From 4c009fc4b26cd5e6270841fe88b58495e3290054 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sat, 11 Feb 2017 21:09:06 +1100 Subject: [PATCH 02/15] README: add stub readme Signed-off-by: Aleksa Sarai --- README.md | 6 ++++++ 1 file changed, 6 insertions(+) create mode 100644 README.md diff --git a/README.md b/README.md new file mode 100644 index 0000000..aa026e1 --- /dev/null +++ b/README.md @@ -0,0 +1,6 @@ +## `govis` ## + +`govis` is a BSD-compatible `vis(3)` and `unvis(3)` encoding implementation +that is unicode aware and written in Go. None of this code comes from the +original BSD code, nor does it come from `go-mtree`'s port of said code. +Because 80s BSD code is not very nice to read. From 1e8de82690725b4dfbebc1a2a111f9fc10e67ae6 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sat, 11 Feb 2017 21:10:12 +1100 Subject: [PATCH 03/15] *: license under Apache 2.0 Signed-off-by: Aleksa Sarai --- COPYING | 202 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ README.md | 21 ++++++ 2 files changed, 223 insertions(+) create mode 100644 COPYING diff --git a/COPYING b/COPYING new file mode 100644 index 0000000..d645695 --- /dev/null +++ b/COPYING @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/README.md b/README.md index aa026e1..d56e8da 100644 --- a/README.md +++ b/README.md @@ -4,3 +4,24 @@ that is unicode aware and written in Go. None of this code comes from the original BSD code, nor does it come from `go-mtree`'s port of said code. Because 80s BSD code is not very nice to read. + +### License ### + +`govis` is licensed under the Apache 2.0 license. + +``` +govis: unicode aware vis(3) encoding implementation +Copyright (C) 2017 SUSE LLC. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +``` From cd1de45ba58cf1bdf556c8cb4da95176a31e2fac Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sun, 12 Feb 2017 04:06:20 +1100 Subject: [PATCH 04/15] unvis: implement partial unvis(3) implementation Also add some unit tests -- one of which currently fails due to ongoing design discussion about how certain escape codes should be handled. Signed-off-by: Aleksa Sarai --- flags.go | 38 ++++++++ unvis.go | 255 ++++++++++++++++++++++++++++++++++++++++++++++++++ unvis_test.go | 92 ++++++++++++++++++ 3 files changed, 385 insertions(+) create mode 100644 flags.go create mode 100644 unvis.go create mode 100644 unvis_test.go diff --git a/flags.go b/flags.go new file mode 100644 index 0000000..e98ed1b --- /dev/null +++ b/flags.go @@ -0,0 +1,38 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +type VisFlag uint + +// vis() has a variety of flags when deciding what encodings to use. While +// mtree only uses one set of flags, implementing them all is necessary in +// order to have compatibility with BSD's vis() and unvis() commands. +const ( + VisOctal VisFlag = (1 << iota) // VIS_OCTAL: Use octal \ddd format. + VisCStyle // VIS_CSTYLE: Use \[nrft0..] where appropriate. + VisSpace // VIS_SP: Also encode space. + VisTab // VIS_TAB: Also encode tab. + VisNewline // VIS_NL: Also encode newline. + VisSafe // VIS_SAFE: Encode unsafe characters. + VisNoSlash // VIS_NOSLASH: Inhibit printing '\'. + VisHTTPStyle // VIS_HTTPSTYLE: HTTP-style escape %xx. + VisGlob // VIS_GLOB: Encode glob(3) magics. + unvisEnd // UNVIS_END: Internal flag used to indicate end of parsing. + + VisWhite VisFlag = (VisSpace | VisTab | VisNewline) +) diff --git a/unvis.go b/unvis.go new file mode 100644 index 0000000..0a3dc44 --- /dev/null +++ b/unvis.go @@ -0,0 +1,255 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +import ( + "fmt" + "strconv" + "unicode/utf8" +) + +// unvisParser stores the current state of the token parser. +type unvisParser struct { + tokens []rune + idx int + flag VisFlag +} + +// Next moves the index to the next character. +func (p *unvisParser) Next() { + p.idx++ +} + +// Peek gets the current token. +func (p *unvisParser) Peek() (rune, error) { + if p.idx >= len(p.tokens) { + return utf8.RuneError, fmt.Errorf("tried to read past end of token list") + } + return p.tokens[p.idx], nil +} + +// End returns whether all of the tokens have been consumed. +func (p *unvisParser) End() bool { + return p.idx >= len(p.tokens) +} + +func newParser(input string, flag VisFlag) *unvisParser { + return &unvisParser{ + tokens: []rune(input), + idx: 0, + flag: flag, + } +} + +// While a recursive descent parser is overkill for parsing simple escape +// codes, this is IMO much easier to read than the ugly 80s coroutine code used +// by the original unvis(3) parser. Here's the EBNF for an unvis sequence: +// +// ::= ()* +// ::= ("\" ) | ("%" ) | +// ::= any rune +// ::= ("x" ) | ("M") | | +// ::= [0-9a-f] [0-9a-f] +// ::= "\" | "n" | "r" | "b" | "a" | "v" | "t" | "f" +// ::= [0-7] ([0-7] ([0-7])?)? + +func unvisPlainRune(p *unvisParser) (string, error) { + ch, err := p.Peek() + if err != nil { + return "", fmt.Errorf("plain rune: %s", ch) + } + p.Next() + return string(ch), nil +} + +func unvisEscapeCStyle(p *unvisParser) (string, error) { + ch, err := p.Peek() + if err != nil { + return "", fmt.Errorf("escape hex: %s", err) + } + + output := "" + switch ch { + case 'n': + output = "\n" + case 'r': + output = "\r" + case 'b': + output = "\b" + case 'a': + output = "\x07" + case 'v': + output = "\v" + case 't': + output = "\t" + case 'f': + output = "\f" + case 's': + output = " " + case 'E': + output = "\x1b" + case '\n': + // Hidden newline. + case '$': + // Hidden marker. + default: + // XXX: We should probably allow falling through and return "\" here... + return "", fmt.Errorf("escape cstyle: unknown escape character: %q", ch) + } + + p.Next() + return output, nil +} + +func unvisEscapeHex(p *unvisParser) (string, error) { + var output rune + + for i := 0; i < 2; i++ { + ch, err := p.Peek() + if err != nil { + return "", fmt.Errorf("escape hex: %s", err) + } + + digit, err := strconv.ParseInt(string(ch), 16, 32) + if err != nil { + return "", fmt.Errorf("escape hex: parse int: %s", err) + } + + output = (output << 4) | rune(digit) + p.Next() + } + + // TODO: We need to handle runes properly to output byte strings again. In + // particular, if rune has 0xf0 set then we know that we're currently + // decoding a messed up string. + return string(output), nil +} + +func unvisEscapeOctal(p *unvisParser) (string, error) { + var output rune + var err error + + for i := 0; i < 3; i++ { + ch, err := p.Peek() + if err != nil { + if i == 0 { + err = fmt.Errorf("escape octal[first]: %s", err) + } + break + } + + digit, err := strconv.ParseInt(string(ch), 8, 32) + if err != nil { + if i == 0 { + err = fmt.Errorf("escape octal[first]: parse int: %s", err) + } + break + } + + output = (output << 3) | rune(digit) + p.Next() + } + + // TODO: We need to handle runes properly to output byte strings again. In + // particular, if rune has 0xf0 set then we know that we're currently + // decoding a messed up string. + return string(output), err +} + +func unvisEscapeSequence(p *unvisParser) (string, error) { + ch, err := p.Peek() + if err != nil { + return "", fmt.Errorf("escape sequence: %s", err) + } + + switch ch { + case '\\': + p.Next() + return "\\", nil + + case '0', '1', '2', '3', '4', '5', '6', '7': + return unvisEscapeOctal(p) + + case 'x': + p.Next() + return unvisEscapeHex(p) + + case 'M': + // TODO + case '^': + // TODO + + default: + return unvisEscapeCStyle(p) + } + + return "", fmt.Errorf("escape