Merge pull request #38 from stephen679/vis

vis: refactor code to use Vis() and Unvis()

check.go

@@ -53,8 +53,11 @@ func Check(root string, dh *DirectoryHierarchy, keywords []string) (*Result, err
 				creator.curSet = nil
 			}
 		case RelativeType, FullType:
-			filename := e.Path()
+			pathname, err := e.Path()
-			info, err := os.Lstat(filename)
+			if err != nil {
+				return nil, err
+			}
+			info, err := os.Lstat(pathname)
 			if err != nil {
 				return nil, err
 			}
@@ -77,23 +80,23 @@ func Check(root string, dh *DirectoryHierarchy, keywords []string) (*Result, err
 
 				keywordFunc, ok := KeywordFuncs[kw]
 				if !ok {
-					return nil, fmt.Errorf("Unknown keyword %q for file %q", kv.Keyword(), e.Path())
+					return nil, fmt.Errorf("Unknown keyword %q for file %q", kv.Keyword(), pathname)
 				}
 				if keywords != nil && !inSlice(kv.Keyword(), keywords) {
 					continue
 				}
-				fh, err := os.Open(filename)
+				fh, err := os.Open(pathname)
 				if err != nil {
 					return nil, err
 				}
-				curKeyVal, err := keywordFunc(filename, info, fh)
+				curKeyVal, err := keywordFunc(pathname, info, fh)
 				if err != nil {
 					fh.Close()
 					return nil, err
 				}
 				fh.Close()
 				if string(kv) != curKeyVal {
-					failure := Failure{Path: e.Path(), Keyword: kv.Keyword(), Expected: kv.Value(), Got: KeyVal(curKeyVal).Value()}
+					failure := Failure{Path: pathname, Keyword: kv.Keyword(), Expected: kv.Value(), Got: KeyVal(curKeyVal).Value()}
 					result.Failures = append(result.Failures, failure)
 				}
 			}
@@ -133,8 +136,12 @@ func TarCheck(tarDH, dh *DirectoryHierarchy, keywords []string) (*Result, error)
 				creator.curSet = nil
 			}
 		case RelativeType, FullType:
+			pathname, err := e.Path()
+			if err != nil {
+				return nil, err
+			}
 			if outOfTree {
-				return &result, fmt.Errorf("No parent node from %s", e.Path())
+				return &result, fmt.Errorf("No parent node from %s", pathname)
 			}
 			// TODO: handle the case where "." is not the first Entry to be found
 			tarEntry := curDir.Descend(e.Name)
@@ -165,15 +172,20 @@ func TarCheck(tarDH, dh *DirectoryHierarchy, keywords []string) (*Result, error)
 			}
 
 			for _, kv := range kvs {
+
 				if _, ok := KeywordFuncs[kv.Keyword()]; !ok {
-					return nil, fmt.Errorf("Unknown keyword %q for file %q", kv.Keyword(), e.Path())
+					return nil, fmt.Errorf("Unknown keyword %q for file %q", kv.Keyword(), pathname)
 				}
 				if keywords != nil && !inSlice(kv.Keyword(), keywords) {
 					continue
 				}
+				tarpath, err := tarEntry.Path()
+				if err != nil {
+					return nil, err
+				}
 				if tarkv := tarkvs.Has(kv.Keyword()); tarkv != emptyKV {
 					if string(tarkv) != string(kv) {
-						failure := Failure{Path: tarEntry.Path(), Keyword: kv.Keyword(), Expected: kv.Value(), Got: tarkv.Value()}
+						failure := Failure{Path: tarpath, Keyword: kv.Keyword(), Expected: kv.Value(), Got: tarkv.Value()}
 						result.Failures = append(result.Failures, failure)
 					}
 				}

