m*p
发帖数: 1331 | 1 在看一个hadoop 的 codebase，发现有如下实现， 为啥捕捉到异常之后再把逻辑执行
一遍？？是hadoop特有的东西么？
private static double[] parseStringToVector(String line, double[]
row) {
try {
StringTokenizer tokenizer = new StringTokenizer(line
, ",");
int size = tokenizer.countTokens();
if (row == null)
row = new double[size];
int i = 0;
while (tokenizer.hasMoreTokens... 阅读全帖 |
|
c****e
发帖数: 1453 | 2 multiple inheritance usually is not encouraged. Template is used everywhere.
Just take a look at boost, almost all reasonably large c++ codebases I
worked on heavily use that. |
|
b**********5
发帖数: 7881 | 3 来自主题: Programming版 - 我老版问我 叫你upgrade个codebase，(from VS 2005 to VS2010), 难道要把thirdparty
libraries也要recompile on VS2010？ 很不make sense啊。。。
这问题， 叫我怎么回答。。。 |
|
F*M
发帖数: 104 | 4 我听到的一个说法是FB在增强PHP的语言特性，大量的codebase是PHP写的。所以只能硬
着头皮PHP下去了。 |
|
p*u
发帖数: 2454 | 5
lol for many firms like service providers, core of their codebase has been
running for a couple of decades... |
|
g*****g
发帖数: 34805 | 6 你也知道动不动号称大型系统是恬不知耻？我们的系统没有Linux大，但Linux同样不是
一个人写的，是整个社区十几年无数人堆出来的。Linux最初是Linus一个人写的的时候
比我们的codebase小多了。
莫非现在又要自比Linus了？ |
|
V*********r
发帖数: 666 | 7 ORM还是用django自己那个么？Paul说你们对django做了重量级定制。早知如此一开始
上Flask甚至Java系会不会灵活很多。不知道你们现在codebase多大了，好不好掉头。 |
|
m******t
发帖数: 635 | 8 恩，他的主要观点就是鼓吹一个小规模的SOA架构。
不过他这句话恶心到我了：
Let’s apply this to our software systems. Instead of building a monolithic
100 000 line codebase, build 100 small services, each 1000 lines long. Fred
George, (the inventor of programmer anarchy) one of the biggest proponents
of this approach, calls these small programs micro-services.
难道JS弱到连1000行都是大程序？
haven'
would
dependency
JS |
|
g*****g
发帖数: 34805 | 9 I don't know if this guy really has architected big system before.
"Let’s apply this to our software systems. Instead of building a monolithic
100 000 line codebase, build 100 small services, each 1000 lines long. Fred
George, (the inventor of programmer anarchy) one of the biggest proponents
of this approach, calls these small programs micro-services."
While I am a proponent of SOA, I wouldn't do micro-services. If your single
service has 1m latency, a call that goes through 100 services has 10... 阅读全帖 |
|
P********l
发帖数: 452 | 10 － If your codebase grows bigger, javascript is not the friendliest language
for your unit tests
True. That's why TypeScript.
- When node leaks memory you don't know why unless you can do this:
Profiling Node.js. Even after intensive profiling, you don't really know
what exactly is wrong. You could only gather ideas. your debugging still
largely remains as guess work.
I don't think it is a valid point. Maybe they went to fast without paying
attention to things like this. Using Java will be better... 阅读全帖 |
|
g*****g
发帖数: 34805 | 11 尼玛收购后还是那个codebase，积累千万用户就是个时间问题，校友录当年就做出来了
，后来都一亿了。怎么着，我说参与开发不行吗？比起你改人现成的模板吐血，不是天
上地下。
求求你下回不要见到个自动文档就出来惊诧好不好。丢人不叫创新。 |
|
n*****t
发帖数: 22014 | 12 你丫说这话脸红不脸红啊，操，真鸡巴好意思。简直打工的穷逼说上亿的 codebase 是
你丫做的，尼玛还有比你丫更不要脸的不?