sequence: unsupported sequence: %q", ch) +} + +func unvisRune(p *unvisParser) (string, error) { + ch, err := p.Peek() + if err != nil { + return "", fmt.Errorf("rune: %s", err) + } + + switch ch { + case '\\': + p.Next() + return unvisEscapeSequence(p) + + case '%': + // % HEX HEX only applies to HTTPStyle encodings. + if p.flag&VisHTTPStyle == VisHTTPStyle { + p.Next() + return unvisEscapeHex(p) + } + fallthrough + + default: + return unvisPlainRune(p) + } +} + +func unvis(p *unvisParser) (string, error) { + output := "" + for !p.End() { + ch, err := unvisRune(p) + if err != nil { + return "", fmt.Errorf("input: %s", err) + } + output += ch + } + return output, nil +} + +// Unvis takes a string formatted with the given Vis flags (though only the +// VisHTTPStyle flag is checked) and output the un-encoded version of the +// encoded string. An error is returned if any escape sequences in the input +// string were invalid. +func Unvis(input string, flag VisFlag) (string, error) { + // TODO: Check all of the VisFlag bits. + p := newParser(input, flag) + output, err := unvis(p) + if err != nil { + return "", fmt.Errorf("unvis: %s", err) + } + if !p.End() { + return "", fmt.Errorf("unvis: trailing characters at end of input") + } + return output, nil +} diff --git a/unvis_test.go b/unvis_test.go new file mode 100644 index 0000000..0b6ca04 --- /dev/null +++ b/unvis_test.go @@ -0,0 +1,92 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +import ( + "testing" +) + +func TestUnvisOctalEscape(t *testing.T) { + for _, test := range []struct { + input string + expected string + }{ + {"", ""}, + {"\\1", "\001"}, + {"\\01\\02\\3", "\001\002\003"}, + {"\\001\\023\\32", "\001\023\032"}, + {"this is a test\\0k1\\133", "this is a test\000k1\133"}, + {"\\170YET\\01another test\\1\\\\82", "\170YET\001another test\001\\82"}, + {"\\177MORE tests\\09a", "\177MORE tests\x009a"}, + {"\\\\710more\\1215testing", "\\710more\1215testing"}, + } { + got, err := Unvis(test.input, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %q", test.input, err) + continue + } + if got != test.expected { + t.Errorf("expected unvis(%q) = %q, got %q", test.input, test.expected, got) + } + } +} + +func TestUnvisHexEscape(t *testing.T) { + for _, test := range []struct { + input string + expected string + }{ + {"", ""}, + {"\\x01", "\x01"}, + {"\\x01\\x02\\x7a", "\x01\x02\x7a"}, + {"this is a test\\x13\\x52\\x6f", "this is a test\x13\x52\x6f"}, + {"\\x170YET\\x01a\\x22nother test\\x11", "\x170YET\x01a\x22nother test\x11"}, + {"\\\\x007more\\\\x215testing", "\\x007more\\x215testing"}, + // Make sure that decoding unicode works properly. + {"\\xf0\\x9f\\x95\\xb4", "\U0001f574"}, + } { + got, err := Unvis(test.input, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %q", test.input, err) + continue + } + if got != test.expected { + t.Errorf("expected unvis(%q) = %q, got %q", test.input, test.expected, got) + } + } +} + +func TestUnvisUnicode(t *testing.T) { + // Ensure that unicode strings are not messed up by Unvis. + for _, test := range []string{ + "", + "this.is.a.normal_string", + "AC_Raíz_Certicámara_S.A..pem", + "NetLock_Arany_=Class_Gold=_Főtanúsítvány.pem", + "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem", + } { + got, err := Unvis(test, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %s", test, err) + continue + } + if got != test { + t.Errorf("expected %q to be unchanged, got %q", test, got) + } + } +} From 44391840b6581322304d03c9ac898601a6bd2c5a Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sun, 12 Feb 2017 04:07:35 +1100 Subject: [PATCH 05/15] vis: partial vis(3) port This is a stopgap while I figure out how I should go about implementing vis(3). It's also important to have some vis(3) implementation so I can do integration tests on round-trips. Signed-off-by: Aleksa Sarai --- vis.go | 162 ++++++++++++++++++++++++++++++++++++++++++++++++++++ vis_test.go | 54 ++++++++++++++++++ 2 files changed, 216 insertions(+) create mode 100644 vis.go create mode 100644 vis_test.go diff --git a/vis.go b/vis.go new file mode 100644 index 0000000..d599ecb --- /dev/null +++ b/vis.go @@ -0,0 +1,162 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +import ( + "fmt" + "unicode" + "unicode/utf8" +) + +func isunsafe(ch rune) bool { + return ch == '\b' || ch == '\007' || ch == '\r' +} + +func isglob(ch rune) bool { + return ch == '*' || ch == '?' || ch == '[' || ch == '#' +} + +func ishttp(ch rune) bool { + return unicode.IsDigit(ch) || unicode.IsLetter(ch) || + // Safe characters. + ch == '$' || ch == '-' || ch == '_' || ch == '.' || ch == '+' || + // Extra characters. + ch == '!' || ch == '*' || ch == '\'' || ch == '(' || + ch == ')' || ch == ',' +} + +func mapRuneBytes(ch rune, fn func(byte) string) string { + bytes := make([]byte, utf8.RuneLen(ch)) + n := utf8.EncodeRune(bytes, ch) + + mapped := "" + for i := 0; i < n; i++ { + mapped += fn(bytes[i]) + } + return mapped +} + +// vis converts a single rune into its encoding, ensuring that it is "safe" +// (for some definition of safe). Note that some visual characters (such as +// accented characters or similar things) can be made up of several runes -- in +// order to maintain my sanity Vis() makes no attempt to handle such cases +// specially. +func vis(ch rune, flag VisFlag) (string, error) { + // XXX: Currently we are just allowing regular multi-byte characters such + // as accents and so on to be passed through without encoding. Is this + // really the best idea? In order to maintain compatibility with + // vis(3) such that an older unvis(3) will do the right thing maybe we + // should only output 7-bit ASCII? I'm not sure. + + if flag&VisHTTPStyle == VisHTTPStyle { + // This is described in RFC 1808. + if !ishttp(ch) { + return mapRuneBytes(ch, func(b byte) string { + return fmt.Sprintf("%.2X", b) + }), nil + } + } + + // Handle all "ordinary" characters which don't need to be encoded. + if !(flag&VisGlob == VisGlob && isglob(ch)) && + ((unicode.IsGraphic(ch) && !unicode.IsSpace(ch)) || + (flag&VisSpace == 0 && ch == ' ') || + (flag&VisTab == 0 && ch == '\t') || + (flag&VisNewline == 0 && ch == '\n') || + (flag&VisSafe == VisSafe && isunsafe(ch))) { + enc := string(ch) + if ch == '\\' && flag&VisNoSlash == 0 { + enc += "\\" + } + return enc, nil + } + + if flag&VisCStyle == VisCStyle { + switch ch { + case '\n': + return "\\n", nil + case '\r': + return "\\r", nil + case '\b': + return "\\b", nil + case '\a': + return "\\a", nil + case '\v': + return "\\v", nil + case '\t': + return "\\t", nil + case '\f': + return "\\f", nil + case 0: + // TODO: Handle isoctal properly. + return "\\000", nil + } + } + + // TODO: ch & 0177 is not implemented... + if flag&VisOctal == VisOctal || unicode.IsGraphic(ch) { + return mapRuneBytes(ch, func(b byte) string { + return fmt.Sprintf("\\%.3o", b) + }), nil + } + + return mapRuneBytes(ch, func(b byte) string { + enc := "" + if flag&VisNoSlash == 0 { + enc += "\\" + } + + // This logic is stolen from cvis, I don't understand any of it. + if b&0200 != 0 { + b &= 0177 + enc += "M" + } + if unicode.IsControl(rune(b)) { + enc += "^" + if b == 0177 { + enc += "?" + } else { + enc += string(b + '@') + } + } else { + enc += fmt.Sprintf("-%s", b) + } + + return enc + }), nil +} + +// Vis encodes the provided string to a BSD-compatible encoding using BSD's +// vis() flags. However, it will correctly handle multi-byte encoding (which is +// not done properly by BSD's vis implementation). +func Vis(src string, flag VisFlag) (string, error) { + if !utf8.ValidString(src) { + return "", fmt.Errorf("vis: input string is invalid utf8 literal") + } + + output := "" + for _, ch := range src { + encodedCh, err := vis(ch, flag) + if err != nil { + return "", err + } + output += encodedCh + } + + return output, nil +} diff --git a/vis_test.go b/vis_test.go new file mode 100644 index 0000000..fbecacb --- /dev/null +++ b/vis_test.go @@ -0,0 +1,54 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +import ( + "testing" +) + +func TestVisUnchanged(t *testing.T) { + for _, test := range []string{ + "helloworld", + "THIS_IS_A_TEST1234", + "SomeEncodingsAreCool", + "AC_Raíz_Certicámara_S.A..pem", + } { + enc, err := Vis(test, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error with %q: %s", test, err) + } + if enc != test { + t.Errorf("expected encoding of %q to be unchanged, got %q", test, enc) + } + } +} + +func TestVisChanged(t *testing.T) { + for _, test := range []string{ + "hello world", + "THIS\\IS_A_TEST1234", + } { + enc, err := Vis(test, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error with %q: %s", test, err) + } + if enc == test { + t.Errorf("expected encoding of %q to be changed") + } + } +} From 3b18d38388fc8b98033fa093d3df14732caffcef Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sun, 12 Feb 2017 04:08:21 +1100 Subject: [PATCH 06/15] govis: add integration tests Signed-off-by: Aleksa Sarai --- govis_test.go | 52 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 52 insertions(+) create mode 100644 govis_test.go diff --git a/govis_test.go b/govis_test.go new file mode 100644 index 0000000..d236f85 --- /dev/null +++ b/govis_test.go @@ -0,0 +1,52 @@ +/* + * govis: unicode aware vis(3) encoding implementation + * Copyright (C) 2017 SUSE LLC. + * + * Licensed under the Apache License, Version 2.0 (the "License"); + * you may not use this file except in compliance with the License. + * You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package govis + +import ( + "testing" +) + +const DefaultVisFlags = VisWhite | VisOctal | VisGlob + +func TestVisUnvis(t *testing.T) { + // Round-trip testing. + for _, test := range []string{ + "", + "this.is.a.normal_string", + "AC_Raíz_Certicámara_S.A..pem", + "NetLock_Arany_=Class_Gold=_Főtanúsítvány.pem", + "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem", + "hello world [ this string needs=enco ding! ]", + "even \n more encoding necessary\a\a ", + "\024 <-- some more weird characters --> 你好，世界", + } { + enc, err := Vis(test, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing vis(%q): %s", test, err) + continue + } + dec, err := Unvis(enc, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %s", enc, err) + continue + } + if dec != test { + t.Errorf("roundtrip failed: unvis(vis(%q) = %q) = %q", test, enc, dec) + } + } +} From 35708696fef4bc564916e5792417c255b8e4716a Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Sun, 12 Feb 2017 23:12:46 +1100 Subject: [PATCH 07/15] unvis: implement proper '\xff' and '\377' escape handling In particular, previously such escape handling would break because we would attempt to encode characters >0x7f as runes -- which would then result in escapes that want to encode multi-byte characters breaking. There's still some work necessary in Vis() to make it act sanely when it comes to arbitrary bit streams. Not to mention that we need to figure out what we actually want to do there... Signed-off-by: Aleksa Sarai --- govis_test.go | 39 +++++++++++++++++++ unvis.go | 106 +++++++++++++++++++++----------------------------- unvis_test.go | 24 +++++++++++- 3 files changed, 107 insertions(+), 62 deletions(-) diff --git a/govis_test.go b/govis_test.go index d236f85..244ae23 100644 --- a/govis_test.go +++ b/govis_test.go @@ -18,6 +18,7 @@ package govis import ( + "bytes" "testing" ) @@ -34,6 +35,7 @@ func TestVisUnvis(t *testing.T) { "hello world [ this string needs=enco ding! ]", "even \n more encoding necessary\a\a ", "\024 <-- some more weird characters --> 你好，世界", + "\\xff\\n double encoding is also great fun \\x", } { enc, err := Vis(test, DefaultVisFlags) if err != nil { @@ -50,3 +52,40 @@ func TestVisUnvis(t *testing.T) { } } } + +func TestByteStrings(t *testing.T) { + // It's important to make sure that we don't mess around with the layout of + // bytes when doing a round-trip. Otherwise we risk outputting visually + // identical but bit-stream non-identical strings (causing much confusion + // when trying to access such files). + + for _, test := range [][]byte{ + []byte("This is a man in business suit levitating: \U0001f574"), + {0x7f, 0x17, 0x01, 0x33}, + // TODO: Test arbitrary byte streams like the one below. Currently this + // fails because Vis() is messing around with it (converting it + // to a rune and spacing it out). + //{'\xef', '\xae', 'h', '\077', 'k'}, + } { + testString := string(test) + enc, err := Vis(testString, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing vis(%q): %s", test, err) + continue + } + dec, err := Unvis(enc, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %s", enc, err) + continue + } + decBytes := []byte(dec) + + if dec != testString { + t.Errorf("roundtrip failed [string comparison]: unvis(vis(%q) = %q) = %q", test, enc, dec) + } + if !bytes.Equal(decBytes, test) { + t.Errorf("roundtrip failed [byte comparison]: unvis(vis(%q) = %q) = %q", test, enc, dec) + } + } + +} diff --git a/unvis.go b/unvis.go index 0a3dc44..9d59b9e 100644 --- a/unvis.go +++ b/unvis.go @@ -20,7 +20,7 @@ package govis import ( "fmt" "strconv" - "unicode/utf8" + "unicode" ) // unvisParser stores the current state of the token parser. @@ -38,7 +38,7 @@ func (p *unvisParser) Next() { // Peek gets the current token. func (p *unvisParser) Peek() (rune, error) { if p.idx >= len(p.tokens) { - return utf8.RuneError, fmt.Errorf("tried to read past end of token list") + return unicode.ReplacementChar, fmt.Errorf("tried to read past end of token list") } return p.tokens[p.idx], nil } @@ -68,19 +68,26 @@ func newParser(input string, flag VisFlag) *unvisParser { // ::= "\" | "n" | "r" | "b" | "a" | "v" | "t" | "f" // ::= [0-7] ([0-7] ([0-7])?)? -func unvisPlainRune(p *unvisParser) (string, error) { +func unvisPlainRune(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { - return "", fmt.Errorf("plain rune: %s", ch) + return nil, fmt.Errorf("plain rune: %s", ch) } p.Next() - return string(ch), nil + + // XXX: Maybe we should not be converting to runes and then back to strings + // here. Are we sure that the byte-for-byte representation is the + // same? If the bytes change, then using these strings for paths will + // break... + + str := string(ch) + return []byte(str), nil } -func unvisEscapeCStyle(p *unvisParser) (string, error) { +func unvisEscapeCStyle(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { - return "", fmt.Errorf("escape hex: %s", err) + return nil, fmt.Errorf("escape hex: %s", err) } output := "" @@ -109,85 +116,62 @@ func unvisEscapeCStyle(p *unvisParser) (string, error) { // Hidden marker. default: // XXX: We should probably allow falling through and return "\" here... - return "", fmt.Errorf("escape cstyle: unknown escape character: %q", ch) + return nil, fmt.Errorf("escape cstyle: unknown escape character: %q", ch) } p.Next() - return output, nil + return []byte(output), nil } -func unvisEscapeHex(p *unvisParser) (string, error) { - var output rune +func unvisEscapeDigits(p *unvisParser, base int, force bool) ([]byte, error) { + var code int - for i := 0; i < 2; i++ { + for i := int(0xFF); i > 0; i /= base { ch, err := p.Peek() if err != nil { - return "", fmt.Errorf("escape hex: %s", err) + if !force && i != 0xFF { + break + } + return nil, fmt.Errorf("escape base %d: %s", base, err) } - digit, err := strconv.ParseInt(string(ch), 16, 32) + digit, err := strconv.ParseInt(string(ch), base, 8) if err != nil { - return "", fmt.Errorf("escape hex: parse int: %s", err) + if !force && i != 0xFF { + break + } + return nil, fmt.Errorf("escape base %d: could not parse digit: %s", base, err) } - output = (output << 4) | rune(digit) + code = (code * base) + int(digit) p.Next() } - // TODO: We need to handle runes properly to output byte strings again. In - // particular, if rune has 0xf0 set then we know that we're currently - // decoding a messed up string. - return string(output), nil -} - -func unvisEscapeOctal(p *unvisParser) (string, error) { - var output rune - var err error - - for i := 0; i < 3; i++ { - ch, err := p.Peek() - if err != nil { - if i == 0 { - err = fmt.Errorf("escape octal[first]: %s", err) - } - break - } - - digit, err := strconv.ParseInt(string(ch), 8, 32) - if err != nil { - if i == 0 { - err = fmt.Errorf("escape octal[first]: parse int: %s", err) - } - break - } - - output = (output << 3) | rune(digit) - p.Next() + if code > unicode.MaxLatin1 { + return nil, fmt.Errorf("escape base %d: code %q outside latin-1 encoding", base, code) } - // TODO: We need to handle runes properly to output byte strings again. In - // particular, if rune has 0xf0 set then we know that we're currently - // decoding a messed up string. - return string(output), err + char := byte(code & 0xFF) + return []byte{char}, nil } -func unvisEscapeSequence(p *unvisParser) (string, error) { +func unvisEscapeSequence(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { - return "", fmt.Errorf("escape sequence: %s", err) + return nil, fmt.Errorf("escape sequence: %s", err) } switch ch { case '\\': p.Next() - return "\\", nil + return []byte("\\"), nil case '0', '1', '2', '3', '4', '5', '6', '7': - return unvisEscapeOctal(p) + return unvisEscapeDigits(p, 8, false) case 'x': p.Next() - return unvisEscapeHex(p) + return unvisEscapeDigits(p, 16, true) case 'M': // TODO @@ -198,13 +182,13 @@ func unvisEscapeSequence(p *unvisParser) (string, error) { return unvisEscapeCStyle(p) } - return "", fmt.Errorf("escape sequence: unsupported sequence: %q", ch) + return nil, fmt.Errorf("escape sequence: unsupported sequence: %q", ch) } -func unvisRune(p *unvisParser) (string, error) { +func unvisRune(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { - return "", fmt.Errorf("rune: %s", err) + return nil, fmt.Errorf("rune: %s", err) } switch ch { @@ -216,7 +200,7 @@ func unvisRune(p *unvisParser) (string, error) { // % HEX HEX only applies to HTTPStyle encodings. if p.flag&VisHTTPStyle == VisHTTPStyle { p.Next() - return unvisEscapeHex(p) + return unvisEscapeDigits(p, 16, true) } fallthrough @@ -226,15 +210,15 @@ func unvisRune(p *unvisParser) (string, error) { } func unvis(p *unvisParser) (string, error) { - output := "" + var output []byte for !p.End() { ch, err := unvisRune(p) if err != nil { return "", fmt.Errorf("input: %s", err) } - output += ch + output = append(output, ch...) } - return output, nil + return string(output), nil } // Unvis takes a string formatted with the given Vis flags (though only the diff --git a/unvis_test.go b/unvis_test.go index 0b6ca04..66d60eb 100644 --- a/unvis_test.go +++ b/unvis_test.go @@ -21,6 +21,20 @@ import ( "testing" ) +func TestUnvisError(t *testing.T) { + for _, test := range []string{ + // Octal escape codes allow you to specify invalid byte values. + "\\777", + "\\420\\322\\455", + "\\652\\233", + } { + got, err := Unvis(test, DefaultVisFlags) + if err == nil { + t.Errorf("expected unvis(%q) to give an error, got %q", test, got) + } + } +} + func TestUnvisOctalEscape(t *testing.T) { for _, test := range []struct { input string @@ -34,6 +48,11 @@ func TestUnvisOctalEscape(t *testing.T) { {"\\170YET\\01another test\\1\\\\82", "\170YET\001another test\001\\82"}, {"\\177MORE tests\\09a", "\177MORE tests\x009a"}, {"\\\\710more\\1215testing", "\\710more\1215testing"}, + // Make sure that decoding unicode works properly, when it's been encoded as single bytes. + {"\\360\\237\\225\\264", "\U0001f574"}, + {"T\\303\\234B\\304\\260TAK_UEKAE_K\\303\\266k_Sertifika_Hizmet_Sa\\304\\237lay\\304\\261c\\304\\261s\\304\\261_-_S\\303\\274r\\303\\274m_3.pem", "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem"}, + // Some invalid characters... + {"\\377\\2\\225\\264", "\xff\x02\x95\xb4"}, } { got, err := Unvis(test.input, DefaultVisFlags) if err != nil { @@ -57,8 +76,11 @@ func TestUnvisHexEscape(t *testing.T) { {"this is a test\\x13\\x52\\x6f", "this is a test\x13\x52\x6f"}, {"\\x170YET\\x01a\\x22nother test\\x11", "\x170YET\x01a\x22nother test\x11"}, {"\\\\x007more\\\\x215testing", "\\x007more\\x215testing"}, - // Make sure that decoding unicode works properly. + // Make sure that decoding unicode works properly, when it's been encoded as single bytes. {"\\xf0\\x9f\\x95\\xb4", "\U0001f574"}, + {"T\\xc3\\x9cB\\xc4\\xb0TAK_UEKAE_K\\xc3\\xb6k_Sertifika_Hizmet_Sa\\xc4\\x9flay\\xc4\\xb1c\\xc4\\xb1s\\xc4\\xb1_-_S\\xc3\\xbcr\\xc3\\xbcm_3.pem", "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem"}, + // Some invalid characters... + {"\\xff\\x02\\x95\\xb4", "\xff\x02\x95\xb4"}, } { got, err := Unvis(test.input, DefaultVisFlags) if err != nil { From 7b16f3a307a28ce24fb478671edf378847cc95f2 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Mon, 13 Feb 2017 07:30:24 +1100 Subject: [PATCH 08/15] travis: add .travis.yml Signed-off-by: Aleksa Sarai --- .travis.yml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) create mode 100644 .travis.yml diff --git a/.travis.yml b/.travis.yml new file mode 100644 index 0000000..ff3b78c --- /dev/null +++ b/.travis.yml @@ -0,0 +1,28 @@ +# govis: unicode aware vis(3) encoding implementation +# Copyright (C) 2017 SUSE LLC. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +language: go + +notifications: + email: false + +go: + - 1.x + - 1.6.x + - 1.7.x + - master + +script: + - go test -v ./... From de223ffc920aa0db838d1bdf00a73e6f37bb4888 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Mon, 13 Feb 2017 07:29:57 +1100 Subject: [PATCH 09/15] unvis: implement meta and ctrl characters ('\M__' and '\^_') While these characters are really weird to handle, here is a fairly simple implementation that need some more testing (and a proper secondary source to compare against). Signed-off-by: Aleksa Sarai --- unvis.go | 67 +++++++++++++++++++++++++++++++++++++++++++++++---- unvis_test.go | 52 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 114 insertions(+), 5 deletions(-) diff --git a/unvis.go b/unvis.go index 9d59b9e..f8dd6b1 100644 --- a/unvis.go +++ b/unvis.go @@ -63,9 +63,12 @@ func newParser(input string, flag VisFlag) *unvisParser { // ::= ()* // ::= ("\" ) | ("%" ) | // ::= any rune -// ::= ("x" ) | ("M") | | -// ::= [0-9a-f] [0-9a-f] +// ::= ("x" ) | ("M" ) | ("^" | +// ::= ("-" ) | ("^" ) +// ::= any rune +// ::= "?" | any rune // ::= "\" | "n" | "r" | "b" | "a" | "v" | "t" | "f" +// ::= [0-9a-f] [0-9a-f] // ::= [0-7] ([0-7] ([0-7])?)