check_test.go

@@ -260,3 +260,38 @@ func TestIgnoreComments(t *testing.T) {
 		t.Fatal(res.Failures)
 	}
 }
+
+func TestCheckNeedsEncoding(t *testing.T) {
+	dir, err := ioutil.TempDir("", "test-needs-encoding")
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer os.RemoveAll(dir)
+
+	fh, err := os.Create(filepath.Join(dir, "file[ "))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := fh.Close(); err != nil {
+		t.Error(err)
+	}
+	fh, err = os.Create(filepath.Join(dir, "    , should work"))
+	if err != nil {
+		t.Fatal(err)
+	}
+	if err := fh.Close(); err != nil {
+		t.Error(err)
+	}
+
+	dh, err := Walk(dir, nil, DefaultKeywords)
+	if err != nil {
+		t.Fatal(err)
+	}
+	res, err := Check(dir, dh, nil)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if len(res.Failures) > 0 {
+		t.Fatal(res.Failures)
+	}
+}

cmd/gomtree/main.go

@@ -202,13 +202,25 @@ func main() {
 			if len(res.Extra) > 0 {
 				defer os.Exit(1)
 				for _, extra := range res.Extra {
-					fmt.Printf("%s extra\n", extra.Path())
+					extrapath, err := extra.Path()
+					if err != nil {
+						log.Println(err)
+						isErr = true
+						return
+					}
+					fmt.Printf("%s extra\n", extrapath)
 				}
 			}
 			if len(res.Missing) > 0 {
 				defer os.Exit(1)
 				for _, missing := range res.Missing {
-					fmt.Printf("%s missing\n", missing.Path())
+					missingpath, err := missing.Path()
+					if err != nil {
+						log.Println(err)
+						isErr = true
+						return
+					}
+					fmt.Printf("%s missing\n", missingpath)
 				}
 			}
 		} else {

entry.go

@@ -48,14 +48,25 @@ func (e Entry) Ascend() *Entry {
 	return e.Parent
 }
 
-// Path provides the full path of the file, despite RelativeType or FullType
+// Path provides the full path of the file, despite RelativeType or FullType. It
-func (e Entry) Path() string {
+// will be in Unvis'd form.
-	if e.Parent == nil || e.Type == FullType {
+func (e Entry) Path() (string, error) {
-		return filepath.Clean(e.Name)
+	decodedName, err := Unvis(e.Name)
+	if err != nil {
+		return "", err
 	}
-	return filepath.Clean(filepath.Join(e.Parent.Path(), e.Name))
+	if e.Parent == nil || e.Type == FullType {
+		return filepath.Clean(decodedName), nil
+	}
+	parentName, err := e.Parent.Path()
+	if err != nil {
+		return "", err
+	}
+	return filepath.Clean(filepath.Join(parentName, decodedName)), nil
 }
 
+// String joins a file with its associated keywords. The file name will be the
+// Vis'd encoded version so that it can be parsed appropriately when Check'd.
 func (e Entry) String() string {
 	if e.Raw != "" {
 		return e.Raw

hierarchy.go

@@ -16,7 +16,8 @@ func (dh DirectoryHierarchy) WriteTo(w io.Writer) (n int64, err error) {
 	sort.Sort(byPos(dh.Entries))
 	var sum int64
 	for _, e := range dh.Entries {
-		i, err := io.WriteString(w, e.String()+"\n")
+		str := e.String()
+		i, err := io.WriteString(w, str+"\n")
 		if err != nil {
 			return sum, err
 		}

tar.go

@@ -97,8 +97,15 @@ func (ts *tarStream) readHeaders() {
 		defer os.Remove(tmpFile.Name())
 
 		// Alright, it's either file or directory
+		encodedName, err := Vis(filepath.Base(hdr.Name))
+		if err != nil {
+			tmpFile.Close()
+			os.Remove(tmpFile.Name())
+			ts.pipeReader.CloseWithError(err)
+			return
+		}
 		e := Entry{
-			Name: filepath.Base(hdr.Name),
+			Name: encodedName,
 			Type: RelativeType,
 		}
 
