On iOS there's a handy function on NSNumber
localizedStringWithStyle:
Which will format a numeric value into a string using digits appropriate to that region (possibly Western, possible Arabic, etc).
I assume in Java on Android a NumberFormat object (or similar) could be configured to do this but I'm not sure how.
So my question. In Android how to you format a number in locale specific digits?
When you instantiate the NumberFormat object, you call it like this:
NumberFormat formatter = NumberFormat.getInstance(Locale.US);
You can pick from the list of static Locales or call NumberFormat.getAvailableLocales() to search for a locale which matches your desired output.
int n;
String.format("%d", n);
That's all. String.format uses the user's locale by default. You can request a particular locale by passing it as the first argument (i.e. before the format string).
Related
I am trying to format a NumberValue as a String in the current locale format.
For example, for Locale.Germany these would be the results I am trying to get:
1 -> 1,00
1.1 -> 1,10
0.1 -> 0,10
What I have tried:
String test1 = NumberFormat.getInstance().format(1.1);
// -> 1,1
String test2 = NumberFormat.getCurrencyInstance().format(1.1);
// -> 1,10 €
How could I get the same format as the second example, without the currency symbol? Or alternativly, how could I make sure the result from the first example always has the amount of decimals needed for the current locale?
Edit: Since there are some currencies that have 0,2 or 3 decimals, I don't think setting the decimals to a fixed amount will have the correct results.
Also, some currencies have their symbol in the front, and some in the back, so cutting off the last characters will probably not work either.
http://apps.cybersource.com/library/documentation/sbc/quickref/currencies.pdf
java.util.Currency provides the necessary information. A reasonable approach to print currency values without the currency sign in pure Java code is to get a general-purpose NumberFormat instance and configure it to use the number of digits defined by the appropriate Currency object:
NumberFormat numberFormat = NumberFormat.getInstance(myLocale);
Currency currency = Currency.getInstance(myLocale);
numberFormat.setMinimumFractionDigits(currency.getDefaultFractionDigits());
If you need a specific currency as opposed to the default currency of the locale you use, use the String override of Currency.getInstance() instead and provide the appropriate currency code.
I was wondering if there was an elegant way to set the currency format of a number in one way but keep the actual number formatting in another. It is Java. Essentially I am doing this
NumberFormat format = NumberFormat.getCurrencyInstance(locale);
This is largely fine except I am writing a system for UK based users and my users are upset that when showing (for example) euros the number is formatted as europeans would use it. So € 500,000 as a UK person would write it is displaying as € 500.000 (i.e. swap , for .). I was going to swap locale for locale.UK but then I will have the wrong currency symbol!
I have a couple of dirty fixes for this but I wondered if there was an elegant way to keep the currency symbol of the locale with the local locale (locale.UK) number format.
Different currencies can also place the currency symbol before or after the string, or have a different number of decimal places (if any). It is not really clear from your question how you want to handle those cases, but assuming you want to preserve those differences, try this.
Instead of just swapping in the currency symbol into your local number format, you could start with the foreign format and substitute the decimal format symbols with your local version. Those also include the currency, so you have to swap that back (don't worry, it's a copy).
public static NumberFormat localStyleForeignFormat(Locale locale) {
NumberFormat format = NumberFormat.getCurrencyInstance(locale);
if (format instanceof DecimalFormat) {
DecimalFormat df = (DecimalFormat) format;
// use local/default decimal symbols with original currency symbol
DecimalFormatSymbols dfs = new DecimalFormat().getDecimalFormatSymbols();
dfs.setCurrency(df.getCurrency());
df.setDecimalFormatSymbols(dfs);
}
return format;
}
This way, you also retain the correct positioning of the currency symbol and the number of decimal places. Some examples, for default-Locale Locale.UK
en_GB £500,000.00 £500,000.00
fr_FR 500 000,00 € 500,000.00 €
it_IT € 500.000,00 € 500,000.00
ja_JP ￥500,000 JPY500,000
hi_IN रू ५००,०००.०० INR 500,000.00
If you also want to preserve the foreign currency symbol, instead of the local equivalent, use
localDfs.setCurrencySymbol(df.getCurrency().getSymbol(locale));
You can specify the currency symbol on the NumberFormat with the setCurrency method.
Then simply use the Locale.UK to have the proper grouping separator displayed.
format.setCurrency(Currency.getInstance("EUR"));
Note that for a better handling of the grouping/decimal separator you might want to use a DecimalFormat instead.
DecimalFormatSymbols custom=new DecimalFormatSymbols();
custom.setDecimalSeparator('.');
custom.setGroupingSeparator(',');
DecimalFormat format = DecimalFormat.getInstance();
format.setDecimalFormatSymbols(custom);
format.setCurrency(Currency.getInstance("EUR"));
Then specify the correct pattern, example "€ ###,###.00".
I have an app which runs the following two lines of code upon starting:
DecimalFormat decim = new DecimalFormat("#.00");
return Double.parseDouble(decim.format(totalNumberOfCredits));
When I start the app on my American phone, the value of decim.format(totalNumberOfCredits) is .00.