? func unvisPlainRune(p *unvisParser) ([]byte, error) { @@ -155,6 +158,57 @@ func unvisEscapeDigits(p *unvisParser, base int, force bool) ([]byte, error) { return []byte{char}, nil } +func unvisEscapeCtrl(p *unvisParser, mask byte) ([]byte, error) { + ch, err := p.Peek() + if err != nil { + return nil, fmt.Errorf("escape ctrl: %s", err) + } + if ch > unicode.MaxLatin1 { + return nil, fmt.Errorf("escape ctrl: code %q outside latin-1 encoding", ch) + } + + char := byte(ch) & 0x1f + if ch == '?' { + char = 0x7f + } + + p.Next() + return []byte{mask | char}, nil +} + +func unvisEscapeMeta(p *unvisParser) ([]byte, error) { + ch, err := p.Peek() + if err != nil { + return nil, fmt.Errorf("escape meta: %s", err) + } + + mask := byte(0x80) + + switch ch { + case '^': + // The same as "\^..." except we apply a mask. + p.Next() + return unvisEscapeCtrl(p, mask) + + case '-': + p.Next() + + ch, err := p.Peek() + if err != nil { + return nil, fmt.Errorf("escape meta1: %s", err) + } + if ch > unicode.MaxLatin1 { + return nil, fmt.Errorf("escape meta1: code %q outside latin-1 encoding", ch) + } + + // Add mask to character. + p.Next() + return []byte{mask | byte(ch)}, nil + } + + return nil, fmt.Errorf("escape meta: unknown escape char: %s", err) +} + func unvisEscapeSequence(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { @@ -173,10 +227,13 @@ func unvisEscapeSequence(p *unvisParser) ([]byte, error) { p.Next() return unvisEscapeDigits(p, 16, true) - case 'M': - // TODO case '^': - // TODO + p.Next() + return unvisEscapeCtrl(p, 0x00) + + case 'M': + p.Next() + return unvisEscapeMeta(p) default: return unvisEscapeCStyle(p) diff --git a/unvis_test.go b/unvis_test.go index 66d60eb..44e0a1a 100644 --- a/unvis_test.go +++ b/unvis_test.go @@ -35,6 +35,58 @@ func TestUnvisError(t *testing.T) { } } +func TestUnvisCStyleEscape(t *testing.T) { + for _, test := range []struct { + input string + expected string + }{ + {"", ""}, + {"\\n\\v\\t\\s", "\n\v\t "}, + {"\\\\n\\tt", "\\n\tt"}, + {"\\b", "\b"}, + {"\\r\\b\\n", "\r\b\n"}, + {"\\a\\a\\b", "\x07\x07\b"}, + {"\\f\\s\\E", "\f \x1b"}, + // Hidden markers. They actually aren't generated by vis(3) but for + // some reason, they're supported... + {"test\\\ning", "testing"}, + {"test\\$\\$ing", "testing"}, + } { + got, err := Unvis(test.input, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %q", test.input, err) + continue + } + if got != test.expected { + t.Errorf("expected unvis(%q) = %q, got %q", test.input, test.expected, got) + } + } +} + +func TestUnvisMetaEscape(t *testing.T) { + for _, test := range []struct { + input string + expected string + }{ + {"", ""}, + {"\\M^ ?\\^ ", "\x80?\x00"}, + {"\\M- ?\\^?", "\xa0?\x7f"}, + {"\\M-x butterfly\\M^?", "\xf8 butterfly\xff"}, + {"\\M^X steady-hand \\^& needle", "\x98 steady-hand \x06 needle"}, + // TODO: Add some more of these tests, but I need to have some + // secondary source to verify these outputs properly. + } { + got, err := Unvis(test.input, DefaultVisFlags) + if err != nil { + t.Errorf("unexpected error doing unvis(%q): %q", test.input, err) + continue + } + if got != test.expected { + t.Errorf("expected unvis(%q) = %q, got %q", test.input, test.expected, got) + } + } +} + func TestUnvisOctalEscape(t *testing.T) { for _, test := range []struct { input string From c2a9f1a56d42488c3c0542f1fbe4e46721e324b7 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Tue, 14 Feb 2017 02:11:11 +1100 Subject: [PATCH 10/15] flags -> govis Signed-off-by: Aleksa Sarai --- flags.go => govis.go | 1 - 1 file changed, 1 deletion(-) rename flags.go => govis.go (94%) diff --git a/flags.go b/govis.go similarity index 94% rename from flags.go rename to govis.go index e98ed1b..2e6b22f 100644 --- a/flags.go +++ b/govis.go @@ -32,7 +32,6 @@ const ( VisNoSlash // VIS_NOSLASH: Inhibit printing '\'. VisHTTPStyle // VIS_HTTPSTYLE: HTTP-style escape %xx. VisGlob // VIS_GLOB: Encode glob(3) magics. - unvisEnd // UNVIS_END: Internal flag used to indicate end of parsing. VisWhite VisFlag = (VisSpace | VisTab | VisNewline) ) From 4045484afbff2b52ad7d6fbc2064ac7289287648 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Tue, 14 Feb 2017 03:52:52 +1100 Subject: [PATCH 11/15] vis: rewrite to use byte slices This results in all multi-byte characters to be encoded in a way that naive unvis(3) implementations will not bork up the encoding. In addition, it also ensures that the output of Vis will always be ASCII *only*. Also test far more cases in *_test.go when it comes to different flags, and do far more tests to ensure that the output of Vis() makes sense. These outputs come directly from vis(3) and so are useful regression tests to ensure that the handling of Vis() is identical to the original. Signed-off-by: Aleksa Sarai --- govis.go | 19 +++---- vis.go | 145 +++++++++++++++++++++++++++++----------------------- vis_test.go | 89 +++++++++++++++++++++++++++++--- 3 files changed, 171 insertions(+), 82 deletions(-) diff --git a/govis.go b/govis.go index 2e6b22f..1e88eb1 100644 --- a/govis.go +++ b/govis.go @@ -23,15 +23,16 @@ type VisFlag uint // mtree only uses one set of flags, implementing them all is necessary in // order to have compatibility with BSD's vis() and unvis() commands. const ( - VisOctal VisFlag = (1 << iota) // VIS_OCTAL: Use octal \ddd format. - VisCStyle // VIS_CSTYLE: Use \[nrft0..] where appropriate. - VisSpace // VIS_SP: Also encode space. - VisTab // VIS_TAB: Also encode tab. - VisNewline // VIS_NL: Also encode newline. - VisSafe // VIS_SAFE: Encode unsafe characters. - VisNoSlash // VIS_NOSLASH: Inhibit printing '\'. - VisHTTPStyle // VIS_HTTPSTYLE: HTTP-style escape %xx. - VisGlob // VIS_GLOB: Encode glob(3) magics. + VisOctal VisFlag = (1 << iota) // VIS_OCTAL: Use octal \ddd format. + VisCStyle // VIS_CSTYLE: Use \[nrft0..] where appropriate. + VisSpace // VIS_SP: Also encode space. + VisTab // VIS_TAB: Also encode tab. + VisNewline // VIS_NL: Also encode newline. + VisSafe // VIS_SAFE: Encode unsafe characters. + VisNoSlash // VIS_NOSLASH: Inhibit printing '\'. + VisHTTPStyle // VIS_HTTPSTYLE: HTTP-style escape %xx. + VisGlob // VIS_GLOB: Encode glob(3) magics. + visMask VisFlag = (1 << iota) - 1 // Mask of all flags. VisWhite VisFlag = (VisSpace | VisTab | VisNewline) ) diff --git a/vis.go b/vis.go index d599ecb..140556a 100644 --- a/vis.go +++ b/vis.go @@ -20,7 +20,6 @@ package govis import ( "fmt" "unicode" - "unicode/utf8" ) func isunsafe(ch rune) bool { @@ -31,7 +30,15 @@ func isglob(ch rune) bool { return ch == '*' || ch == '?' || ch == '[' || ch == '#' } +// ishttp is defined by RFC 1808. func ishttp(ch rune) bool { + // RFC1808 does not really consider characters outside of ASCII, so just to + // be safe always treat characters outside the ASCII character set as "not + // HTTP". + if ch > unicode.MaxASCII { + return false + } + return unicode.IsDigit(ch) || unicode.IsLetter(ch) || // Safe characters. ch == '$' || ch == '-' || ch == '_' || ch == '.' || ch == '+' || @@ -40,54 +47,56 @@ func ishttp(ch rune) bool { ch == ')' || ch == ',' } -func mapRuneBytes(ch rune, fn func(byte) string) string { - bytes := make([]byte, utf8.RuneLen(ch)) - n := utf8.EncodeRune(bytes, ch) - - mapped := "" - for i := 0; i < n; i++ { - mapped += fn(bytes[i]) - } - return mapped +func isgraph(ch rune) bool { + return unicode.IsGraphic(ch) && !unicode.IsSpace(ch) && ch <= unicode.MaxASCII } -// vis converts a single rune into its encoding, ensuring that it is "safe" -// (for some definition of safe). Note that some visual characters (such as -// accented characters or similar things) can be made up of several runes -- in -// order to maintain my sanity Vis() makes no attempt to handle such cases -// specially. -func vis(ch rune, flag VisFlag) (string, error) { - // XXX: Currently we are just allowing regular multi-byte characters such - // as accents and so on to be passed through without encoding. Is this - // really the best idea? In order to maintain compatibility with - // vis(3) such that an older unvis(3) will do the right thing maybe we - // should only output 7-bit ASCII? I'm not sure. +// vis converts a single *byte* into its encoding. While Go supports the +// concept of runes (and thus native utf-8 parsing), in order to make sure that +// the bit-stream will be completely maintained through an Unvis(Vis(...)) +// round-trip. The downside is that Vis() will never output unicode -- but on +// the plus side this is actually a benefit on the encoding side (it will +// always work with the simple unvis(3) implementation). It also means that we +// don't have to worry about different multi-byte encodings. +func vis(b byte, flag VisFlag) (string, error) { + // Treat the single-byte character as a rune. + ch := rune(b) + // XXX: This is quite a horrible thing to support. if flag&VisHTTPStyle == VisHTTPStyle { - // This is described in RFC 1808. if !ishttp(ch) { - return mapRuneBytes(ch, func(b byte) string { - return fmt.Sprintf("%.2X", b) - }), nil + return "%" + fmt.Sprintf("%.2X", ch), nil } } - // Handle all "ordinary" characters which don't need to be encoded. - if !(flag&VisGlob == VisGlob && isglob(ch)) && - ((unicode.IsGraphic(ch) && !unicode.IsSpace(ch)) || - (flag&VisSpace == 0 && ch == ' ') || - (flag&VisTab == 0 && ch == '\t') || - (flag&VisNewline == 0 && ch == '\n') || - (flag&VisSafe == VisSafe && isunsafe(ch))) { - enc := string(ch) + // Figure out if the character doesn't need to be encoded. Effectively, we + // encode most "normal" (graphical) characters as themselves unless we have + // been specifically asked not to. Note though that we *ALWAYS* encode + // everything outside ASCII. + // TODO: Switch this to much more logical code. + + if ch > unicode.MaxASCII { + /* ... */ + } else if flag&VisGlob == VisGlob && isglob(ch) { + /* ... */ + } else if isgraph(ch) || + (flag&VisSpace != VisSpace && ch == ' ') || + (flag&VisTab != VisTab && ch == '\t') || + (flag&VisNewline != VisNewline && ch == '\n') || + (flag&VisSafe != 0 && isunsafe(ch)) { + + encoded := string(ch) if ch == '\\' && flag&VisNoSlash == 0 { - enc += "\\" + encoded += "\\" } - return enc, nil + return encoded, nil } + // Try to use C-style escapes first. if flag&VisCStyle == VisCStyle { switch ch { + case ' ': + return "\\s", nil case '\n': return "\\n", nil case '\r': @@ -102,55 +111,61 @@ func vis(ch rune, flag VisFlag) (string, error) { return "\\t", nil case '\f': return "\\f", nil - case 0: - // TODO: Handle isoctal properly. + case '\x00': + // Output octal just to be safe. return "\\000", nil } } - // TODO: ch & 0177 is not implemented... - if flag&VisOctal == VisOctal || unicode.IsGraphic(ch) { - return mapRuneBytes(ch, func(b byte) string { - return fmt.Sprintf("\\%.3o", b) - }), nil + // For graphical characters we generate octal output (and also if it's + // being forced by the caller's flags). Also spaces should always be + // encoded as octal. + if flag&VisOctal == VisOctal || isgraph(ch) || ch&0x7f == ' ' { + // Always output three-character octal just to be safe. + return fmt.Sprintf("\\%.3o", ch), nil } - return mapRuneBytes(ch, func(b byte) string { - enc := "" - if flag&VisNoSlash == 0 { - enc += "\\" - } + // Now we have to output meta or ctrl escapes. As far as I can tell, this + // is not actually defined by any standard -- so this logic is basically + // copied from the original vis(3) implementation. Hopefully nobody + // actually relies on this (octal and hex are better). - // This logic is stolen from cvis, I don't understand any of it. - if b&0200 != 0 { - b &= 0177 - enc += "M" - } - if unicode.IsControl(rune(b)) { - enc += "^" - if b == 0177 { - enc += "?" - } else { - enc += string(b + '@') - } + encoded := "" + if flag&VisNoSlash == 0 { + encoded += "\\" + } + + // Meta characters have 0x80 set, but are otherwise identical to control + // characters. + if b&0x80 != 0 { + b &= 0x7f + encoded += "M" + } + + if unicode.IsControl(rune(b)) { + encoded += "^" + if b == 0x7f { + encoded += "?" } else { - enc += fmt.Sprintf("-%s", b) + encoded += fmt.Sprintf("%c", b+'@') } + } else { + encoded += fmt.Sprintf("-%c", b) + } - return enc - }), nil + return encoded, nil } // Vis encodes the provided string to a BSD-compatible encoding using BSD's // vis() flags. However, it will correctly handle multi-byte encoding (which is // not done properly by BSD's vis implementation). func Vis(src string, flag VisFlag) (string, error) { - if !utf8.ValidString(src) { - return "", fmt.Errorf("vis: input string is invalid utf8 literal") + if flag&visMask != flag { + return "", fmt.Errorf("vis: flag %q contains unknown or unsupported flags", flag) } output := "" - for _, ch := range src { + for _, ch := range []byte(src) { encodedCh, err := vis(ch, flag) if err != nil { return "", err diff --git a/vis_test.go b/vis_test.go index fbecacb..e4ef083 100644 --- a/vis_test.go +++ b/vis_test.go @@ -22,18 +22,90 @@ import ( ) func TestVisUnchanged(t *testing.T) { - for _, test := range []string{ - "helloworld", - "THIS_IS_A_TEST1234", - "SomeEncodingsAreCool", - "AC_Raíz_Certicámara_S.A..pem", + for _, test := range []struct { + input string + flag VisFlag + }{ + {"", DefaultVisFlags}, + {"helloworld", DefaultVisFlags}, + {"THIS_IS_A_TEST1234", DefaultVisFlags}, + {"SomeEncodingsAreCool", DefaultVisFlags}, + {"spaces are totally safe", DefaultVisFlags &^ VisSpace}, + {"tabs\tare\talso\tsafe!!", DefaultVisFlags &^ VisTab}, + {"just\a\atrustme\r\b\b!!", DefaultVisFlags | VisSafe}, } { - enc, err := Vis(test, DefaultVisFlags) + enc, err := Vis(test.input, test.flag) if err != nil { t.Errorf("unexpected error with %q: %s", test, err) } - if enc != test { - t.Errorf("expected encoding of %q to be unchanged, got %q", test, enc) + if enc != test.input { + t.Errorf("expected encoding of %q (flag=%q) to be unchanged, got %q", test.input, test.flag, enc) + } + } +} + +func TestVisFlags(t *testing.T) { + for _, test := range []struct { + input string + output string + flag VisFlag + }{ + // Default + {"AC_Ra\u00edz_Certic\u00e1mara_S.A..pem", "AC_Ra\\M-C\\M--z_Certic\\M-C\\M-!mara_S.A..pem", 0}, + {"z^i3i$\u00d3\u008anqgh5/t\u00e5<86>\u00b2kzla\\e^lv\u00df\u0093nv\u00df\u00aea|3}\u00d8\u0088\u00d6\u0084", `z^i3i$\M-C\M^S\M-B\M^Jnqgh5/t\M-C\M-%<86>\M-B\M-2kzla\\e^lv\M-C\M^_\M-B\M^Snv\M-C\M^_\M-B\M-.a|3}\M-C\M^X\M-B\M^H\M-C\M^V\M-B\M^D`, 0}, + {"@?e1xs+.R_Kjo]7s8pgRP:*nXCE4{!c", "@?e1xs+.R_Kjo]7s8pgRP:*nXCE4{!c", 0}, + {"62_\u00c6\u00c62\u00ae\u00b7m\u00db\u00c3r^\u00bfp\u00c6u'q\u00fbc2\u00f0u\u00b8\u00dd\u00e8v\u00ff\u00b0\u00dc\u00c2\u00f53\u00db-k\u00f2sd4\\p\u00da\u00a6\u00d3\u00eea<\u00e6s{\u00a0p\u00f0\u00ffj\u00e0\u00e8\u00b8\u00b8\u00bc\u00fcb", `62_\M-C\M^F\M-C\M^F2\M-B\M-.\M-B\M-7m\M-C\M^[\M-C\M^Cr^\M-B\M-?p\M-C\M^Fu'q\M-C\M-;c2\M-C\M-0u\M-B\M-8\M-C\M^]\M-C\M-(v\M-C\M-?