也是，那种傻逼自动文档就是给你这种傻逼看的，你丫不要脸最创新 |
|
n*****t
发帖数: 22014 | 13 别不服，人 2000 年写的 chinaren 到现在还在用他的 codebase，现在有上亿的客户
，快赶上三退了 |
|
n*****t
发帖数: 22014 | 14 好，拉的屎自己吃回去了，有进步。
下次不说 chinaren 的 codebase 是你了哦，乖 |
|
g*****g
发帖数: 34805 | 15 我老从来没说过codebase是我的，参与开发而已，个人没版权。你精分了连记性都不好
了。 |
|
p*****2
发帖数: 21240 | 16
Grails is a full-stack web application framework for the JVM. With its
mature codebase backed by a thriving community, Grails is nonetheless facing
new architectural challenges. Events were introduced into Grails through
the platform-core plugin. But Events are so powerful that this feature
really belongs in the core; so from version 2.3 on, Grails applications will
have a built-in, extremely powerful yet easy to use, convention-based
Events API that looks very similar to the current implementa... 阅读全帖 |
|
h*****a
发帖数: 1718 | 17 经常是为了维护很大的legacy codebase的，尤其是VB6这样比较老的语言的code.
另外就是startup常常需要新人能快速上手，所以对语言和framework的熟悉要求高一些。
大公司一般来说语言不太会是主要的考虑。但熟悉公司正在用的或者准备用的语言和技
术一般来说也是一个plus。 |
|
V*********r
发帖数: 666 | 18
原先有2to3的转化程序，但py3进一步进化后，现在已经被证明那个程序不可行了
现在唯一可靠的方法是只维持一套同时兼容2/3的codebase
不过这是对于library writer的要求
普通用户不用太担心，一直用py2.7就行了
前几天官方已经宣布把py2的“死亡日期”推迟到2020年，到时候说不定还可能继续后推 |
|
l*********s
发帖数: 5409 | 19 和个人选择没啥关系。我做过的公司开发环境都是在unix服务器上，codebase以c++为
主，eclipse太慢没法用。 |
|
c*******0
发帖数: 5247 | 20
事实上到目前为止，改头门第的人，来自Python和Ruby圈的最多。反而是因为Go的GC特
性让很多C/C++不愿意改投门第。
商业逻辑这方面转Go的确实少，Go本来也没打算在这方面和Java争。Go大部分是做
infra的。
个习惯问题，比如Swift style和Go类似，很多objc的人转了几周也就适应了。
我老早就在这里说了，Go不可能取代Java，Go的目标也不是取代Java。Java dominate
的地位，只要Android不倒，JVM不倒，是不会被动摇的。
同意
既然说到这了，我就多说几句。
Go是一门设计理念非常清晰但是顽固的语言。和别的语言不同，Go并不希望成为一个
one language rule them all。在一开始设计的时候，Go team的人的目标很简单，就
是满足Google内部做多team大规模infra的需求。注意，这个“多team大规模”是目标
，不是期盼。因为这个目标，所以才会有Go boring的语法，少得可怜的特性，和及其
简单的上手，以及超强的可读性。
Go社区的人喜欢Go，完全不是因为什么语言牛逼的特性，也不是什么best pe... 阅读全帖 |
|
L***s
发帖数: 1148 | 21 “一个codebase通吃所有平台”还是挺吸引人的
过阵子我也拿来练练手～
也许不如native app流畅，但做startup、缺人手和时间的情况下，还是很吸引人的 |
|
m****a
发帖数: 2593 | 22 git 的命令行设计的一塌糊涂，简直就是业余的水平。
http://stevebennett.me/2012/02/24/10-things-i-hate-about-git/
. Crazy command line syntax
The command line syntax is completely arbitrary and inconsistent. Some “
shortcuts” are graced with top level commands: “git pull” is exactly
equivalent to “git fetch” followed by “git merge”. But the shortcut for
“git branch” combined with “git checkout”? “git checkout -b”.
Specifying filenames completely changes the semantics of some commands (“
git commit” ignores local, unstag... 阅读全帖 |
|
f******2
发帖数: 2455 | 23 错，flink主要codebase是java，scala那部分是两部分：用户语言binding，akka使用
因为核心是java，给广大java高手提供了广阔天地，大有作为的平台。这对于一个开源
型公司非常重要。因此社区运作好了绝对可以比spark后发先至。
我准备上去修个bug，个contributor当当。象goodbug这样的估计直接就是committer了 |
|
f********x
发帖数: 99 | 24 Batch is a special case of streaming
Posted on September 15, 2015 by Kostas Tzoumas
In recent blog posts, we introduced what we deem as requirements for systems
to classify as stream processors, and followed up with a detailed
comparison of current approaches to data streaming, including extensive
experiments comparing Apache Flink and Apache Storm.