@@ -213,8 +220,13 @@ func populateTree(root, e *Entry, hdr *tar.Header, ts *tarStream) {
 			if isDir {
 				newEntry = e
 			} else {
+				encodedName, err := Vis(name)
+				if err != nil {
+					ts.setErr(err)
+					return
+				}
 				newEntry = &Entry{
-					Name: name,
+					Name: encodedName,
 					Type: RelativeType,
 				}
 			}
@@ -230,8 +242,13 @@ func populateTree(root, e *Entry, hdr *tar.Header, ts *tarStream) {
 		parent.Children = append([]*Entry{e}, parent.Children...)
 		e.Parent = parent
 	} else {
+		commentpath, err := e.Path()
+		if err != nil {
+			ts.setErr(err)
+			return
+		}
 		commentEntry := Entry{
-			Raw:  "# " + e.Path(),
+			Raw:  "# " + commentpath,
 			Type: CommentType,
 		}
 		e.Prev = &commentEntry

tar_test.go

@@ -119,12 +119,20 @@ func TestTar(t *testing.T) {
 			errors += "Keyword validation errors\n"
 		case len(res.Missing) > 0:
 			for _, m := range res.Missing {
-				t.Errorf("Missing file: %s\n", m.Path())
+				missingpath, err := m.Path()
+				if err != nil {
+					t.Fatal(err)
+				}
+				t.Errorf("Missing file: %s\n", missingpath)
 			}
 			errors += "Missing files not expected for this test\n"
 		case len(res.Extra) > 0:
 			for _, e := range res.Extra {
-				t.Errorf("Extra file: %s\n", e.Path())
+				extrapath, err := e.Path()
+				if err != nil {
+					t.Fatal(err)
+				}
+				t.Errorf("Extra file: %s\n", extrapath)
 			}
 			errors += "Extra files not expected for this test\n"
 		}