\M-B\M-0\M-C\M^\\M-C\M^B\M-C\M-53\M-C\M^[-k\M-C\M-2sd4\\p\M-C\M^Z\M-B\M-&\M-C\M^S\M-C\M-.a<\M-C\M-&s{\M-B\240p\M-C\M-0\M-C\M-?j\M-C\240\M-C\M-(\M-B\M-8\M-B\M-8\M-B\M-<\M-C\M-\u00b2kzla\\e^lv\u00df\u0093nv\u00df\u00aea|3}\u00d8\u0088\u00d6\u0084", `z^i3i$\303\223\302\212nqgh5/t\303\245<86>\302\262kzla\\e^lv\303\237\302\223nv\303\237\302\256a|3}\303\230\302\210\303\226\302\204`, VisOctal}, + {"62_\u00c6\u00c62\u00ae\u00b7m\u00db\u00c3r^\u00bfp\u00c6u'q\u00fbc2\u00f0u\u00b8\u00dd\u00e8v\u00ff\u00b0\u00dc\u00c2\u00f53\u00db-k\u00f2sd4\\p\u00da\u00a6\u00d3\u00eea<\u00e6s{\u00a0p\u00f0\u00ffj\u00e0\u00e8\u00b8\u00b8\u00bc\u00fcb", `62_\303\206\303\2062\302\256\302\267m\303\233\303\203r^\302\277p\303\206u'q\303\273c2\303\260u\302\270\303\235\303\250v\303\277\302\260\303\234\303\202\303\2653\303\233-k\303\262sd4\\p\303\232\302\246\303\223\303\256a<\303\246s{\302\240p\303\260\303\277j\303\240\303\250\302\270\302\270\302\274\303\274b`, VisOctal}, + {"\u9003\"9v1)T798|o;fly jnKX\u0489Be=", `\351\200\203"9v1)T798|o;fly jnKX\322\211Be=`, VisOctal}, + // VisCStyle + {"\x00 \f \a \n\v\b \r \t\r", "\\000 \\f \\a \n\\v\\b \\r \t\\r", VisCStyle}, + {"\t \n\v\b", "\\t \n\\v\\b", VisTab | VisCStyle}, + {"\n\v\t ", "\n\\v\t\\s\\s\\s", VisSpace | VisCStyle}, + {"\n \n ", "\\n \\n ", VisNewline | VisCStyle}, + {"z^i3i$\u00d3\u008anqgh5/t\u00e5<86>\u00b2kzla\\e^lv\u00df\u0093nv\u00df\u00aea|3}\u00d8\u0088\u00d6\u0084", `z^i3i$\M-C\M^S\M-B\M^Jnqgh5/t\M-C\M-%<86>\M-B\M-2kzla\\e^lv\M-C\M^_\M-B\M^Snv\M-C\M^_\M-B\M-.a|3}\M-C\M^X\M-B\M^H\M-C\M^V\M-B\M^D`, VisCStyle}, + {"62_\u00c6\u00c62\u00ae\u00b7m\u00db\u00c3r^\u00bfp\u00c6u'q\u00fbc2\u00f0u\u00b8\u00dd\u00e8v\u00ff\u00b0\u00dc\u00c2\u00f53\u00db-k\u00f2sd4\\p\u00da\u00a6\u00d3\u00eea<\u00e6s{\u00a0p\u00f0\u00ffj\u00e0\u00e8\u00b8\u00b8\u00bc\u00fcb", `62_\M-C\M^F\M-C\M^F2\M-B\M-.\M-B\M-7m\M-C\M^[\M-C\M^Cr^\M-B\M-?p\M-C\M^Fu'q\M-C\M-;c2\M-C\M-0u\M-B\M-8\M-C\M^]\M-C\M-(v\M-C\M-?\M-B\M-0\M-C\M^\\M-C\M^B\M-C\M-53\M-C\M^[-k\M-C\M-2sd4\\p\M-C\M^Z\M-B\M-&\M-C\M^S\M-C\M-.a<\M-C\M-&s{\M-B\240p\M-C\M-0\M-C\M-?j\M-C\240\M-C\M-(\M-B\M-8\M-B\M-8\M-B\M-<\M-C\M- Date: Thu, 16 Feb 2017 01:03:29 +1100 Subject: [PATCH 12/15] govis: extend roundtrip testing This is necessary to ensure that strings I don't have examples of work properly. The same thing applies for double-encode and double-decode usecases. Signed-off-by: Aleksa Sarai --- govis_test.go | 147 ++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 125 insertions(+), 22 deletions(-) diff --git a/govis_test.go b/govis_test.go index 244ae23..312cec3 100644 --- a/govis_test.go +++ b/govis_test.go @@ -19,36 +19,139 @@ package govis import ( "bytes" + "crypto/rand" "testing" ) const DefaultVisFlags = VisWhite | VisOctal | VisGlob +func TestRandomVisUnvis(t *testing.T) { + // Randomly generate N strings. + const N = 100 + + for i := 0; i < N; i++ { + testBytes := make([]byte, 256) + if n, err := rand.Read(testBytes); n != cap(testBytes) || err != nil { + t.Fatalf("could not read enough bytes: err=%v n=%d", err, n) + } + test := string(testBytes) + + for flag := VisFlag(0); flag <= visMask; flag++ { + // VisNoSlash is frankly just a dumb flag, and it is impossible for us + // to actually preserve things in a round-trip. + if flag&VisNoSlash == VisNoSlash { + continue + } + + enc, err := Vis(test, flag) + if err != nil { + t.Errorf("unexpected error doing vis(%q, %b): %s", test, flag, err) + continue + } + dec, err := Unvis(enc, flag) + if err != nil { + t.Errorf("unexpected error doing unvis(%q, %b): %s", enc, flag, err) + continue + } + if dec != test { + t.Errorf("roundtrip failed: unvis(vis(%q, %b) = %q, %b) = %q", test, flag, enc, flag, dec) + } + } + } +} + +func TestRandomVisVisUnvisUnvis(t *testing.T) { + // Randomly generate N strings. + const N = 100 + + for i := 0; i < N; i++ { + testBytes := make([]byte, 256) + if n, err := rand.Read(testBytes); n != cap(testBytes) || err != nil { + t.Fatalf("could not read enough bytes: err=%v n=%d", err, n) + } + test := string(testBytes) + + for flag := VisFlag(0); flag <= visMask; flag++ { + // VisNoSlash is frankly just a dumb flag, and it is impossible for us + // to actually preserve things in a round-trip. + if flag&VisNoSlash == VisNoSlash { + continue + } + + enc, err := Vis(test, flag) + if err != nil { + t.Errorf("unexpected error doing vis(%q, %b): %s", test, flag, err) + continue + } + enc2, err := Vis(enc, flag) + if err != nil { + t.Errorf("unexpected error doing vis(%q, %b): %s", enc, flag, err) + continue + } + dec, err := Unvis(enc2, flag) + if err != nil { + t.Errorf("unexpected error doing unvis(%q, %b): %s", enc2, flag, err) + continue + } + dec2, err := Unvis(dec, flag) + if err != nil { + t.Errorf("unexpected error doing unvis(%q, %b): %s", dec, flag, err) + continue + } + if dec2 != test { + t.Errorf("roundtrip failed: unvis(unvis(vis(vis(%q) = %q) = %q) = %q, %b) = %q", test, enc, enc2, dec, flag, dec2) + } + } + } +} + func TestVisUnvis(t *testing.T) { - // Round-trip testing. - for _, test := range []string{ - "", - "this.is.a.normal_string", - "AC_Raíz_Certicámara_S.A..pem", - "NetLock_Arany_=Class_Gold=_Főtanúsítvány.pem", - "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem", - "hello world [ this string needs=enco ding! ]", - "even \n more encoding necessary\a\a ", - "\024 <-- some more weird characters --> 你好，世界", - "\\xff\\n double encoding is also great fun \\x", - } { - enc, err := Vis(test, DefaultVisFlags) - if err != nil { - t.Errorf("unexpected error doing vis(%q): %s", test, err) + for flag := VisFlag(0); flag <= visMask; flag++ { + // VisNoSlash is frankly just a dumb flag, and it is impossible for us + // to actually preserve things in a round-trip. + if flag&VisNoSlash == VisNoSlash { continue } - dec, err := Unvis(enc, DefaultVisFlags) - if err != nil { - t.Errorf("unexpected error doing unvis(%q): %s", enc, err) - continue - } - if dec != test { - t.Errorf("roundtrip failed: unvis(vis(%q) = %q) = %q", test, enc, dec) + + // Round-trip testing. + for _, test := range []string{ + "", + "hello world", + "THIS\\IS_A_TEST1234", + "this.is.a.normal_string", + "AC_Ra\u00edz_Certic\u00e1mara_S.A..pem", + "NetLock_Arany_=Class_Gold=_F\u0151tan\u00fas\u00edtv\u00e1ny.pem", + "T\u00dcB\u0130TAK_UEKAE_K\u00f6k_Sertifika_Hizmet_Sa\u011flay\u0131c\u0131s\u0131_-_S\u00fcr\u00fcm_3.pem", + "hello world [ this string needs=enco ding! ]", + "even \n more encoding necessary\a\a ", + "\024 <-- some more weird characters --> \u4f60\u597d\uff0c\u4e16\u754c", + "\\xff\\n double encoding is also great fun \\x", + "AC_Ra\\M-C\\M--z_Certic\\M-C\\M-!mara_S.A..pem", + "z^i3i$\u00d3\u008anqgh5/t\u00e5<86>\u00b2kzla\\e^lv\u00df\u0093nv\u00df\u00aea|3}\u00d8\u0088\u00d6\u0084", + `z^i3i$\M-C\M^S\M-B\M^Jnqgh5/t\M-C\M-%<86>\M-B\M-2kzla\\e^lv\M-C\M^_\M-B\M^Snv\M-C\M^_\M-B\M-.a|3}\M-C\M^X\M-B\M^H\M-C\M^V\M-B\M^D`, + "@?e1xs+.R_Kjo]7s8pgRP:*nXCE4{!c", + "62_\u00c6\u00c62\u00ae\u00b7m\u00db\u00c3r^\u00bfp\u00c6u'q\u00fbc2\u00f0u\u00b8\u00dd\u00e8v\u00ff\u00b0\u00dc\u00c2\u00f53\u00db-k\u00f2sd4\\p\u00da\u00a6\u00d3\u00eea<\u00e6s{\u00a0p\u00f0\u00ffj\u00e0\u00e8\u00b8\u00b8\u00bc\u00fcb", + `62_\M-C\M^F\M-C\M^F2\M-B\M-.\M-B\M-7m\M-C\M^[\M-C\M^Cr^\M-B\M-?p\M-C\M^Fu'q\M-C\M-;c2\M-C\M-0u\M-B\M-8\M-C\M^]\M-C\M-(v\M-C\M-?