1
0
mirror of https://github.com/golang/go synced 2024-11-12 10:30:23 -07:00

gofmt-ify unicode

R=r
http://go/go-review/1018051
This commit is contained in:
Robert Griesemer 2009-11-04 21:39:55 -08:00
parent 55ba9d6a2c
commit 841c18a95a
4 changed files with 402 additions and 409 deletions

View File

@ -5,8 +5,8 @@
package unicode_test
import (
"testing";
. "unicode";
"testing";
. "unicode";
)
var testDigit = []int{

View File

@ -5,8 +5,8 @@
package unicode_test
import (
"testing";
. "unicode";
"testing";
. "unicode";
)
var upperTest = []int{

View File

@ -30,7 +30,7 @@ func main() {
printCases();
}
var dataUrl = flag.String("data", "", "full URL for UnicodeData.txt; defaults to --url/UnicodeData.txt");
var dataUrl = flag.String("data", "", "full URL for UnicodeData.txt; defaults to --url/UnicodeData.txt")
var url = flag.String("url",
"http://www.unicode.org/Public/5.1.0/ucd/",
"URL of Unicode database directory")
@ -51,9 +51,9 @@ var test = flag.Bool("test",
"test existing tables; can be used to compare web data with package data")
var scriptRe = regexp.MustCompile(`([0-9A-F]+)(\.\.[0-9A-F]+)? *; ([A-Za-z_]+)`)
var die = log.New(os.Stderr, nil, "", log.Lexit|log.Lshortfile)
var die = log.New(os.Stderr, nil, "", log.Lexit | log.Lshortfile)
var category = map[string] bool{ "letter":true } // Nd Lu etc. letter is a special case
var category = map[string]bool{"letter": true} // Nd Lu etc. letter is a special case
// UnicodeData.txt has form:
// 0037;DIGIT SEVEN;Nd;0;EN;;7;7;7;N;;;;;
@ -61,7 +61,7 @@ var category = map[string] bool{ "letter":true } // Nd Lu etc. letter is a speci
// See http://www.unicode.org/Public/5.1.0/ucd/UCD.html for full explanation
// The fields:
const (
FCodePoint = iota;
FCodePoint = iota;
FName;
FGeneralCategory;
FCanonicalCombiningClass;
@ -78,7 +78,7 @@ const (
FSimpleTitlecaseMapping;
NumField;
MaxChar = 0x10FFFF; // anything above this shouldn't exist
MaxChar = 0x10FFFF; // anything above this shouldn't exist
)
var fieldName = []string{
@ -96,13 +96,13 @@ var fieldName = []string{
"ISOComment",
"SimpleUppercaseMapping",
"SimpleLowercaseMapping",
"SimpleTitlecaseMapping"
"SimpleTitlecaseMapping",
}
// This contains only the properties we're interested in.
type Char struct {
field []string; // debugging only; could be deleted if we take out char.dump()
codePoint uint32; // if zero, this index is not a valid code point.
field []string; // debugging only; could be deleted if we take out char.dump()
codePoint uint32; // if zero, this index is not a valid code point.
category string;
upperCase int;
lowerCase int;
@ -120,8 +120,8 @@ type Script struct {
}
var chars = make([]Char, MaxChar+1)
var scripts = make(map[string] []Script)
var props = make(map[string] []Script) // a property looks like a script; can share the format
var scripts = make(map[string][]Script)
var props = make(map[string][]Script) // a property looks like a script; can share the format
var lastChar uint32 = 0
@ -130,8 +130,9 @@ var lastChar uint32 = 0
// 4DB5;<CJK Ideograph Extension A, Last>;Lo;0;L;;;;;N;;;;;
// parseCategory returns a state variable indicating the weirdness.
type State int
const (
SNormal State = iota; // known to be zero for the type
SNormal State = iota; // known to be zero for the type
SFirst;
SLast;
SMissing;
@ -144,17 +145,17 @@ func parseCategory(line string) (state State) {
}