unvis.c

@@ -0,0 +1,293 @@
+/*-
+ * Copyright (c) 1989, 1993
+ *      The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 4. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#if defined(LIBC_SCCS) && !defined(lint)
+static char sccsid[] = "@(#)unvis.c     8.1 (Berkeley) 6/4/93";
+#endif /* LIBC_SCCS and not lint */
+
+#include <sys/cdefs.h>
+
+#include <sys/types.h>
+#include <ctype.h>
+#include "vis.h"
+
+/*
+ * decode driven by state machine
+ */
+#define S_GROUND        0       /* haven't seen escape char */
+#define S_START         1       /* start decoding special sequence */
+#define S_META          2       /* metachar started (M) */
+#define S_META1         3       /* metachar more, regular char (-) */
+#define S_CTRL          4       /* control char started (^) */
+#define S_OCTAL2        5       /* octal digit 2 */
+#define S_OCTAL3        6       /* octal digit 3 */
+#define S_HEX2          7       /* hex digit 2 */
+
+#define S_HTTP          0x080   /* %HEXHEX escape */
+
+#define isoctal(c)      (((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
+#define ishex(c)        ((((u_char)(c)) >= '0' && ((u_char)(c)) <= '9') || (((u_char)(c)) >= 'a' && ((u_char)(c)) <= 'f'))
+
+/*
+ * unvis - decode characters previously encoded by vis
+ */
+int
+unvis(char *cp, int c, int *astate, int flag)
+{
+
+        if (flag & UNVIS_END) {
+                if (*astate == S_OCTAL2 || *astate == S_OCTAL3) {
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                }
+                return (*astate == S_GROUND ? UNVIS_NOCHAR : UNVIS_SYNBAD);
+        }
+
+        switch (*astate & ~S_HTTP) {
+
+        case S_GROUND:
+                *cp = 0;
+                if (c == '\\') {
+                        *astate = S_START;
+                        return (0);
+                }
+                if (flag & VIS_HTTPSTYLE && c == '%') {
+                        *astate = S_START | S_HTTP;
+                        return (0);
+                }
+                *cp = c;
+                return (UNVIS_VALID);
+
+        case S_START:
+                if (*astate & S_HTTP) {
+                    if (ishex(tolower(c))) {
+                        *cp = isdigit(c) ? (c - '0') : (tolower(c) - 'a');
+                        *astate = S_HEX2;
+                        return (0);
+                    }
+                }
+                switch(c) {
+                case '\\':
+                        *cp = c;
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case '0': case '1': case '2': case '3':
+                case '4': case '5': case '6': case '7':
+                        *cp = (c - '0');
+                        *astate = S_OCTAL2;
+                        return (0);
+                case 'M':
+                        *cp = 0200;
+                        *astate = S_META;
+                        return (0);
+                case '^':
+                        *astate = S_CTRL;
+                        return (0);
+                case 'n':
+                        *cp = '\n';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'r':
+                        *cp = '\r';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'b':
+                        *cp = '\b';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'a':
+                        *cp = '\007';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'v':
+                        *cp = '\v';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 't':
+                        *cp = '\t';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'f':
+                        *cp = '\f';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 's':
+                        *cp = ' ';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case 'E':
+                        *cp = '\033';
+                        *astate = S_GROUND;
+                        return (UNVIS_VALID);
+                case '\n':
+                        /*
+                         * hidden newline
+                         */
+                        *astate = S_GROUND;
+                        return (UNVIS_NOCHAR);
+                case '$':
