Add package path/filepath; Add os.stat for windows (TODO: unix)

This commit is contained in:
gingerBill
2020-09-25 20:20:53 +01:00
parent 6b634d5e46
commit 8cc5cd1494
13 changed files with 997 additions and 18 deletions
+350
View File
@@ -0,0 +1,350 @@
package filepath
import "core:os"
import "core:sort"
import "core:strings"
import "core:unicode/utf8"
Match_Error :: enum {
None,
Syntax_Error,
}
// match states whether "name" matches the shell pattern
// Pattern syntax is:
// pattern:
// {term}
// term:
// '*' matches any sequence of non-/ characters
// '?' matches any single non-/ character
// '[' ['^'] { character-range } ']'
// character classification (cannot be empty)
// c matches character c (c != '*', '?', '\\', '[')
// '\\' c matches character c
//
// character-range
// c matches character c (c != '\\', '-', ']')
// '\\' c matches character c
// lo '-' hi matches character c for lo <= c <= hi
//
// match requires that the pattern matches the entirety of the name, not just a substring
// The only possible error returned is .Syntax_Error
//
// NOTE(bill): This is effectively the shell pattern matching system found
//
match :: proc(pattern, name: string) -> (matched: bool, err: Match_Error) {
pattern, name := pattern, name;
pattern_loop: for len(pattern) > 0 {
star: bool;
chunk: string;
star, chunk, pattern = scan_chunk(pattern);
if star && chunk == "" {
return !strings.contains(name, SEPARATOR_STRING), .None;
}
t: string;
ok: bool;
t, ok, err = match_chunk(chunk, name);
if ok && (len(t) == 0 || len(pattern) > 0) {
name = t;
continue;
}
if err != .None {
return;
}
if star {
for i := 0; i < len(name) && name[i] != SEPARATOR; i += 1 {
t, ok, err = match_chunk(chunk, name[i+1:]);
if ok {
if len(pattern) == 0 && len(t) > 0 {
continue;
}
name = t;
continue pattern_loop;
}
if err != .None {
return;
}
}
}
return false, .None;
}
return len(name) == 0, .None;
}
@(private="file")
scan_chunk :: proc(pattern: string) -> (star: bool, chunk, rest: string) {
pattern := pattern;
for len(pattern) > 0 && pattern[0] == '*' {
pattern = pattern[1:];
star = true;
}
in_range := false;
i: int;
scan_loop: for i = 0; i < len(pattern); i += 1 {
switch pattern[i] {
case '\\':
when ODIN_OS != "windows" {
if i+1 < len(pattern) {
i += 1;
}
}
case '[':
in_range = true;
case ']':
in_range = false;
case '*':
if !in_range {
break scan_loop;
}
}
}
return star, pattern[:i], pattern[i:];
}
@(private="file")
match_chunk :: proc(chunk, s: string) -> (rest: string, ok: bool, err: Match_Error) {
chunk, s := chunk, s;
for len(chunk) > 0 {
if len(s) == 0 {
return;
}
switch chunk[0] {
case '[':
r, w := utf8.decode_rune_in_string(s);
s = s[w:];
chunk = chunk[1:];
is_negated := false;
if len(chunk) > 0 && chunk[0] == '^' {
is_negated = true;
chunk = chunk[1:];
}
match := false;
range_count := 0;
for {
if len(chunk) > 0 && chunk[0] == ']' && range_count > 0 {
chunk = chunk[1:];
break;
}
lo, hi: rune;
if lo, chunk, err = get_escape(chunk); err != .None {
return;
}
hi = lo;
if chunk[0] == '-' {
if hi, chunk, err = get_escape(chunk[1:]); err != .None {
return;
}
}
if lo <= r && r <= hi {
match = true;
}
range_count += 1;
}
if match == is_negated {
return;
}
case '?':
if s[0] == SEPARATOR {
return;
}
_, w := utf8.decode_rune_in_string(s);
s = s[w:];
chunk = chunk[1:];
case '\\':
when ODIN_OS != "windows" {
chunk = chunk[1:];
if len(chunk) == 0 {
err = .Syntax_Error;
return;
}
}
fallthrough;
case:
if chunk[0] != s[0] {
return;
}
s = s[1:];
chunk = chunk[1:];
}
}
return s, true, .None;
}
@(private="file")
get_escape :: proc(chunk: string) -> (r: rune, next_chunk: string, err: Match_Error) {
if len(chunk) == 0 || chunk[0] == '-' || chunk[0] == ']' {
err = .Syntax_Error;
return;
}
chunk := chunk;
if chunk[0] == '\\' && ODIN_OS != "windows" {
chunk = chunk[1:];
if len(chunk) == 0 {
err = .Syntax_Error;
return;
}
}
w: int;
r, w = utf8.decode_rune_in_string(chunk);
if r == utf8.RUNE_ERROR && w == 1 {
err = .Syntax_Error;
}
next_chunk = chunk[w:];
if len(next_chunk) == 0 {
err = .Syntax_Error;
}
return;
}
// glob returns the names of all files matching pattern or nil if there are no matching files
// The syntax of patterns is the same as "match".