point, err := strconv.Btoui64(field[FCodePoint], 16);
if err != nil {
die.Log("%.5s...:", err)
die.Log("%.5s...:", err);
}
lastChar = uint32(point);
if point == 0 {
return // not interesting and we use 0 as unset
return; // not interesting and we use 0 as unset
}
if point > MaxChar {
return;
}
char := &chars[point];
char.field=field;
char.field = field;
if char.codePoint != 0 {
die.Logf("point U+%04x reused\n");
}
@ -179,16 +180,16 @@ func parseCategory(line string) (state State) {
}
switch {
case strings.Index(field[FName], ", First>") > 0:
state = SFirst
state = SFirst;
case strings.Index(field[FName], ", Last>") > 0:
state = SLast
state = SLast;
}
return
return;
}
func (char *Char) dump(s string) {
fmt.Print(s, " ");
for i:=0;i<len(char.field);i++ {
for i := 0; i < len(char.field); i++ {
fmt.Printf("%s:%q ", fieldName[i], char.field[i]);
}
fmt.Print("\n");
@ -202,14 +203,14 @@ func (char *Char) letter(u, l, t string) {
func (char *Char) letterValue(s string, cas string) int {
if s == "" {
return 0
return 0;
}
v, err := strconv.Btoui64(s, 16);
if err != nil {
char.dump(cas);
die.Logf("U+%04x: bad letter(%s): %s", char.codePoint, s, err)
die.Logf("U+%04x: bad letter(%s): %s", char.codePoint, s, err);
}
return int(v)
return int(v);
}
func allCategories() []string {
@ -222,7 +223,7 @@ func allCategories() []string {
return a;
}
func all(scripts map[string] []Script) []string {
func all(scripts map[string][]Script) []string {
a := make([]string, len(scripts));
i := 0;
for k := range scripts {
@ -238,7 +239,7 @@ func version() string {
fields := strings.Split(*url, "/", 0);
for _, f := range fields {
if len(f) > 0 && '0' <= f[0] && f[0] <= '9' {
return f
return f;
}
}
die.Log("unknown version");
@ -248,9 +249,9 @@ func version() string {
func letterOp(code int) bool {
switch chars[code].category {
case "Lu", "Ll", "Lt", "Lm", "Lo":
return true
return true;
}
return false
return false;
}
func loadChars() {
@ -274,25 +275,25 @@ func loadChars() {
}
die.Log(err);
}
switch parseCategory(line[0:len(line)-1]) {
switch parseCategory(line[0 : len(line)-1]) {
case SNormal:
if first != 0 {
die.Logf("bad state normal at U+%04X", lastChar)
die.Logf("bad state normal at U+%04X", lastChar);
}
case SFirst:
if first != 0 {
die.Logf("bad state first at U+%04X", lastChar)
die.Logf("bad state first at U+%04X", lastChar);
}
first = lastChar
first = lastChar;
case SLast:
if first == 0 {
die.Logf("bad state last at U+%04X", lastChar)
die.Logf("bad state last at U+%04X", lastChar);
}
for i := first+1; i <= lastChar; i++ {
chars[i] = chars[first];
chars[i].codePoint = i;
}
first = 0
first = 0;
}
}
resp.Body.Close();
@ -300,16 +301,16 @@ func loadChars() {
func printCategories() {
if *tablelist == "" {
return
return;
}
// Find out which categories to dump
list := strings.Split(*tablelist, ",", 0);
if *tablelist == "all" {
list = allCategories()
list = allCategories();
}
if *test {
fullCategoryTest(list);
return
return;
}
fmt.Printf(
"// Generated by running\n"
@ -317,15 +318,14 @@ func printCategories() {
"// DO NOT EDIT\n\n"
"package unicode\n\n",
*tablelist,
*dataUrl
);
*dataUrl);
fmt.Println("// Version is the Unicode edition from which the tables are derived.");
fmt.Printf("const Version = %q\n\n", version());
if *tablelist == "all" {
fmt.Println("// Categories is the set of Unicode data tables.");
fmt.Println("var Categories = map[string] []Range {");
