MacKuba

This week I went to Berlin for 2 days to attend the MongoBerlin conference organized by 10gen, the creators of MongoDB. I took a lot of notes from the presentations, and I figured this may be useful for someone if they missed a presentation (or just couldn’t go to the conference).

Feel free to correct me if I got something wrong :)

Update: A new post from December 2010 with updated stats is available here.

For a few weeks I’ve been working on a new iPhone application.

Like most of other Cocoa apps I’ve written so far, this app also includes a JSON parser to load some kind of data from a server. The first thing the application does when it starts is connect to the JSON API, download a data file (about 100 KB) and parse it. This used to take about 10-15 seconds on the device, and I thought it was reasonable until I noticed that the HTTP response actually arrives after a second or so. So what was it doing for the rest of the time? I had to find out.

So I did some debugging, and it turned out that the 10 seconds are spent just on parsing the downloaded JSON file. That’s pretty bad… Like in all previous apps, I used an open source library BSJSONAdditions, which I knew wasn’t the fastest one available, but I never had any major problems with it before. On the other hand, I never tested it on a 100 KB file…

I knew there were a few other JSON parser libraries in ObjC, so I decided to make a small benchmark and see how well they all compared to the one I used. The other libraries I tried were: JSON Framework, TouchJSON and YAJL.

I’m working on a new version of this blog, in which I want to use RightJS. This week I wanted to add some kind of pretty JavaScript tooltips there; there is a Tooltip class in RightJS, but I don’t like the way these tooltips look. However, I know a jQuery library called “TipTip” which adds very attractive black tooltips. So I took the TipTip code and rewrote it using RightJS (and renamed to RipTip, for obvious reasons) – code is available on GitHub, as usual.

This is how the tooltips look:

I came across a problem recently. I have a project called xBlip which I’ve described before – it’s an iPhone client for a Polish Twitter-like service Blip. This project has a backend part which I keep in a subdirectory “ObjectiveBlip” and which I’ve tried to keep as separate from the rest as possible, with the intention that it might be one day extracted as a separate project.

Now I got the idea that I could write a desktop application for Mac that does the same – and of course I could reuse that backend part for that. I would also like to create a separate project on Github with just the backend, so that theoretically someone might use it in future for some purpose.

But this means that I would be maintaining three separate copies of the same code, which I’d have to keep in sync somehow. So the question is, how to do this best?

There are a few ways in Git to share code between projects (for example, git submodules) – but most of them are intended only for one-way communication, i.e. downloading updates to a library maintained by someone else into your project. Here, I want to have a two-way communication: I could extend the backend code while working either on the iPhone or the Mac application (working directly on the backend-only project wouldn’t usually make sense), and then broadcast the changes into the other two projects. I also don’t want the solution to be inconvenient to people who download the project, as is the case with git submodules – you have to manually update them once you download the main code, initially their directories are empty.

So I started looking for a way to pull this off – and I found a script called “git subtree” which seems to do exactly what I need (confusingly, there’s another plugin also called git subtree which is completely unrelated to the first one…). It took me some time (and a few emails to the author, Avery Pennarun) to figure out how to use it, so I thought I’d post a tutorial here in case anyone has a similar situation.

So, here’s what we need to do… (grab a coffee, it’s going to be long):

This year, for several months I’ve been working on a project which involved quite a lot of JavaScript. I’ve already written about how this prompted me to start writing JavaScript unit tests. But as I found out later, there were some kinds of JavaScript errors which the unit tests didn’t help me find. Let me give you an example…

(TLDR: link to the Github project page)

When I work on a Rails web app, I don’t usually test the project in Internet Explorer all the time – if I did that, I’d have gone mad long time ago. Instead, I do everything in Firefox, and leave IE testing for more patient people – either our tester, or even the client in smaller projects. This seems to work well most of the time; however, in this project every 2 weeks or so I used to get such bug report: “The site crashes in IE”. Crashes here means that it doesn’t load at all. You see, if JavaScript is just a nice optional add-on to your project, it’s not a big thing if something doesn’t work; but if your entire application depends on JavaScript for everything it does, then one tiny mistake and you’re screwed.

The problem is that IE has this nasty habit of breaking on code that has a comma at the end of a hash – you know, something like this: { a: 1, b: 2, c: 3, }. And I do this mistake surprisingly often, because that tiny little comma is so easy to miss, and no one ever complains about it except IE – Firefox works fine, unit tests work fine, but next morning I get that dreadful bug report that IE just explodes. Then I wait for VirtualBox to boot Windows and eat all remaining memory, and spend half an hour debugging, only to find that this was all because of one comma…

Yeah, I know that there’s already about 5.4 gazillions of such posts on the web. So what :) I believe that cool apps – especially cool free apps – deserve promotion like this; and although some of the apps listed here are rather well known and popular, some are not.

Most of these apps are free; it’s not like I can’t afford 0.79 €, but I’ve been a Linux user for a few years and I got used to the “free / open source = good, paid = evil” way of thinking. It’s slowly changing, but I still prefer a good free app to a great paid app – unless there aren’t any good free apps available.

I haven’t included any games here – the reason is that I don’t play them a lot on my iPhone; I just don’t have time for that. The only times when I want to play games on it are when I have a bit of time and completely nothing else to do – but in such cases simple logic or board games are enough for me.

So here’s the list (in a rather random order):

A few days ago I spent a couple of hours trying to install the full text search engine Sphinx on my work computer in order to use it in my current project. I’m posting the details here, maybe this will save someone some time…

Firstly, if you’re like me and you look for some kind of binary package (like .dmg) first – sorry, there isn’t one. This leaves us two options: compiling Sphinx manually from the source, or using a package manager like MacPorts. I always avoid the first method if I have any other option, so I tried the ports.

Here’s the first problem: Sphinx requires MySQL, and while I have it on the disk of course, I’ve installed it from a dmg package, not from ports – which means it’s in a different path than ports expect and they have no idea where that is. I got scary errors like:

$ sudo port install sphinx
---> Building sphinx with target all
Error: Target org.macports.build returned: shell command " cd "/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.​macports.​org_release_ports_textproc_sphinx/​work/​sphinx-0.9.7" && make all " returned error 2
Command output: sphinx.cpp: In member function 'virtual const char* CSphSource_MySQL::SqlError()':
sphinx.cpp:9397: error: 'm_tMysqlDriver' was not declared in this scope
sphinx.cpp:9397: error: 'mysql_error' was not declared in this scope
sphinx.cpp: In member function 'virtual bool CSphSource_MySQL::SqlConnect()':
...