// The pattern may describe hierarchical names such as /usr/*/bin (assuming '/' is a separator)
//
// glob ignores file system errors
//
glob :: proc(pattern: string, allocator := context.allocator) -> (matches: []string, err: Match_Error) {
if !has_meta(pattern) {
// TODO(bill): os.lstat on here to check for error
m := make([]string, 1, allocator);
m[0] = pattern;
return m[:], .None;
}
temp_buf: [8]byte;
dir, file := split(pattern);
volume_len := 0;
when ODIN_OS == "windows" {
volume_len, dir = clean_glob_path_windows(dir, temp_buf[:]);
} else {
dir = clean_glob_path(dir);
}
if !has_meta(dir[volume_len:]) {
m, e := _glob(dir, file, nil);
return m[:], e;
}
m: []string;
m, err = glob(dir);
if err != .None {
return;
}
dmatches := make([dynamic]string, 0, 0, allocator);
for d in m {
dmatches, err = _glob(d, file, &dmatches);
if err != .None {
break;
}
}
if len(dmatches) > 0 {
matches = dmatches[:];
}
return;
}
_glob :: proc(dir, pattern: string, matches: ^[dynamic]string) -> (m: [dynamic]string, e: Match_Error) {
if matches != nil {
m = matches^;
} else {
m = make([dynamic]string, 0, 0, context.allocator);
}
d, derr := os.open(dir);
if derr != 0 {
return;
}
defer os.close(d);
fi, ferr := os.stat(d);
if ferr != 0 {
os.file_info_delete(fi);
return;
}
if !fi.is_dir {
os.file_info_delete(fi);
return;
}
fis, _ := os.read_dir(d, -1);
sort.quick_sort_proc(fis, proc(a, b: os.File_Info) -> int {
return sort.compare_strings(a.name, b.name);
});
defer {
for fi in fis {
os.file_info_delete(fi);
}
delete(fis);
}
for fi in fis {
n := fi.name;
matched, err := match(pattern, n);
if err != nil {
return m, err;
}
if matched {
append(&m, join(dir, n));
}
}
return;
}
@(private)
has_meta :: proc(path: string) -> bool {
when ODIN_OS == "windows" {
CHARS :: `*?[`;
} else {
CHARS :: `*?[\`;
}
return strings.contains_any(path, CHARS);
}
@(private)
clean_glob_path :: proc(path: string) -> string {
switch path {
case "":
return ".";
case SEPARATOR_STRING:
return path;
}
return path[:len(path)-1];
}
@(private)
clean_glob_path_windows :: proc(path: string, temp_buf: []byte) -> (prefix_len: int, cleaned: string) {
vol_len := volume_name_len(path);
switch {
case path == "":
return 0, ".";
case vol_len+1 == len(path) && is_separator(path[len(path)-1]): // /, \, C:\, C:/
return vol_len+1, path;
case vol_len == len(path) && len(path) == 2: // C:
copy(temp_buf[:], path);
temp_buf[2] = '.';
return vol_len, string(temp_buf[:3]);
}
if vol_len >= len(path) {
vol_len = len(path) -1;
}
return vol_len, path[:len(path)-1];
}
+199
View File
@@ -0,0 +1,199 @@
package filepath
import "core:os"
import "core:strings"
// is_separator checks whether the byte is a valid separator character
is_separator :: proc(c: byte) -> bool {
switch c {
case '/': return true;
case '\\': return ODIN_OS == "windows";
}
return false;
}
@(private)
is_slash :: proc(c: byte) -> bool {
return c == '\\' || c == '/';
}
split :: proc(path: string) -> (dir, file: string) {
vol := volume_name(path);
i := len(path) - 1;
for i >= len(vol) && !is_separator(path[i]) {
i -= 1;
}
return path[:i+1], path[i+1:];
}
volume_name :: proc(path: string) -> string {
return path[:volume_name_len(path)];
}
volume_name_len :: proc(path: string) -> int {
if len(path) < 2 {
return 0;
}
c := path[0];
if path[1] == ':' {
switch c {
case 'a'..'z', 'A'..'Z':
return 2;
}
}
if l := len(path); l >= 5 && is_slash(path[0]) && is_slash(path[1]) &&
!is_slash(path[2]) && path[2] != '.' {
for n := 3; n < l-1; n += 1 {
if is_slash(path[n]) {
n += 1;
if !is_slash(path[n]) {
if path[n] == '.' {
break;
}
}
for ; n < l; n += 1 {
if is_slash(path[n]) {
break;
}
}
return n;
}
break;
}
}
return 0;
}
clean :: proc(path: string, allocator := context.allocator) -> string {
context.allocator = allocator;
path := path;
original_path := path;
vol_len := volume_name_len(path);
path = path[vol_len:];
if path == "" {