fmt.Println("var Categories = map[string] []Range {");
for k, _ := range category {
fmt.Printf("\t%q: %s,\n", k, k);
}
@ -358,22 +358,19 @@ func printCategories() {
if name != "letter" {
varDecl += fmt.Sprintf(
"\t%s = _%s; // %s is the set of Unicode characters in category %s.\n",
name, name, name, name
);
name, name, name, name);
}
decl[ndecl] = varDecl;
ndecl++;
if name == "letter" { // special case
dumpRange(
"var letter = []Range {\n",
letterOp
);
letterOp);
continue;
}
dumpRange(
fmt.Sprintf("var _%s = []Range {\n", name),
func(code int) bool { return chars[code].category == name }
);
func(code int) bool { return chars[code].category == name });
}
decl.Sort();
fmt.Println("var (");
@ -384,7 +381,8 @@ func printCategories() {
}
type Op func(code int) bool
const format = "\tRange{0x%04x, 0x%04x, %d},\n";
const format = "\tRange{0x%04x, 0x%04x, %d},\n"
func dumpRange(header string, inCategory Op) {
fmt.Print(header);
@ -393,11 +391,11 @@ func dumpRange(header string, inCategory Op) {
for {
// look for start of range
for next < len(chars) && !inCategory(next) {
next++
next++;
}
if next >= len(chars) {
// no characters remain
break
break;
}
// start of range
@ -408,7 +406,7 @@ func dumpRange(header string, inCategory Op) {
next++;
// look for another character to set the stride
for next < len(chars) && !inCategory(next) {
next++
next++;
}
if next >= len(chars) {
// no more characters
@ -416,22 +414,22 @@ func dumpRange(header string, inCategory Op) {
break;
}
// set stride
stride = next - lo;
stride = next-lo;
// check for length of run. next points to first jump in stride
for i := next; i < len(chars); i++ {
if inCategory(i) == (((i-lo)%stride) == 0) {
// accept
if inCategory(i) {
hi = i
hi = i;
}
} else {
// no more characters in this run
break
break;
}
}
fmt.Printf(format, lo, hi, stride);
// next range: start looking where this range ends
next = hi + 1;
next = hi+1;
}
fmt.Print("}\n\n");
}
@ -451,8 +449,7 @@ func fullCategoryTest(list []string) {
verifyRange(
name,
func(code int) bool { return chars[code].category == name },
r
);
r);
}
}
}
@ -467,14 +464,14 @@ func verifyRange(name string, inCategory Op, table []unicode.Range) {
}
}
func parseScript(line string, scripts map[string] []Script) {
func parseScript(line string, scripts map[string][]Script) {
comment := strings.Index(line, "#");
if comment >= 0 {
line = line[0:comment]
line = line[0:comment];
}
line = strings.TrimSpace(line);
if len(line) == 0 {
return
return;
}
field := strings.Split(line, ";", -1);
if len(field) != 2 {
@ -486,13 +483,13 @@ func parseScript(line string, scripts map[string] []Script) {
}
lo, err := strconv.Btoui64(matches[1], 16);
if err != nil {
die.Log("%.5s...:", err)
die.Log("%.5s...:", err);
}
hi := lo;
if len(matches[2]) > 2 { // ignore leading ..
hi, err = strconv.Btoui64(matches[2][2:len(matches[2])], 16);
if err != nil {
die.Log("%.5s...:", err)
die.Log("%.5s...:", err);
}
}
name := matches[3];
@ -500,12 +497,12 @@ func parseScript(line string, scripts map[string] []Script) {
if !ok || len(s) == cap(s) {
ns := make([]Script, len(s), len(s)+100);
for i, sc := range s {
ns[i] = sc
ns[i] = sc;
}
s = ns;
}
s = s[0:len(s)+1];
s[len(s)-1] = Script{ uint32(lo), uint32(hi), name };
s = s[0 : len(s)+1];
s[len(s)-1] = Script{uint32(lo), uint32(hi), name};
scripts[name] = s;
}
@ -514,10 +511,10 @@ func foldAdjacent(r []Script) []unicode.Range {
