Where Wonder Begins

I’ve been implementing a email dispatching and archiving program recently. To avoid reinventing the wheel, I simply did several searches on Google, one open source software called MailArchiva can get the archiving job done perfectly. It is built to support millions of emails and thousands of mailboxes. So without doubt its performance can easily meet my requirement – simply listen to one single mailbox that won’t have more than 1000 emails a day.

So I just download the open source edition of MailArchiva (They sell the enterprise version which offers more advanced features like WebServices), upload it to my local server, run the install, then all are settled down. It runs smoothly and I can see the emails won’t stay more than 3 seconds within the inbox.

However, this software doesn’t ship with a configuration to let me modify which folder of the IMAP account the application listens to. Before the archiving, there should be another program written by myself to parse the incoming messages and dispatch accordingly. But now it turns out my program don’t even get to read the message before it is archived.

I’ve considered quite a few approaches to solve this problem, such as giving up this MailArchiva and implement a archiving program myself, or move the desired message to other folders on the server side via Exchange rules.

And I finally get things done by modifying one class of MailArchiva which is in charge of connecting to IMAP, changing default folder ‘INBOX’ to my own specified one. It’s my first time to modify .class file directly. There’re some existing tools outside, but I chose to edit it with UltraEdit as Vim doesn’t work correctly somehow. (The file modified by Vim cannot be loaded due to some ‘Extra bytes at the end of file’ problem.) Also, as I don’t want to spend the time to dig out where the length of String is, so instead of ‘ARCHIVE’, I named the folder ‘MAGIC’ as it’s of the same length with ‘INBOX’

Sometimes, things can be so straightforward. You just haven’t found the right way. That’s what I’ve learned from today’s experience. Keep DRY(Don’t Repeat Yourself) principle in the mind all the time.

Yesterday I came across a weird problem which took me a whole day to troubleshoot but without luck. Speaking simply, one YAML configuration cannot be parsed due to whatever ParserException..

Alright this morning, I finally got it solved by editing the file with UltraEdit instead of Vim. As always I’d like to dig out what exactly was going on there. Then I found out that when I insert a new line, both Vim and UltraEdit can indent automatically for me, the difference is, the latter uses spaces but the former users tabs!!

But what’s wrong with Tabs???

Then I was embarrassed to find one of the only two items of YAML’s FAQ talks about this:

Why does YAML forbid tabs?
Tabs have been outlawed since they are treated differently by different editors and tools. And since indentation is so critical to proper interpretation of YAML, this issue is just too tricky to even attempt. Indeed Guido van Rossum of Python has acknowledged that allowing TABs in Python source is a headache for many people and that were he to design Python again, he would forbid them.

So that’s it! It is a classical “religious war” within computer science.(Just like the war between Emacs & Vi). Check out this good article.

I instinctively feel that Tabs should be the one, like the article above said

Simply put, tabs is proper, and spaces are improper. Why? This may seem ridiculously simple given the de facto ball of confusion: the semantics of tabs is what indenting is about, while, using spaces to align code
is a hack.
Now, tech geekers may object this simple conclusion because they itch to drivel about different editors and so on. The alleged problem created by tabs as seen by the industry coders are caused by two things: (1) tech geeker’s sloppiness and lack of critical thinking which lead them to not understanding the semantic purposes of tab and space characters. (2) Due to the first reason, they have created and
propagated a massive none-understanding and mis-use, to the degree that many tools (e.g. vi) does not deal with tabs well and using spaces to align code has become widely practiced, so that in the end spaces seem to be actually better by popularity and seeming simplicity.
In short, this is a phenomenon of misunderstanding begetting a snowball of misunderstanding, such that it created a cultural milieu to embrace this malpractice and kick what is true or proper. Situations like this
happens a lot in unix. For one non-unix example, is the file name’s suffix known as “extension”, where the code of file’s type became part of the file name. (e.g. “.txt”, “.html”, “.jpg”).
Another well-known example is HTML practices in the industry, where badly designed tags from corporation’s competitive greed, and stupid coding and misunderstanding by coders and their tools are so
wide-spread such that they force the correct way to the side by the eventual standardization caused by sheer quantity of inproper but set practice.

It’s good to get my problem solved, and it’s of so much fun to read these stories

To write very robust and safe code, you should always treat the client as a potential attacker, a bad guy. Consider he will do his best to break your function with invalid arguments.Therefore the good practice is, always validate the input arguments, and throw NullPointerException or IllegalArgumentException accordingly.

Usually for sake of reuse and simplicity, there would be a utility class consisting of several static methods to do this. Then we call it like this

StringUtil.isNullOrEmpty(inputString);


However after read Google Guice’s code accidentally, I now realized this can be rewritten with utilization of import static

checkNotNullNorEmpty(cells);

As you need have the validation inserted at the beginning of every public method, the new approach is surely a time saver. Moreover, you can extend the utility class into a complete Preconditions class, and use it everywhere you want to ‘assert’ something.

Well this is just a very small tip about usage of import static, but life does get better, right? I bet you can also recall JUnit’s assertion way.