We are not the only ones to make the point that streaming systems are
reaching a point of maturity which makes older batch systems and
architectures... 阅读全帖 |
|
x***4
发帖数: 1815 | 25 我觉得学东西好相对容易，最大的问题是不同背景的人写的scala code的范式会非常不
一样。最后整个codebase很难理解。 |
|
h******b
发帖数: 6055 | 26 The reality is that it’s really hard for most companies to staff and build
native mobile apps. This explains why most large companies outsource mobile
development. It also explains why so many companies complain about the cost
and time required to build apps, despite their generally smaller scope.
This, despite the fact that many large companies have tons of web developers
on staff! After we launched Ionic and started to get feedback, I realized
our true calling was to empower those developers t... 阅读全帖 |
|
S*********t
发帖数: 78 | 27 Is Angular 2.0 Worth It
Author: Torgeir Helgevold
Published: Sat Sep 05 2015
I have over the past months spent quite a bit of time working with Angular 2
.0 by writing sample components and going through the Angular 2.0 source
code. At this point I have decided to write down some of my thoughts around
my experiences with Angular 2.0. Keep in mind, these are just my own
personal experiences, so I am very interested in hearing from other
developers who might have totally different views on this.
S... 阅读全帖 |
|
w*********y
发帖数: 251 | 28 use case写的人不爽，接手的人喜欢。 最近接手了一个java老项目木有UT。现在每天
早上上班都要挣扎一番不想去，想换工作得了免得面对那个不可收拾的codebase. |
|
w*********y
发帖数: 251 | 29 use case写的人不爽，接手的人喜欢。 最近接手了一个java老项目木有UT。现在每天
早上上班都要挣扎一番不想去，想换工作得了免得面对那个不可收拾的codebase. |
|
x***4
发帖数: 1815 | 30 同意。我觉得scala的codebase要做得容易理解，对码农的要求很高。从成本效益来说
不是很划算。 |
|
t*********r
发帖数: 387 | 31 不光是这几点,每次上新的codebase有了静态类一般很快能上手
Python/JS要看半天才能开码 |
|
z**********3
发帖数: 11979 | 32 【 以下文字转载自 PDA 讨论区 】
发信人: zhizunbao123 (你在红楼我在西游), 信区: PDA
标 题: Google confirms next Android version won’t implement Oracle’s proprietary Java APIs
发信站: BBS 未名空间站 (Wed Dec 30 12:35:01 2015, 美东)
Google is replacing its implementation of the Java application programming
interfaces (APIs) in Android with OpenJDK, the open source version of Oracle
’s Java Development Kit (JDK). The news first came by a “mysterious
Android codebase commit” from last month submitted to Hacker News. Google
confirmed to... 阅读全帖 |
|
|
s******u
发帖数: 501 | 34 烂。OpenMP的scaling明显有问题，72核心280线程但是scaling能到50-60x就很不错了
。总而言之，OpenMP对海量线程的优化还是不行，sweet spot停留在8-32线程并行。也
许是kernel thread的模型决定了OpenMP thread的overhead太高，不像GPU那么
lightweight。MPI倒是能做的不错，但是要这么多的进程内存又不够。最大的优点是可
以直接用现有的x86代码（绝大多数已经支持MPI＋OpenMP了），不用像GPU需要重新
fork出来写CUDA，然后maintain两套codebase |
|
发帖数: 1 | 35 Which language has the brightest future in replacement of C between D, Go
and Rust? And Why?
In spite of my status and obvious bias as co-creator of D, I'll do my best
to answer candidly; I follow Go and Rust, and I also definitely know where D
's dirty laundry is. I'd encourage people with similar positions in the Rust
and Go communities to share their honest opinion as well. So here goes.