s := make([]unicode.Range, 0, len(r));
j := 0;
for i := 0; i < len(r); i++ {
if j>0 && int(r[i].lo) == s[j-1].Hi+1 {
if j > 0 && int(r[i].lo) == s[j-1].Hi + 1 {
s[j-1].Hi = int(r[i].hi);
} else {
s = s[0:j+1];
s = s[0 : j+1];
s[j] = unicode.Range{int(r[i].lo), int(r[i].hi), 1};
j++;
}
@ -525,7 +522,7 @@ func foldAdjacent(r []Script) []unicode.Range {
return s;
}
func fullScriptTest(list []string, installed map[string] []unicode.Range, scripts map[string] []Script) {
func fullScriptTest(list []string, installed map[string][]unicode.Range, scripts map[string][]Script) {
for _, name := range list {
if _, ok := scripts[name]; !ok {
die.Log("unknown script", name);
@ -559,7 +556,7 @@ func printScriptOrProperty(doProps bool) {
installed = unicode.Properties;
}
if flaglist == "" {
return
return;
}
var err os.Error;
resp, _, err := http.Get(*url + file);
@ -578,7 +575,7 @@ func printScriptOrProperty(doProps bool) {
}
die.Log(err);
}
parseScript(line[0:len(line)-1], table);
parseScript(line[0 : len(line)-1], table);
}
resp.Body.Close();
@ -598,8 +595,7 @@ func printScriptOrProperty(doProps bool) {
"// DO NOT EDIT\n\n",
flag,
flaglist,
*url
);
*url);
if flaglist == "all" {
if doProps {
fmt.Println("// Properties is the set of Unicode property tables.");
@ -620,13 +616,11 @@ func printScriptOrProperty(doProps bool) {
if doProps {
decl[ndecl] = fmt.Sprintf(
"\t%s = _%s;\t// %s is the set of Unicode characters with property %s.\n",
name, name, name, name
);
name, name, name, name);
} else {
decl[ndecl] = fmt.Sprintf(
"\t%s = _%s;\t// %s is the set of Unicode characters in script %s.\n",
name, name, name, name
);
name, name, name, name);
}
ndecl++;
fmt.Printf("var _%s = []Range {\n", name);
@ -645,16 +639,16 @@ func printScriptOrProperty(doProps bool) {
}
const (
CaseUpper = 1 << iota;
CaseUpper = 1<<iota;
CaseLower;
CaseTitle;
CaseNone = 0; // must be zero
CaseMissing = -1; // character not present; not a valid case state
CaseNone = 0; // must be zero
CaseMissing = -1; // character not present; not a valid case state
)
type caseState struct {
point int;
_case int;
point int;
_case int;
deltaToUpper int;
deltaToLower int;
deltaToTitle int;
@ -663,25 +657,25 @@ type caseState struct {
// Is d a continuation of the state of c?
func (c *caseState) adjacent(d *caseState) bool {
if d.point < c.point {
c, d = d, c
c, d = d, c;
}
switch {
case d.point != c.point+1: // code points not adjacent (shouldn't happen)
return false
case d.point != c.point + 1: // code points not adjacent (shouldn't happen)
return false;
case d._case != c._case: // different cases
return c.upperLowerAdjacent(d);
case c._case == CaseNone:
return false
return false;
case c._case == CaseMissing:
return false
return false;
case d.deltaToUpper != c.deltaToUpper:
return false
return false;
case d.deltaToLower != c.deltaToLower:
return false
return false;
case d.deltaToTitle != c.deltaToTitle:
return false
return false;
}
return true;
return true;
}
// Is d the same as c, but opposite in upper/lower case? this would make it
@ -690,32 +684,32 @@ func (c *caseState) upperLowerAdjacent(d *caseState) bool {
// check they're a matched case pair. we know they have adjacent values
switch {
case c._case == CaseUpper && d._case != CaseLower:
return false
return false;
case c._case == CaseLower && d._case != CaseUpper:
return false
return false;
}