+                        /*
+                         * hidden marker
+                         */
+                        *astate = S_GROUND;
+                        return (UNVIS_NOCHAR);
+                }
+                *astate = S_GROUND;
+                return (UNVIS_SYNBAD);
+
+        case S_META:
+                if (c == '-')
+                        *astate = S_META1;
+                else if (c == '^')
+                        *astate = S_CTRL;
+                else {
+                        *astate = S_GROUND;
+                        return (UNVIS_SYNBAD);
+                }
+                return (0);
+
+        case S_META1:
+                *astate = S_GROUND;
+                *cp |= c;
+                return (UNVIS_VALID);
+
+        case S_CTRL:
+                if (c == '?')
+                        *cp |= 0177;
+                else
+                        *cp |= c & 037;
+                *astate = S_GROUND;
+                return (UNVIS_VALID);
+
+        case S_OCTAL2:  /* second possible octal digit */
+                if (isoctal(c)) {
+                        /*
+                         * yes - and maybe a third
+                         */
+                        *cp = (*cp << 3) + (c - '0');
+                        *astate = S_OCTAL3;
+                        return (0);
+                }
+                /*
+                 * no - done with current sequence, push back passed char
+                 */
+                *astate = S_GROUND;
+                return (UNVIS_VALIDPUSH);
+
+        case S_OCTAL3:  /* third possible octal digit */
+                *astate = S_GROUND;
+                if (isoctal(c)) {
+                        *cp = (*cp << 3) + (c - '0');
+                        return (UNVIS_VALID);
+                }
+                /*
+                 * we were done, push back passed char
+                 */
+                return (UNVIS_VALIDPUSH);
+
+        case S_HEX2:    /* second mandatory hex digit */
+                if (ishex(tolower(c))) {
+                        *cp = (isdigit(c) ? (*cp << 4) + (c - '0') : (*cp << 4) + (tolower(c) - 'a' + 10));
+                }
+                *astate = S_GROUND;
+                return (UNVIS_VALID);
+
+        default:
+                /*
+                 * decoder in unknown state - (probably uninitialized)
+                 */
+                *astate = S_GROUND;
+                return (UNVIS_SYNBAD);
+        }
+}
+
+/*
+ * strunvis - decode src into dst
+ *
+ *      Number of chars decoded into dst is returned, -1 on error.
+ *      Dst is null terminated.
+ */
+
+int
+strunvis(char *dst, const char *src)
+{
+        char c;
+        char *start = dst;
+        int state = 0;
+
+        while ( (c = *src++) ) {
+        again:
+                switch (unvis(dst, c, &state, 0)) {
+                case UNVIS_VALID:
+                        dst++;
+                        break;
+                case UNVIS_VALIDPUSH:
+                        dst++;
+                        goto again;
+                case 0:
+                case UNVIS_NOCHAR:
+                        break;
+                default:
+                        return (-1);
+                }
+        }
+        if (unvis(dst, c, &state, UNVIS_END) == UNVIS_VALID)
+                dst++;
+        *dst = '\0';
+        return (dst - start);
+}
+
+int
+strunvisx(char *dst, const char *src, int flag)
+{
+        char c;
+        char *start = dst;
+        int state = 0;
+    
+        while ( (c = *src++) ) {
+        again:
+                switch (unvis(dst, c, &state, flag)) {
+                case UNVIS_VALID:
+                        dst++;
+                        break;
+                case UNVIS_VALIDPUSH:
+                        dst++;
+                        goto again;
+                case 0:
+                case UNVIS_NOCHAR:
+                        break;
+                default:
+                        return (-1);
+                }
+        }
+        if (unvis(dst, c, &state, UNVIS_END) == UNVIS_VALID)
+                dst++;
+        *dst = '\0';
+        return (dst - start);
+}