First off, C++ needs to be somewhere in the question. Whether it's to be
replaced alongside C, or be one o... 阅读全帖 |
|
x***4
发帖数: 1815 | 36 我公司来了新人，之前做java的，进来就说为什么我们的scala FP codebase里面不用
mock object。哈哈哈。 |
|
D***e
发帖数: 247 | 37 写10万行perl真的不能证明说是好程序员，当然你们是好的那另说。我也曾写过将近（
估计）十万行Matlab，perl，和各种scripting languages，我们实验室也有生物学生
python写的特好。但面试是人家问的是what's the largest codebase have you ever
worked on, and how many lines of code did you contribute to it？当系统变得大
而复杂的时候，就不是你聪明不聪明的问题了。Google招过不止一个两个生物PhD，但
肯定不是靠他们做系统的设计和实现的。 |
|
D***e
发帖数: 247 | 38 写10万行perl真的不能证明说是好程序员，当然你们是好的那另说。我也曾写过将近（
估计）十万行Matlab，perl，和各种scripting languages，我们实验室也有生物学生
python写的特好。但面试是人家问的是what's the largest codebase have you ever
worked on, and how many lines of code did you contribute to it？当系统变得大
而复杂的时候，就不是你聪明不聪明的问题了。Google招过不止一个两个生物PhD，但
肯定不是靠他们做系统的设计和实现的。 |
|
j****c
发帖数: 19908 | 39 【 以下文字转载自 Quant 讨论区 】
发信人: jjjstc (买买提就是一个垃圾处理站), 信区: Quant
标 题: 猎头公司发给我的关于高盛五个quant职位的email
发信站: BBS 未名空间站 (Tue Feb 15 14:09:32 2011, 美东)
两星期前发的，估计这人喜欢没事在各大学校物理系网站瞎逛，看到要毕业的就发
email，我第一次直接拖到垃圾箱里了，过了五天又发了一次。我对这个行业没兴趣，
今天刚知道这个版，转贴到这里，不知道这些信息是否有用
From: Clint Tankersley [mailto:c***[email protected]]
Sent: Monday, January 31, 2011 9:16 AM
To: 'j****[email protected]'
Subject: Clint / Ignited
jjjstc:
We’ve not spoken, so my name won’t be familiar. I visited your mitbbs Web
page, however, so I know you’re... 阅读全帖 |
|
j****c
发帖数: 19908 | 40 两星期前发的，估计这人喜欢没事在各大学校物理系网站瞎逛，看到要毕业的就发
email，我第一次直接拖到垃圾箱里了，过了五天又发了一次。我对这个行业没兴趣，
今天刚知道这个版，转贴到这里，不知道这些信息是否有用
From: Clint Tankersley [mailto:c***[email protected]]
Sent: Monday, January 31, 2011 9:16 AM
To: 'j****[email protected]'
Subject: Clint / Ignited
jjjstc:
We’ve not spoken, so my name won’t be familiar. I visited your mitbbs Web
page, however, so I know you’re a physics Ph.D. student there. I’m a
headhunter with 12 years of experience recruiting technologists, technology/
project managers, busines... 阅读全帖 |
|
b******n
发帖数: 851 | 41 up. "barbyuan" can't forward no posts anymore. 错误的参数。 u nerds need
to do a "grep" on the entire codebase for "错误的参数", and see why i get
this message! |
|
j**4
发帖数: 10425 | 42 i couldn't think of anything. I have been able to use "barbyuan" to forward
posts for a while. Then after i forwarded a bunch of posts in severeal
weeks, it stopped working both at work and at home.
is there a forwarding limit? or do a grep on the codebase, and see what
leads to the error message |
|
z*******n
发帖数: 1034 | 43 New Android Compilers for More Performance by Abel Avram on Dec 04, 2014
Eric Lafortune, Technical Director of Saikoa, maker of DexGuard, has
discovered new tools in the SDK for Android 5.0 Revision 1 (API 21.1),
namely two compilers called Jack and Jill. These tools are meant to
streamline the compilation process for Android.
Google introduced the Android RunTime (ART) as an experimental utility in
KitKat, and made it the default runtime for Lollipop, replacing Dalvik. ART
takes DEX bytecode an... 阅读全帖 |
|
C******g
发帖数: 2930 | 44 不能用url的链接，要用下面那个embed里的，然后找src=后面的那个地址
|
|
s***a
发帖数: 4921 | 46 什么是malware?
HTML:Codebase-B [Expl] |
|
