y % 256;
y -= buf[2];
y = y / 256;
buf[3] = y % 256;
y -= buf[3];
y = y / 256;
buf[4] = y % 256;
y -= buf[4];
y = y / 256;
buf[5] = y % 256;
y -= buf[5];
y = y / 256;
buf[6] = y % 256;
y -= buf[6];
y = y / 256;
buf[7] = y % 256;
if (x < 0) buf[7] |= 0x80;
}
int bsdiff_main(int argc, char *argv[]) {
int fd;
u_char *old, *new;
off_t oldsize, newsize;
off_t *I, *V;
off_t scan, pos, len;
off_t lastscan, lastpos, lastoffset;
off_t oldscore, scsc;
off_t s, Sf, lenf, Sb, lenb;
off_t overlap, Ss, lens;
off_t i;
off_t dblen, eblen;
u_char *db, *eb;
u_char buf[8];
u_char header[32];
FILE *pf;
BZFILE *pfbz2;
int bz2err;
if (argc != 4) errx(1, "usage: %s oldfile newfile patchfile\n", argv[0]);
/* Allocate oldsize+1 bytes instead of oldsize bytes to ensure
that we never try to malloc(0) and get a NULL pointer */
if (((fd = open(argv[1], O_RDONLY, 0)) < 0) ||
((oldsize = lseek(fd, 0, SEEK_END)) == -1) ||
((old = malloc(oldsize + 1)) == NULL) ||
(lseek(fd, 0, SEEK_SET) != 0) ||
(read(fd, old, oldsize) != oldsize) ||
(close(fd) == -1))
err(1, "%s", argv[1]);
if (((I = malloc((oldsize + 1) * sizeof(off_t))) == NULL) ||
((V = malloc((oldsize + 1) * sizeof(off_t))) == NULL))
err(1, NULL);
qsufsort(I, V, old, oldsize);
free(V);
/* Allocate newsize+1 bytes instead of newsize bytes to ensure
that we never try to malloc(0) and get a NULL pointer */
if (((fd = open(argv[2], O_RDONLY, 0)) < 0) ||
((newsize = lseek(fd, 0, SEEK_END)) == -1) ||
((new = malloc(newsize + 1)) == NULL) ||
(lseek(fd, 0, SEEK_SET) != 0) ||
(read(fd, new, newsize) != newsize) ||
(close(fd) == -1))
err(1, "%s", argv[2]);
if (((db = malloc(newsize + 1)) == NULL) ||
((eb = malloc(newsize + 1)) == NULL))
err(1, NULL);
dblen = 0;
eblen = 0;
/* Create the patch file */
if ((pf = fopen(argv[3], "w")) == NULL)
err(1, "%s", argv[3]);
/* Header is
0 8 "BSDIFF40"
8 8 length of bzip2ed ctrl block
16 8 length of bzip2ed diff block
24 8 length of new file */
/* File is
0 32 Header
32 ?? Bzip2ed ctrl block
?? ?? Bzip2ed diff block
?? ?? Bzip2ed extra block */
memcpy(header, "BSDIFF40", 8);
offtout(0, header + 8);
offtout(0, header + 16);
offtout(newsize, header + 24);
if (fwrite(header, 32, 1, pf) != 1)
err(1, "fwrite(%s)", argv[3]);
/* Compute the differences, writing ctrl as we go */
if ((pfbz2 = BZ2_bzWriteOpen(&bz2err, pf, 9, 0, 0)) == NULL)
errx(1, "BZ2_bzWriteOpen, bz2err = %d", bz2err);
scan = 0;
len = 0;
lastscan = 0;
lastpos = 0;
lastoffset = 0;
while (scan < newsize) {
oldscore = 0;
for (scsc = scan += len; scan < newsize; scan++) {
len = search(I, old, oldsize, new + scan, newsize - scan,
0, oldsize, &pos);
for (; scsc < scan + len; scsc++)
if ((scsc + lastoffset < oldsize) &&
(old[scsc + lastoffset] == new[scsc]))
oldscore++;
if (((len == oldscore) && (len != 0)) ||
(len > oldscore + 8))
break;
if ((scan + lastoffset < oldsize) &&
(old[scan + lastoffset] == new[scan]))
oldscore--;
};