unvis.go

@@ -0,0 +1,22 @@
+package mtree
+
+// #include "vis.h"
+// #include <stdlib.h>
+import "C"
+import (
+	"fmt"
+	"unsafe"
+)
+
+// Unvis is a wrapper for the C implementation of unvis, which decodes a string
+// that potentially has characters that are encoded with Vis
+func Unvis(src string) (string, error) {
+	cDst, cSrc := C.CString(string(make([]byte, len(src)+1))), C.CString(src)
+	defer C.free(unsafe.Pointer(cDst))
+	defer C.free(unsafe.Pointer(cSrc))
+	ret := C.strunvis(cDst, cSrc)
+	if ret == -1 {
+		return "", fmt.Errorf("failed to decode: %q", src)
+	}
+	return C.GoString(cDst), nil
+}

vis.c

@@ -0,0 +1,202 @@
+/*-
+ * Copyright (c) 1989, 1993
+ *      The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 4. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#if defined(LIBC_SCCS) && !defined(lint)
+static char sccsid[] = "@(#)vis.c       8.1 (Berkeley) 7/19/93";
+#endif /* LIBC_SCCS and not lint */
+
+
+#include <sys/cdefs.h>
+
+#include <sys/types.h>
+#include <limits.h>
+#include <ctype.h>
+#include <stdio.h>
+#include "vis.h"
+
+#define isoctal(c)      (((u_char)(c)) >= '0' && ((u_char)(c)) <= '7')
+
+/*
+ * vis - visually encode characters
+ */
+char *
+vis(dst, c, flag, nextc)
+        char *dst;
+        int c, nextc;
+        int flag;
+{
+        c = (unsigned char)c;
+
+        if (flag & VIS_HTTPSTYLE) {
+                /* Described in RFC 1808 */
+                if (!(isalnum(c) /* alpha-numeric */
+                    /* safe */
+                    || c == '$' || c == '-' || c == '_' || c == '.' || c == '+'
+                    /* extra */
+                    || c == '!' || c == '*' || c == '\'' || c == '('
+                    || c == ')' || c == ',')) {
+                        *dst++ = '%';
+                        snprintf(dst, 4, (c < 16 ? "0%X" : "%X"), c);
+                        dst += 2;
+                        goto done;
+                }
+        }
+
+        if ((flag & VIS_GLOB) &&
+            (c == '*' || c == '?' || c == '[' || c == '#'))
+                ;
+        else if (isgraph(c) ||
+           ((flag & VIS_SP) == 0 && c == ' ') ||
+           ((flag & VIS_TAB) == 0 && c == '\t') ||
+           ((flag & VIS_NL) == 0 && c == '\n') ||
+           ((flag & VIS_SAFE) && (c == '\b' || c == '\007' || c == '\r'))) {
+                *dst++ = c;
+                if (c == '\\' && (flag & VIS_NOSLASH) == 0)
+                        *dst++ = '\\';
+                *dst = '\0';
+                return (dst);
+        }
+
+        if (flag & VIS_CSTYLE) {
+                switch(c) {
+                case '\n':
+                        *dst++ = '\\';
+                        *dst++ = 'n';
+                        goto done;
+                case '\r':
+                        *dst++ = '\\';
+                        *dst++ = 'r';
+                        goto done;
+                case '\b':
+                        *dst++ = '\\';
+                        *dst++ = 'b';
+                        goto done;
+                case '\a':
+                        *dst++ = '\\';
+                        *dst++ = 'a';
+                        goto done;
+                case '\v':
+                        *dst++ = '\\';
+                        *dst++ = 'v';
+                        goto done;
+                case '\t':
+                        *dst++ = '\\';
+                        *dst++ = 't';
+                        goto done;
+                case '\f':
+                        *dst++ = '\\';
+                        *dst++ = 'f';
+                        goto done;
+                case ' ':
+                        *dst++ = '\\';
+                        *dst++ = 's';
+                        goto done;
+                case '\0':
+                        *dst++ = '\\';
+                        *dst++ = '0';
+                        if (isoctal(nextc)) {
+                                *dst++ = '0';
+                                *dst++ = '0';
+                        }
+                        goto done;
+                }
+        }
+        if (((c & 0177) == ' ') || isgraph(c) || (flag & VIS_OCTAL)) {
+                *dst++ = '\\';
+                *dst++ = ((u_char)c >> 6 & 07) + '0';
+                *dst++ = ((u_char)c >> 3 & 07) + '0';
+                *dst++ = ((u_char)c & 07) + '0';
+                goto done;
+        }
+        if ((flag & VIS_NOSLASH) == 0)
+                *dst++ = '\\';
+        if (c & 0200) {
+                c &= 0177;
+                *dst++ = 'M';
+        }
+        if (iscntrl(c)) {
+                *dst++ = '^';
+                if (c == 0177)
+                        *dst++ = '?';
+                else
+                        *dst++ = c + '@';
+        } else {
+                *dst++ = '-';
+                *dst++ = c;
+        }
+done:
+        *dst = '\0';
+        return (dst);
+}
+
+/*
+ * strvis, strvisx - visually encode characters from src into dst
+ *
+ *      Dst must be 4 times the size of src to account for possible
+ *      expansion.  The length of dst, not including the trailing NUL,
+ *      is returned.
+ *
+ *      Strvisx encodes exactly len bytes from src into dst.
+ *      This is useful for encoding a block of data.
+ */
+int
+strvis(dst, src, flag)
+        char *dst;
+        const char *src;
+        int flag;
+{
+        char c;
+        char *start;
+
+        for (start = dst; (c = *src); )
+                dst = vis(dst, c, flag, *++src);
+        *dst = '\0';
+        return (dst - start);
+}
+
+int
+strvisx(dst, src, len, flag)
+        char *dst;
+        const char *src;
+        size_t len;
+        int flag;
+{
+        int c;
+        char *start;
+
+        for (start = dst; len > 1; len--) {
+                c = *src;
+                dst = vis(dst, c, flag, *++src);
+        }
+        if (len)
+                dst = vis(dst, *src, flag, '\0');
+        *dst = '\0';
+
+        return (dst - start);
+}

