Reputation: 2199
In reading a file in Go I am attempting to skip all of the white spaces; however, I am having issues finding the correct method to do this. Any assistance would be appreciated
file, err := os.Open(filename) // For read access.
this.file = file
if err != nil {
log.Fatal(err)
}
//skip white space
c := make([]byte, 1)
char, err := this.file.Read(c)
//skip white space
for {
//catch unintended errors
if err != nil && err != io.EOF {
panic(err)
}
if err == io.EOF || !unicode.IsSpace(int(c)) {
break
}
//get next
char, err := this.file.Read(c)
}
I am just simply attempting to create a scanner for a file to read a single character at a time and ignore whitespace
EDIT
I changed a few things around to make use of bufio.Reader; however I have still fallen into issue What is the correct way to read a file character by character so that it might be compared to a specific symbol such as 'A' but also can ignore whitespace i.e unicode.isSpace(rune)
char, size, err := this.reader.ReadRune()
//skip white space and comments
for {
//catch unintended errors
if err != nil && err != io.EOF {
panic(err)
}
if err == io.EOF {
break
}
//skip it when their is no data or a space
if size != 0 && char == '{' {
//Ignore Comments
//Documentation specifies no nested comments
for char != '}' {
char, size, err = this.reader.ReadRune()
}
} else if !unicode.IsSpace(char) {
break
}
// Do something with the byte
fmt.Print(char)
//get next
char, size, err = this.reader.ReadRune()
}
Upvotes: 0
Views: 3351
Reputation: 2568
Unless I'm misunderstanding your question, it would seem that you'd want a continue
statement when encountering a space.
c := make([]byte, 100)
n, err := this.file.Read(c)
//skip white space
for {
//catch unintended errors
if err != nil && err != io.EOF {
panic(err)
}
if err == io.EOF {
break
}
for i := 0; i < n; i++ {
ch := c[i]
switch ch {
case '{': // Do something
case '}': // Do something else
default:
if unicode.IsSpace(int(ch)) {
continue
}
// Do whatever
}
}
//get next
n, err = this.file.Read(c)
}
I don't know why you're reading one byte at a time, but I left it that way in case it's intentional. At the very least, I'd think you'd want to read full unicode characters instead of individual bytes.
Upvotes: 2