if vol_len > 1 && original_path[1] != ':' {
return from_slash(original_path);
}
return strings.concatenate({original_path, "."});
}
rooted := is_separator(path[0]);
n := len(path);
out := &Lazy_Buffer{
s = path,
vol_and_path = original_path,
vol_len = vol_len,
};
r, dot_dot := 0, 0;
if rooted {
lazy_buffer_append(out, '/');
r, dot_dot = 1, 1;
}
for r < n {
switch {
case is_separator(path[r]):
r += 1;
case path[r] == '.' && (r+1 == n || is_separator(path[r+1])):
r += 1;
case path[r] == '.' && path[r+1] == '.' && (r+2 == n || is_separator(path[r+2])):
r += 2;
switch {
case out.w > dot_dot:
out.w -= 1;
for out.w > dot_dot && !is_separator(lazy_buffer_index(out, out.w)) {
out.w -= 1;
}
case !rooted:
if out.w > 0 {
lazy_buffer_append(out, '/');
}
lazy_buffer_append(out, '.');
lazy_buffer_append(out, '.');
dot_dot = out.w;
}
case:
if rooted && out.w != 1 || !rooted && out.w != 0 {
lazy_buffer_append(out, '/');
}
for ; r < n && !is_separator(path[r]); r += 1 {
lazy_buffer_append(out, path[r]);
}
}
}
if out.w == 0 {
lazy_buffer_append(out, '.');
}
s := lazy_buffer_string(out);
cleaned := from_slash(s);
return cleaned;
}
from_slash :: proc(path: string, allocator := context.allocator) -> string {
if SEPARATOR == '/' {
return path;
}
s, ok := strings.replace_all(path, "/", SEPARATOR_STRING, allocator);
if !ok {
s = strings.clone(s, allocator);
}
return s;
}
/*
Lazy_Buffer is a lazily made path buffer
When it does allocate, it uses the context.allocator
*/
@(private)
Lazy_Buffer :: struct {
s: string,
b: []byte,
w: int, // write index
vol_and_path: string,
vol_len: int,
}
@(private)
lazy_buffer_index :: proc(lb: ^Lazy_Buffer, i: int) -> byte {
if lb.b != nil {
return lb.b[i];
}
return lb.s[i];
}
@(private)
lazy_buffer_append :: proc(lb: ^Lazy_Buffer, c: byte) {
if lb.b == nil {
if lb.w < len(lb.s) && lb.s[lb.w] == c {
lb.w += 1;
return;
}
lb.b = make([]byte, len(lb.s));
copy(lb.b, lb.s[:lb.w]);
}
lb.b[lb.w] = c;
lb.w += 1;
}
@(private)
lazy_buffer_string :: proc(lb: ^Lazy_Buffer) -> string {
if lb.b == nil {
return strings.clone(lb.vol_and_path[:lb.vol_len+lb.w]);
}
x := lb.vol_and_path[:lb.vol_len];
y := string(lb.b[:lb.w]);
z := make([]byte, len(x)+len(y));
copy(z, x);
copy(z[len(x):], y);
return string(z);
}
+5
View File
@@ -0,0 +1,5 @@
//+build linux, darwin, freebsd
package filepath
SEPARATOR :: '/';
SEPARATOR_STRING :: `/`;
+71
View File
@@ -0,0 +1,71 @@
package filepath
import "core:strings"
SEPARATOR :: '\\';
SEPARATOR_STRING :: `\`;
reserved_names := []string{
"CON", "PRN", "AUX", "NUL",
"COM1", "COM2", "COM3", "COM4", "COM5", "COM6", "COM7", "COM8", "COM9",
"LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
};
is_reserved_name :: proc(path: string) -> bool {
if len(path) == 0 {
return false;
}
for reserved in reserved_names {
if strings.equal_fold(path, reserved) {
return true;
}
}
return false;
}
is_UNC :: proc(path: string) -> bool {
return volume_name_len(path) > 2;
}
join :: proc(elems: ..string, allocator := context.allocator) -> string {
for e, i in elems {
if e != "" {
return join_non_empty(elems[i:]);
}
}
return "";
}
join_non_empty :: proc(elems: []string) -> string {
if len(elems[0]) == 2 && elems[0][1] == ':' {
i := 1;
for ; i < len(elems); i += 1 {
if elems[i] != "" {
break;
}
}
s := strings.join(elems[i:], SEPARATOR_STRING, context.temp_allocator);
s = strings.concatenate({elems[0], s}, context.temp_allocator);
return clean(s);
}
s := strings.join(elems, SEPARATOR_STRING, context.temp_allocator);
p := clean(s);
if !is_UNC(p) {
return p;
}
head := clean(elems[0], context.temp_allocator);
if is_UNC(head) {
return p;
}
delete(p); // It is not needed now
tail := clean(strings.join(elems[1:], SEPARATOR_STRING, context.temp_allocator), context.temp_allocator);
if head[len(head)-1] == SEPARATOR {
return strings.concatenate({head, tail});
}
return strings.concatenate({head, SEPARATOR_STRING, tail});
}