vis.go

@@ -0,0 +1,26 @@
+package mtree
+
+// #include "vis.h"
+// #include <stdlib.h>
+import "C"
+import (
+	"fmt"
+	"math"
+	"unsafe"
+)
+
+// Vis is a wrapper of the C implementation of the function vis, which encodes
+// a character with a particular format/style
+func Vis(src string) (string, error) {
+	// dst needs to be 4 times the length of str, must check appropriate size
+	if uint32(len(src)*4+1) >= math.MaxUint32/4 {
+		return "", fmt.Errorf("failed to encode: %q", src)
+	}
+	dst := string(make([]byte, 4*len(src)+1))
+	cDst, cSrc := C.CString(dst), C.CString(src)
+	defer C.free(unsafe.Pointer(cDst))
+	defer C.free(unsafe.Pointer(cSrc))
+	C.strvis(cDst, cSrc, C.VIS_WHITE|C.VIS_OCTAL|C.VIS_GLOB)
+
+	return C.GoString(cDst), nil
+}

vis.h

@@ -0,0 +1,90 @@
+/*-
+ * Copyright (c) 1990, 1993
+ *      The Regents of the University of California.  All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. All advertising materials mentioning features or use of this software
+ *    must display the following acknowledgement:
+ *      This product includes software developed by the University of
+ *      California, Berkeley and its contributors.
+ * 4. Neither the name of the University nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ *
+ *      @(#)vis.h       8.1 (Berkeley) 6/2/93
+ * $FreeBSD$
+ */
+
+#ifndef _VIS_H_
+#define _VIS_H_
+
+#include <sys/types.h>
+
+/*
+ * to select alternate encoding format
+ */
+#define VIS_OCTAL       0x01    /* use octal \ddd format */
+#define VIS_CSTYLE      0x02    /* use \[nrft0..] where appropriate */
+
+/*
+ * to alter set of characters encoded (default is to encode all
+ * non-graphic except space, tab, and newline).
+ */
+#define VIS_SP          0x04    /* also encode space */
+#define VIS_TAB         0x08    /* also encode tab */
+#define VIS_NL          0x10    /* also encode newline */
+#define VIS_WHITE       (VIS_SP | VIS_TAB | VIS_NL)
+#define VIS_SAFE        0x20    /* only encode "unsafe" characters */
+
+/*
+ * other
+ */
+#define VIS_NOSLASH     0x40    /* inhibit printing '\' */
+#define VIS_HTTPSTYLE   0x80    /* http-style escape % HEX HEX */
+#define VIS_GLOB        0x100   /* encode glob(3) magics */
+
+/*
+ * unvis return codes
+ */
+#define UNVIS_VALID      1      /* character valid */
+#define UNVIS_VALIDPUSH  2      /* character valid, push back passed char */
+#define UNVIS_NOCHAR     3      /* valid sequence, no character produced */
+#define UNVIS_SYNBAD    -1      /* unrecognized escape sequence */
+#define UNVIS_ERROR     -2      /* decoder in unknown state (unrecoverable) */
+
+/*
+ * unvis flags
+ */
+#define UNVIS_END       1       /* no more characters */
+
+#include <sys/cdefs.h>
+
+__BEGIN_DECLS
+char    *vis(char *, int, int, int);
+int     strvis(char *, const char *, int);
+int     strvisx(char *, const char *, size_t, int);
+int     strunvis(char *, const char *);
+int     strunvisx(char *, const char *, int);
+int     unvis(char *, int, int *, int);
+__END_DECLS
+
+#endif /* !_VIS_H_ */

vis_test.go

@@ -0,0 +1,49 @@
+package mtree
+
+import "testing"
+
+func TestVis(t *testing.T) {
+	testset := []struct {
+		Src, Dest string
+	}{
+		{"[", "\\133"},
+		{" ", "\\040"},
+		{"	", "\\011"},
+		{"dir with space", "dir\\040with\\040space"},
+		{"consec   spaces", "consec\\040\\040\\040spaces"},
+		{"trailingsymbol[", "trailingsymbol\\133"},
+		{" [ leadingsymbols", "\\040\\133\\040leadingsymbols"},
+		{"no_need_for_encoding", "no_need_for_encoding"},
+	}
+
+	for i := range testset {
+		got, err := Vis(testset[i].Src)
+		if err != nil {
+			t.Errorf("working with %q: %s", testset[i].Src, err)
+		}
+		if got != testset[i].Dest {
+			t.Errorf("expected %#v; got %#v", testset[i].Dest, got)
+			continue
+		}
+
+		got, err = Unvis(got)
+		if err != nil {
+			t.Errorf("working with %q: %s", testset[i].Src, err)
+			continue
+		}
+		if got != testset[i].Src {
+			t.Errorf("expected %#v; got %#v", testset[i].Src, got)
+			continue
+		}
+	}
+}
+
+// The resulting string of Vis output could potentially be four times longer than
+// the original. Vis must handle this possibility.
+func TestVisLength(t *testing.T) {
+	testString := "All work and no play makes Jack a dull boy\n"
+	for i := 0; i < 20; i++ {
+		Vis(testString)
+		testString = testString + testString
+	}
+}

walk.go

@@ -47,9 +47,13 @@ func Walk(root string, exlcudes []ExcludeFunc, keywords []string) (*DirectoryHie
 
 			// Insert a comment of the full path of the directory's name
 			if creator.curDir != nil {
+				dirname, err := creator.curDir.Path()
+				if err != nil {
+					return err
+				}
 				creator.DH.Entries = append(creator.DH.Entries, Entry{
 					Pos:  len(creator.DH.Entries),
-					Raw:  "# " + filepath.Join(creator.curDir.Path(), entryPathName),
+					Raw:  "# " + filepath.Join(dirname, entryPathName),
 					Type: CommentType,
 				})
 			} else {
@@ -147,9 +151,12 @@ func Walk(root string, exlcudes []ExcludeFunc, keywords []string) (*DirectoryHie
 				}
 			}
 		}
-
+		encodedEntryName, err := Vis(entryPathName)
+		if err != nil {
+			return err
+		}
 		e := Entry{
-			Name:   entryPathName,
+			Name:   encodedEntryName,
 			Pos:    len(creator.DH.Entries),
 			Type:   RelativeType,
 			Set:    creator.curSet,