RSS Feed :: Ben Humphreys

どんどん vs だんだん, ますます

Mon, 08 Oct 2007 04:19:00 -0700

だんだん vs どんどん

I was presented with this example sentence in a JLPT1 book.

「聞いたそばから、＿＿覚えていくとは、彼の記憶力は抜群だ。」

a) どんどん　b) だんだん　c) ゆっくり

I was kind of stumped as I knew that ゆっくり meant slow, but didn't know the difference between どんどん and だんだん. Thankfully スイカさん explained it to me like this:

どんどん is faster than だんだん.
どんどん is about the thing itself increasing (therefore faster).
だんだん has an emphasis on time passing (therefore slower).

Here's how she explained it: そばから　の意味は「次々に同じことを繰り返してる様子を表す」。繰り返すだから、「だんだん」だと思った

でも正解はどんどん

どんどんのほうが、だんだんより、早い

どんどんは、物が増えていくかんじ。

だんだんは、_時間_がたっていくかんじ。

聞いたそばから　っていうのは　「聞いたあと、すぐ」　みたいな意味だから、だんだんだとすぐではない。ちょっと、時間が必要なかんじ。

どんどん　だったら物が増えていく感じだから、聞いたあとすぐに、どんどん覚えても大丈夫。

Hard to Translate Phrases

のりがいい　Good sense of humour/timing? How would you define “のりがいい”? I'd really appreciate some feedback on this, as I'm still not sure how Japanese people perceive the phrase.

右に出るものがない insurpassable, there isn't anything better than...

不貞寝 The dictionary says “futene - staying in bed out of spite” Does anyone use this nowadays?

Odd Japanese English

I found out a while back that Japanese people use two different English words to describe stripes. For vertical stripes it's ストライプ, whereas horizontal stripes are usually called ボーダー.

Notes on ますます

Summarized, big things can get bigger, small things can get smaller.

○ 大きい物が、ますます大きくなれる
○ 小さいものも、ますます小さくなれる
× おおきいものが、ますます小さくなる
× ちいさいものがますますおおきくなる、できない
× ないものは、ますます増える
○ big → bigger
○ small → smaller
× big → smaller
× small → bigger

I hope people find this interesting. If you have any ideas on what I've mentioned, or want to point out any mistakes I've made, I look forward to hearing from you :)

Japanese Nuances

Sun, 07 Oct 2007 02:43:00 -0700

Comparisons

I often come across words in Anki that seem very similar, and I have a hard time remembering which kanji is associated with two seemingly identical descriptions. To clear up the different nuances I asked some Japanese friends (thanks Genさん and other 2ch-ers).

拒絶　vs　辞退

拒絶 is stronger, more severe. “No way in hell.” kind of feeling.

辞退 is relatively soft. The answer is no, but there’s a reason behind it.

逃げる　vs　逃れる

逃げる　人から逃げる　　You run away from a person.

逃れる　災いから逃れる　You flee from a disaster.

I'm still not 100% sure about this difference.

適応　vs　適用

適応 Natural adaptation to external conditions (becoming accustomed to the situation).

適用 Actively adapting to external conditions (problem solving).

When you consider the meanings of the 2nd kanji, it becomes more easy to remember:

用いる = to use

応じる = to respond, to satisfy, to comply with

Some examples ○ 僕が海外に行って生活習慣に適応した。　I became used to life in a foreign country.

○ プログラムを適用した

× プログラムを適応した　Because you are actively adapting the program, should be 適用

褒美　vs　獲物

獲物 is the target, the thing you’re trying to get. Reflected by its translations: game, spoils, trophy

褒美 is more like a present, the reward you get for winning.

New Vocabulary

品切れ　しなぎれ　sold out
間近　　まぢか　　nearing
まるごと　　　　　whole
結成　　けっせい　formation
節目　　ふしめ　　turning point
放つ　　はなつ　　let loose, let fly, release
傑作　　けっさく　masterpiece
円満　　えんまん　satisfactory
秘訣　　ひけつ　　secret
過酷　　かこく　　stern
酔っぱ　よっぱ　　drunk (compare 漢字)
葛藤　　かっとう　conflict
瞬時　　しゅんじ　instant
壮麗　　そうれい　magnificent
引用　　いんよう　quotation
うなぎのぼり　　　rapid promotion

Weird Japanese English

ビッグ・タイアップ・シングル　Same as the English, the main single (song) that’s associated with the movie.