However, in my Google Play Developer Console, I have a dozen crashes, all of which look like this:
Caused by: java.lang.NumberFormatException: Invalid double: ",00"
at java.lang.StringToReal.invalidReal(StringToReal.java:63)
at java.lang.StringToReal.parseDouble(StringToReal.java:269)
at java.lang.Double.parseDouble(Double.java:295)
Is it really possible that DecimalFormat is producing a comma version of the decimal on European phones?
Is it really possible that DecimalFormat is producing a comma version of the decimal on European phones?
Yes, absolutely. That's what it's meant to do, after all:
Creates a DecimalFormat using the given pattern and the symbols for the default locale. This is a convenient way to obtain a DecimalFormat when internationalization is not the main concern.
To obtain standard formats for a given locale, use the factory methods on NumberFormat such as getNumberInstance. These factories will return the most appropriate sub-class of NumberFormat for a given locale.
Note that this isn't a matter of a "European version of Android" - it's just a matter of using Android in a context where the default locale uses , as the decimal separator.
If you want to use the symbols for a particular locale, but using a specific pattern, you could use:
DecimalFormatSymbols symbols = DecimalFormatSymbols.getInstance(Locale.US);
DecimalFormat format = new DecimalFormat("#.00", symbols);
Having said that, it's not at all clear what you're trying to do in the first place - why would you format and then parse a number? You should almost always avoid string conversions when you don't really need them. Why not just convert it directly? (We don't know what totalNumberOfCredits is, which doesn't help.)
public double getTwoPointDecimal(double value) {
DecimalFormatSymbols symbols = new DecimalFormatSymbols(Locale.US);
return Double.parseDouble(new DecimalFormat("##.##", symbols).format(value));
}
try it, its help me in my project
double unit = Float.parseFloat(String);
DecimalFormat decimal = new DecimalFormat("##.###").format(unit);
try this it help me in my project
I'm doing some javascript work inside a ColdFusion shopping cart, and I need to be able to format some numbers in js which will mimic LScurrencyFormat() in CF.
Currently we are taking the first (left,1) character of a formatted string but that doesn't work for currencies like Yen or Euro which come after the number, not to mention any multiple character currency symbols.
What I need to find, based on the current CF locale, is
currency symbol
decimal delimiter (, or .)
leading or trailing (before or after the number)
From there i can run my own js formatting to make the formatted numbers come out as expected on the page.In php we can use localeconv() to get these values... how can I find them in CF?
I am not aware of any built in functions. However, you can obtain the first two items from java. As far as the third, the closest suggestion I have seen is to parse the localized number pattern and detect the position of the currency sign ie \u00A4. Note: It is just a mask placeholder. It is not the same as the actual currency symbols like "$" or "£".
Edit:
As discussed in the comments, getLocale() returns some user friendly name which unfortunately does not quite line up with java's. The easiest way to get the java locale object for the current request is using getPageContext().getResponse().getLocale().
<cfscript>
// Get the current locale as a java object
javaLocale = getPageContext().getResponse().getLocale();
// get numeric settings for that locale
currency = createObject("java", "java.text.DecimalFormat").getCurrencyInstance(javaLocale);
symbols = currency.getDecimalFormatSymbols();
// 164 => decimal code point for currency sign
currencyPattern = currency.toLocalizedPattern();
result.hasTrailingCurrencySymbol = currencyPattern.indexOf(javacast("int", 164)) > 0;
result.currencySymbol = symbols.getCurrencySymbol();
result.decimalSeparator= symbols.getDecimalSeparator();
WriteDump(result);
</cfscript>
getLocale() returns the old cf5 style locale "names" but only for those locales supported by cf5. if you dump out the supported locales (Server.Coldfusion.SupportedLocales) you'll see the goofy old cf5 style locale names as well as the core java locale IDs (ie both "Chinese(China)" and "zh_CN"). if your locale wasn't one of the cf5 supported locales you should see the core java locale ID (ie th_TH for thai, thailand). see
http://cfbugs.adobe.com/cfbugreport/flexbugui/cfbugtracker/main.html#bugId=82474
as a small tweak to leigh's answer, you should also be concerned with the currency/locale's fraction digits. for instance in normal practice, you can't have part of a yen (ie 1.1 isn't quite kosher). you can get that info from the Currency class's getDefaultFractionDigits() method:
result.fractionDigits=currency.getDefaultFractionDigits();
I have a java string variable in my groovy app. The variable contains a user input of price in possibly different currency formats:
val = "1,250.50"
val = "1.250,50"
val = "1250,50"
val = "1250.50"
(etc.. I don't know if there are anymore funny way other countries write this)
Is there a way to parse this to the appropriate double value regardless of the format? Looking at this at the moment but not sure if it'll help. My current method only works for the US format:
total = Double.parseDouble(val.replace('$','').replaceAll(",","").trim())
You cannot parse it without knowing what the user will use as decimal separator, grouping separator, ... . For example if I type 1,250 you do not know whether I mean one thousand two hundred fifty (1,250.00), or one point two hundred fifty (1.250) .
That's why the NumberFormat/DecimalFormat class of Java allows you to specify the grouping and decimal separator.
What you could do is hoping that the user inputs his values using the conventions corresponding to his Locale settings, and use the
NumberFormat.getInstance( Locale )
with the current Locale of the JVM.
Note: with the NumberFormat you can also parse a currency. See NumberFormat#getCurrencyInstance