// matched pair (at least in upper/lower). make the order Upper Lower
if c._case == CaseLower {
c, d = d, c
c, d = d, c;
}
// for an Upper Lower sequence the deltas have to be in order
// c: 0 1 0
// d: -1 0 -1
switch {
case c.deltaToUpper != 0:
return false
return false;
case c.deltaToLower != 1:
return false
return false;
case c.deltaToTitle != 0:
return false
return false;
case d.deltaToUpper != -1:
return false
return false;
case d.deltaToLower != 0:
return false
return false;
case d.deltaToTitle != -1:
return false
return false;
}
return true
return true;
}
// Does this character start an UpperLower sequence?
@ -724,13 +718,13 @@ func (c *caseState) isUpperLower() bool {
// c: 0 1 0
switch {
case c.deltaToUpper != 0:
return false
return false;
case c.deltaToLower != 1:
return false
return false;
case c.deltaToTitle != 0:
return false
return false;
}
return true
return true;
}
// Does this character start a LowerUpper sequence?
@ -739,17 +733,17 @@ func (c *caseState) isLowerUpper() bool {
// c: -1 0 -1
switch {
case c.deltaToUpper != -1:
return false
return false;
case c.deltaToLower != 0:
return false
return false;
case c.deltaToTitle != -1:
return false
return false;
}
return true
return true;
}
func getCaseState(i int) (c *caseState) {
c = &caseState{ point: i, _case: CaseNone };
c = &caseState{point: i, _case: CaseNone};
ch := &chars[i];
switch int(ch.codePoint) {
case 0:
@ -763,24 +757,24 @@ func getCaseState(i int) (c *caseState) {
c._case = CaseTitle;
}
if ch.upperCase != 0 {
c.deltaToUpper = ch.upperCase - i
c.deltaToUpper = ch.upperCase - i;
}
if ch.lowerCase != 0 {
c.deltaToLower = ch.lowerCase - i
c.deltaToLower = ch.lowerCase - i;
}
if ch.titleCase != 0 {
c.deltaToTitle = ch.titleCase - i
c.deltaToTitle = ch.titleCase - i;
}
return;
}
func printCases() {
if !*cases {
return
return;
}
if *test {
fullCaseTest();
return
return;
}
fmt.Printf(
"// Generated by running\n"
@ -790,8 +784,7 @@ func printCases() {
"// non-self mappings.\n"
"var CaseRanges = _CaseRanges\n"
"var _CaseRanges = []CaseRange {\n",
*dataUrl
);
*dataUrl);
var startState *caseState; // the start of a run; nil for not active
var prevState = &caseState{}; // the state of the previous character
@ -814,33 +807,33 @@ func printCases() {
func printCaseRange(lo, hi *caseState) {
if lo == nil {
return
return;
}
if lo.deltaToUpper == 0 && lo.deltaToLower == 0 && lo.deltaToTitle == 0 {
// character represents itself in all cases - no need to mention it
return
return;
}
switch {
case hi.point > lo.point && lo.isUpperLower():
fmt.Printf("\tCaseRange{0x%04X, 0x%04X, d{UpperLower, UpperLower, UpperLower}},\n",
lo.point, hi.point)
lo.point, hi.point);
case hi.point > lo.point && lo.isLowerUpper():
die.Log("LowerUpper sequence: should not happen: U+%04X. If it's real, need to fix To()", lo.point);
fmt.Printf("\tCaseRange{0x%04X, 0x%04X, d{LowerUpper, LowerUpper, LowerUpper}},\n",
lo.point, hi.point)
lo.point, hi.point);
default:
fmt.Printf("\tCaseRange{0x%04X, 0x%04X, d{%d, %d, %d}},\n",
lo.point, hi.point,
lo.deltaToUpper, lo.deltaToLower, lo.deltaToTitle)
lo.deltaToUpper, lo.deltaToLower, lo.deltaToTitle);
}
}
// If the cased value in the Char is 0, it means use the rune itself.
func caseIt(rune, cased int) int {
if cased == 0 {
return rune
return rune;
}
return cased
return cased;
}
func fullCaseTest() {

File diff suppressed because it is too large Load Diff