Question

in a text file I have following text.

add device 1: /dev/input/event7
  name:     "evfwd"
add device 2: /dev/input/event6
  name:     "aev_abs"
add device 3: /dev/input/event5
  name:     "light-prox"
add device 4: /dev/input/event4
  name:     "qtouch-touchscreen"
add device 5: /dev/input/event2
  name:     "cpcap-key"
add device 6: /dev/input/event1
  name:     "accelerometer"
add device 7: /dev/input/event0
  name:     "compass"
add device 8: /dev/input/event3
  name:     "omap-keypad"

What i need to do is that for each name element i need to extract the corresponding event number and put them on a hashmap.

for example from the text we see that name evfwd is linked with event7, so the hashmap will look like this

<"evfwd", 7>

similarly

<"compass", 0>

How can I efficiently parse the text and build a hashmap like that in java?

Answer 1

You should probably use a regex (regular expression) to parse out the info. To use a regex, look at the Pattern and Matcher classes, here’s an example:

Pattern pattern = Pattern.compile("test");
Matcher matcher = pattern.matcher("some testing thing with a test in it");

You can then call matcher.find() to set the matchers contents to the next match (it returns true if it found another one, false if it hit the end).

You need a regex with what are called capture groups, basically whenever you put ( ) around something in a regex, you can call matcher.group(index) to find the contents of those ( ) (it should be noted that .group(0) returns the whole thing, .group(1) contents of 1st ( ), etc.)

Here’s a regex I just made that should work for you:

add device \d*?: /dev/input/event(\d*?)\n name: \"(.*?)\"

This says in english:

add device [some number]: /dev/input/event[some number, capture group #1]\n name: \"[some string, capture group #2]\"

To get it’s event number, call Integer.parseInt(matcher.group(1)) and to get it’s name, call matcher.group(2).

Tell me if anything fails, hope I helped 🙂