if ((len != oldscore) || (scan == newsize)) {
s = 0;
Sf = 0;
lenf = 0;
for (i = 0; (lastscan + i < scan) && (lastpos + i < oldsize);) {
if (old[lastpos + i] == new[lastscan + i]) s++;
i++;
if (s * 2 - i > Sf * 2 - lenf) {
Sf = s;
lenf = i;
};
};
lenb = 0;
if (scan < newsize) {
s = 0;
Sb = 0;
for (i = 1; (scan >= lastscan + i) && (pos >= i); i++) {
if (old[pos - i] == new[scan - i]) s++;
if (s * 2 - i > Sb * 2 - lenb) {
Sb = s;
lenb = i;
};
};
};
if (lastscan + lenf > scan - lenb) {
overlap = (lastscan + lenf) - (scan - lenb);
s = 0;
Ss = 0;
lens = 0;
for (i = 0; i < overlap; i++) {
if (new[lastscan + lenf - overlap + i] ==
old[lastpos + lenf - overlap + i])
s++;
if (new[scan - lenb + i] ==
old[pos - lenb + i])
s--;
if (s > Ss) {
Ss = s;
lens = i + 1;
};
};
lenf += lens - overlap;
lenb -= lens;
};
for (i = 0; i < lenf; i++)
db[dblen + i] = new[lastscan + i] - old[lastpos + i];
for (i = 0; i < (scan - lenb) - (lastscan + lenf); i++)
eb[eblen + i] = new[lastscan + lenf + i];
dblen += lenf;
eblen += (scan - lenb) - (lastscan + lenf);
offtout(lenf, buf);
BZ2_bzWrite(&bz2err, pfbz2, buf, 8);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWrite, bz2err = %d", bz2err);
offtout((scan - lenb) - (lastscan + lenf), buf);
BZ2_bzWrite(&bz2err, pfbz2, buf, 8);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWrite, bz2err = %d", bz2err);
offtout((pos - lenb) - (lastpos + lenf), buf);
BZ2_bzWrite(&bz2err, pfbz2, buf, 8);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWrite, bz2err = %d", bz2err);
lastscan = scan - lenb;
lastpos = pos - lenb;
lastoffset = pos - scan;
};
};
BZ2_bzWriteClose(&bz2err, pfbz2, 0, NULL, NULL);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWriteClose, bz2err = %d", bz2err);
/* Compute size of compressed ctrl data */
if ((len = ftello(pf)) == -1)
err(1, "ftello");
offtout(len - 32, header + 8);
/* Write compressed diff data */
if ((pfbz2 = BZ2_bzWriteOpen(&bz2err, pf, 9, 0, 0)) == NULL)
errx(1, "BZ2_bzWriteOpen, bz2err = %d", bz2err);
BZ2_bzWrite(&bz2err, pfbz2, db, dblen);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWrite, bz2err = %d", bz2err);
BZ2_bzWriteClose(&bz2err, pfbz2, 0, NULL, NULL);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWriteClose, bz2err = %d", bz2err);
/* Compute size of compressed diff data */
if ((newsize = ftello(pf)) == -1)
err(1, "ftello");
offtout(newsize - len, header + 16);
/* Write compressed extra data */
if ((pfbz2 = BZ2_bzWriteOpen(&bz2err, pf, 9, 0, 0)) == NULL)
errx(1, "BZ2_bzWriteOpen, bz2err = %d", bz2err);
BZ2_bzWrite(&bz2err, pfbz2, eb, eblen);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWrite, bz2err = %d", bz2err);
BZ2_bzWriteClose(&bz2err, pfbz2, 0, NULL, NULL);
if (bz2err != BZ_OK)
errx(1, "BZ2_bzWriteClose, bz2err = %d", bz2err);
/* Seek to the beginning, write the header, and close the file */
if (fseeko(pf, 0, SEEK_SET))
err(1, "fseeko");
