The String Methods You Will Actually Use

The page says Total: Rs 1,299.00. Your expected value is the number 1299. Between those two facts sits about a third of the work in real test automation. Almost nothing arrives ready to compare. A price is scraped with a currency symbol, a thousands separator, and two decimal places you do not care about. A name comes out of a spreadsheet with a trailing space nobody can see. A status arrives from an API in a different case than your expectation. Every one of these is a text problem, and getting them wrong produces the worst kind of test result: a failure that is not a defect.

Start with the raw material and look at what is available:

String scraped = "  Total: Rs 1,299.00  ";

raw          : '  Total: Rs 1,299.00  '
trim()       : 'Total: Rs 1,299.00'
length       : 18
contains Rs? : true
startsWith   : true
indexOf 'Rs' : 7
upper        : TOTAL: RS 1,299.00

Note the single quotes in that output. Printing a value inside quote marks is a small habit worth adopting permanently, because it makes invisible whitespace visible. ' DEF-101 ' and 'DEF-101' look identical in a log without them, and they are not equal.

The everyday set is short. trim() removes leading and trailing whitespace, and strip() is its modern equivalent that also handles unusual space characters. length() counts characters. contains, startsWith, and endsWith answer yes-or-no questions. indexOf gives the position of the first match, or -1 when there is none. toUpperCase and toLowerCase normalize case — which, as Chapter 6 showed, is what makes a comparison survive real data. And isEmpty() asks whether the length is zero, while isBlank() also treats a string of only spaces as empty, which is usually the question you meant — test data from spreadsheets is full of cells holding a single space, so isBlank is the check you usually want.

Cleaning: from scraped text to a number

Put them together and the opening problem solves itself:

String cleaned = scraped.trim()
                        .replace("Total:", "")
                        .replace("Rs", "")
                        .replace(",", "")
                        .trim();
double amount = Double.parseDouble(cleaned);

cleaned : '1299.00'
as number: 1299.0
rupees   : 1299

Two things to notice. The calls chain, because each one returns a String that the next one operates on. And the final trim() is not redundant — removing "Total:" from the middle leaves a space behind, and the parse would fail without it. Cleaning is usually a small pipeline, not a single call.

The last step converts text to a number, which is where a test suite most often meets reality. Integer.parseInt and Double.parseDouble are strict: they accept digits and a sign, and reject everything else with the NumberFormatException you met in Chapter 9. The comma is exactly why the cleaning had to come first.

Every one of those calls just used quietly returned a brand-new String and left the original alone. That is not incidental — it is a rule with real consequences, and it explains a bug you've already met.