キャッチ・コピー　Taking something and using it. The example was using a line from the title song as the main catchphrase of a movie.

attr_protected and Foreign Keys

Sun, 16 Sep 2007 05:09:00 -0700

Say you have a site that allows users to create favourites, and a favourite is composed of a url and a user_id. How is it best to stop users trying to add favourites to other people's accounts? I know this seems like an unlikely scenario but I think it's a point that highlights other security issues.

One approach I thought about was using attrprotected :userid, which I think would do the job, but am I missing something about the way Rails works?

As always, I'd appreciate any thoughts.

Learn from Existing Code: Typo

Thu, 06 Sep 2007 03:29:00 -0700

I find it really useful to learn from example code and look through existing applications to see how they tick. I came across Typo a while ago, but didn't really think much of it as I planned to write my own blog for practice. It occurred to me the other day that as Ruby is an interpreted language, I could look at the source code of free apps like Typo.

Typo is a large application and so it's a bit hard to get your head round at first, but if you're looking for a big example, try downloading Typo, and then using it to write your own ;)

Caching in Rails

Wed, 05 Sep 2007 18:43:00 -0700

Caching in Rails is kind of like C++, if you know what you're doing it's possible to make a small elegant solution, but if you get it wrong then god help you when you try to work out what's going on.

There are plenty of tutorials online and sections in books that go into the basics of caching and some of the finer points, but here are some particular problems I encountered.

Moving Parameters from Session

Most tutorials hammer in that Rails caching only works based on the URL, so session variables and parameters are completely forgotten. The original version of my site stored the user's preferred language in the session and retrieved the language data based on that.

By changing the language to be specified by the URL (e.g. benhumphreys.co.uk/en/...) instead of the session, I solved the caching problem. Then it was just a matter of updating the

routes.rb

map.home ':language/', :controller => 'entries', :requirements => { :language => /(en|ja)/}

map.en 'en', :controller => 'entries', :language => 'en' map.ja 'ja', :controller => 'entries', :language => 'ja'

map.connect ':language/:controller/:action/:id', :requirements => { :language => /(en|ja)/ }

In the same way as when you specify a link_to with only the action Rails assumes you're in the same controller/path, Rails assumes you're in the same language/path if you don't specify the language. So link_tos automatically point to the controller in the right language.

Get Your Paths Right

If like me you have your admin section in a subdirectory double-check that the path in your expire_page commands have a preceding backslash or you'll be trying to expire stuff in the same admin directory (thanks to rsl in #rubyonrails for helping me decypher this).

Also, as far as I know you have to specify every part of the route. When using link_to, Rails can use the current location to work out unspecified parts of the route. However as sweepers can be called from any location, you must specify all parts of the route (lines 21 & 24). class EntrySweeper < ActionController::Caching::Sweeper

observe Entry

# QUESTION: Does anyone know how to roll these 3 into one method?
def after_create(entry)
    expire(entry)
end 

def after_update(entry)
    expire(entry)
end

def after_destroy(entry)
    expire(entry)
end

private
def expire(entry)
    # expire single page
    expire_page(:controller => '/entries', :action => 'view', :language => entry.language, :type => entry.type, :slug => entry.slug)

    # expire list page for type
    expire_page(:controller => '/entries', :action => 'list', :language => language, :type => entry.type)

    # expire general list
    expire_page(:language => language, :controller => '/entries', :action => 'list')

    # expire rss feed
    expire_page(:language => language, :controller => '/entries', :action => 'rss')
end

end

Use the Logs!

If you ever don't know what's going on in your Rails app, be sure to check the logs. The same applies for when you're implementing caching. Make sure you turn on caching in config/environment/development.rb, and use the logs to work out what pages are being expired and whether the paths are correct.

Expired page: /en/articles/caching-in-rails.html (0.00367) Expired page: /en/entries/list.html (0.00037) Expired page: /en/rss.xml (0.00046)

I'm not sure how this compares with other people's experience, but developing on a iBook the time taken to expire existing pages (the first entry) and the time taken to expire non-existant pages (usually indicative of a bug) are noticably different and can help you debug. Not rocket science but I thought I'd point it out.

General Tips

Here are some general points to remember when implementing caching, some of them seem obvious but they might save you some time trying to solve unusual bugs:

render vs redirect_to

Cache files are created from the URL, irrespective of how the URL was reached. Remember that the following code will stay at the original URL. list render :action => 'list' Whereas the following changes the URL. redirect_to :action => 'list'

Multiple Paths to Enlightenment

If you have many ways of reaching the same content, e.g view/id and permalink, you need to be especially careful.

Final Word

I'm still a long way from knowing how to do caching on a large scale and I've yet to do detailed fragment caching, so I'd appreciate any corrections anyone has on the points I've made.