if (fwrite(header, 32, 1, pf) != 1)
err(1, "fwrite(%s)", argv[3]);
if (fclose(pf))
err(1, "fclose");
/* Free the memory we used */
free(db);
free(eb);
free(I);
free(old);
free(new);
return 0;
}
//JNI 调用
JNIEXPORT jstring JNICALL Java_sven_com_apkpatchserver_DiffUtils_genDiff
(JNIEnv *env, jclass jcls, jstring oldfile_jstr, jstring newfile_jstr,
jstring patchfile_jstr) {
int argc = 4;
char *oldfile = (char *) (*env)->GetStringUTFChars(env, oldfile_jstr, NULL);
char *newfile = (char *) (*env)->GetStringUTFChars(env, newfile_jstr, NULL);
char *patchfile = (char *) (*env)->GetStringUTFChars(env, patchfile_jstr, NULL);
//参数（第一个参数无效）
char *argv[4];
argv[0] = "bsdiff";
argv[1] = oldfile;
argv[2] = newfile;
argv[3] = patchfile;
//这个方法是和新方法，他会根据 old.apk new.apk 进行拆分，然后生成 patch 放到指定文件夹下
bsdiff_main(argc, argv);
(*env)->ReleaseStringUTFChars(env, oldfile_jstr, oldfile);
(*env)->ReleaseStringUTFChars(env, newfile_jstr, newfile);
(*env)->ReleaseStringUTFChars(env, patchfile_jstr, patchfile);
return (*env)->NewStringUTF(env, "hello hello");
}
//void main(int argc, char **argv) {
//
//}
复制代码到这里 拆分就完成了。 只需要在java代码里执行如下代码就行
findViewById(R.id.diff).setOnClickListener(new View.OnClickListener() {
@Override
public void onClick(View v) {
String s = DiffUtils.genDiff(Constants.OLD_APK_PATH, Constants.NEW_APK_PATH, Constants.PATCH_FILE);
}
});
Constants.OLD_APK_PATH： old.apk 路径
Constants.NEW_APK_PATH： new.apk 路径
Constants.PATCH_FILE： 生成的patch 路径
复制代码到这里 拆分 就都讲完了。合并 的代码 大同小异这里就不赘述了。我会把代码放到github上去。
使用patch.so 进行apk 与patch的合并。
新建一个项目，假设这个项目就是我们公司的app了。把刚才生成好的 patch.so文件 拿过来 复制到这里
然后我们开始编写 PatchUtils的代码了
注意
- 我们如果想完美的使用so文件。我们带么的 名字 和路径 都要和 生成so 项目的代码的路径一致
比如这样
引用成功的话。基本就完成了。 我们只需要在代码里 简单调用就可以了
class ApkUpdateTask extends AsyncTask<Void, Void, Boolean> {
@Override
protected Boolean doInBackground(Void... params) {
try {
//1.下载差分包
String oldfile = ApkUtils.getSourceApkPath(MainActivity.this, getPackageName());
//2.合并得到最新版本的APK文件
String newfile = Constants.OLD_2_NEW_APK;
String patchfile = Constants.PATCH_FILE;
PatchUtils.patch(oldfile, newfile, patchfile);
} catch (Exception e) {
e.printStackTrace();
return false;
}
return true;
}
@Override
protected void onPostExecute(Boolean result) {
super.onPostExecute(result);
//3.安装
if (result) {
Toast.makeText(MainActivity.this, "您正在进行无流量更新", Toast.LENGTH_SHORT).show();
ApkUtils.installApk(MainActivity.this, Constants.OLD_2_NEW_APK);
}
}
}
复制代码到这里 我们就完成了apk 的 增量更新
覆盖安装的一些坑
- 增量更新，在手机上安装的 apk。一定要是 release key 打包的，并且 v1 v2 对勾都要选中
- 覆盖安装 要适配7.0 https://juejin.im/post/5a72bc53f265da3e364197c0
- 覆盖安装不要忘记 这句代码
intent.setFlags(Intent.FLAG_ACTIVITY_NEW_TASK);
复制代码- 如果怎么都不行，可以查看 你的apk包是否有读写权限 可以这样
String[] command = {"chmod", "777", Constants.OLD_2_NEW_APK};
ProcessBuilder builder = new ProcessBuilder(command);
try {
builder.start();
} catch (IOException e) {
e.printStackTrace();
}
复制代码不懂得可以查看代码，最直观 https://github.com/wanghao200906/NdkBsDiffPractise