\M-B\M-0\M-C\M^\\M-C\M^B\M-C\M-53\M-C\M^[-k\M-C\M-2sd4\\p\M-C\M^Z\M-B\M-&\M-C\M^S\M-C\M-.a<\M-C\M-&s{\M-B\240p\M-C\M-0\M-C\M-?j\M-C\240\M-C\M-(\M-B\M-8\M-B\M-8\M-B\M-<\M-C\M- Date: Thu, 16 Feb 2017 01:11:32 +1100 Subject: [PATCH 13/15] README: add travis-ci build Signed-off-by: Aleksa Sarai --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index d56e8da..dfdbda1 100644 --- a/README.md +++ b/README.md @@ -1,4 +1,5 @@ ## `govis` ## +[![Travis CI](https://travis-ci.org/cyphar/govis.svg?branch=master)](https://travis-ci.org/cyphar/govis) `govis` is a BSD-compatible `vis(3)` and `unvis(3)` encoding implementation that is unicode aware and written in Go. None of this code comes from the From c889416068e8e3dcc54910382c5be40e56640b08 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Thu, 16 Feb 2017 03:08:07 +1100 Subject: [PATCH 14/15] *: fix go-vet errors Signed-off-by: Aleksa Sarai --- unvis.go | 4 +--- vis_test.go | 2 +- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/unvis.go b/unvis.go index f8dd6b1..8a26218 100644 --- a/unvis.go +++ b/unvis.go @@ -74,7 +74,7 @@ func newParser(input string, flag VisFlag) *unvisParser { func unvisPlainRune(p *unvisParser) ([]byte, error) { ch, err := p.Peek() if err != nil { - return nil, fmt.Errorf("plain rune: %s", ch) + return nil, fmt.Errorf("plain rune: %c", ch) } p.Next() @@ -238,8 +238,6 @@ func unvisEscapeSequence(p *unvisParser) ([]byte, error) { default: return unvisEscapeCStyle(p) } - - return nil, fmt.Errorf("escape sequence: unsupported sequence: %q", ch) } func unvisRune(p *unvisParser) ([]byte, error) { diff --git a/vis_test.go b/vis_test.go index e4ef083..5177a58 100644 --- a/vis_test.go +++ b/vis_test.go @@ -121,7 +121,7 @@ func TestVisChanged(t *testing.T) { t.Errorf("unexpected error with %q: %s", test, err) } if enc == test { - t.Errorf("expected encoding of %q to be changed") + t.Errorf("expected encoding of %q to be changed", test) } } } From c5ec1c9f3a767fb6c69d48cd2924885179e30314 Mon Sep 17 00:00:00 2001 From: Aleksa Sarai Date: Thu, 16 Feb 2017 03:10:30 +1100 Subject: [PATCH 15/15] *: switch everything to govis Now that we have govis, move everything to using govis.{Vis,Unvis} and then remove the cvis build tags (because that code no longer exists). Signed-off-by: Aleksa Sarai --- Makefile | 4 +- cvis/cvis_test.go | 15 --- cvis/unvis.c | 293 ---------------------------------------------- cvis/unvis.go | 22 ---- cvis/vis.c | 202 -------------------------------- cvis/vis.go | 28 ----- cvis/vis.h | 90 -------------- entry.go | 6 +- keywordfunc.go | 5 +- keywords.go | 6 + keywords_linux.go | 5 +- tar.go | 8 +- unvis.go | 7 -- unvis_c.go | 11 -- unvis_go.go | 233 ------------------------------------ unvis_go_test.go | 93 --------------- vis.go | 89 -------------- vis_c.go | 11 -- vis_go.go | 107 ----------------- vis_test.go | 39 ------ walk.go | 4 +- 21 files changed, 26 insertions(+), 1252 deletions(-) delete mode 100644 cvis/cvis_test.go delete mode 100644 cvis/unvis.c delete mode 100644 cvis/unvis.go delete mode 100644 cvis/vis.c delete mode 100644 cvis/vis.go delete mode 100644 cvis/vis.h delete mode 100644 unvis.go delete mode 100644 unvis_c.go delete mode 100644 unvis_go.go delete mode 100644 unvis_go_test.go delete mode 100644 vis.go delete mode 100644 vis_c.go delete mode 100644 vis_go.go delete mode 100644 vis_test.go diff --git a/Makefile b/Makefile index 72c21f6..8967673 100644 --- a/Makefile +++ b/Makefile @@ -4,10 +4,10 @@ BUILDPATH := github.com/vbatts/go-mtree/cmd/gomtree CWD := $(shell pwd) SOURCE_FILES := $(shell find . -type f -name "*.go") CLEAN_FILES := *~ -TAGS := cvis +TAGS := ARCHES := linux,386 linux,amd64 linux,arm linux,arm64 openbsd,amd64 windows,amd64 darwin,amd64 -default: build validation +default: build validation .PHONY: validation validation: .test .lint .vet .cli.test diff --git a/cvis/cvis_test.go b/cvis/cvis_test.go deleted file mode 100644 index 75e3884..0000000 --- a/cvis/cvis_test.go +++ /dev/null @@ -1,15 +0,0 @@ -// +build cgo,!govis - -package cvis - -import "testing" - -// 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, DefaultVisFlags) - testString = testString + testString - } -} diff --git a/cvis/unvis.c b/cvis/unvis.c deleted file mode 100644 index ab16298..0000000 --- a/cvis/unvis.c +++ /dev/null @@ -1,293 +0,0 @@ -/*- - * 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 - -#include -#include -#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); -} diff --git a/cvis/unvis.go b/cvis/unvis.go deleted file mode 100644 index 398a53b..0000000 --- a/cvis/unvis.go +++ /dev/null @@ -1,22 +0,0 @@ -package cvis - -// #include "vis.h" -// #include -import "C" -import ( - "fmt" - "unsafe" -) - -// Unvis decodes the Vis() string encoding -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) - // TODO(vbatts) this needs to be confirmed against UnvisError - if ret == -1 { - return "", fmt.Errorf("failed to decode: %q", src) - } - return C.GoString(cDst), nil -} diff --git a/cvis/vis.c b/cvis/vis.c deleted file mode 100644 index c3390b4..0000000 --- a/cvis/vis.c +++ /dev/null @@ -1,202 +0,0 @@ -/*- - * 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 - -#include -#include -#include -#include -#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); -} diff --git a/cvis/vis.go b/cvis/vis.go deleted file mode 100644 index c10073d..0000000 --- a/cvis/vis.go +++ /dev/null @@ -1,28 +0,0 @@ -package cvis - -// #include "vis.h" -// #include -import "C" -import ( - "fmt" - "math" - "unsafe" -) - -// Vis is a wrapper around the C implementation -func Vis(src string, flags int) (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.int(flags)) - - return C.GoString(cDst), nil -} - -// DefaultVisFlags are the common flags used in mtree string encoding -var DefaultVisFlags = C.VIS_WHITE | C.VIS_OCTAL | C.VIS_GLOB diff --git a/cvis/vis.h b/cvis/vis.h deleted file mode 100644 index 0798008..0000000 --- a/cvis/vis.h +++ /dev/null @@ -1,90 +0,0 @@ -/*- - * 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 - -/* - * 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 - -__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_ */ diff --git a/entry.go b/entry.go index 558d1cd..adf2592 100644 --- a/entry.go +++ b/entry.go @@ -4,6 +4,8 @@ import ( "fmt" "path/filepath" "strings" + + "github.com/vbatts/go-mtree/pkg/govis" ) type byPos []Entry @@ -47,7 +49,7 @@ func (e Entry) Descend(filename string) *Entry { func (e Entry) Find(filepath string) *Entry { resultnode := &e for _, path := range strings.Split(filepath, "/") { - encoded, err := Vis(path, DefaultVisFlags) + encoded, err := govis.Vis(path, DefaultVisFlags) if err != nil { return nil } @@ -68,7 +70,7 @@ func (e Entry) Ascend() *Entry { // Path provides the full path of the file, despite RelativeType or FullType. It // will be in Unvis'd form. func (e Entry) Path() (string, error) { - decodedName, err := Unvis(e.Name) + decodedName, err := govis.Unvis(e.Name, DefaultVisFlags) if err != nil { return "", err } diff --git a/keywordfunc.go b/keywordfunc.go index c9a0f77..e88fa2e 100644 --- a/keywordfunc.go +++ b/keywordfunc.go @@ -11,6 +11,7 @@ import ( "io" "os" + "github.com/vbatts/go-mtree/pkg/govis" "golang.org/x/crypto/ripemd160" ) @@ -119,7 +120,7 @@ var ( linkKeywordFunc = func(path string, info os.FileInfo, r io.Reader) (KeyVal, error) { if sys, ok := info.Sys().(*tar.Header); ok { if sys.Linkname != "" { - linkname, err := Vis(sys.Linkname, DefaultVisFlags) + linkname, err := govis.Vis(sys.Linkname, DefaultVisFlags) if err != nil { return emptyKV, err } @@ -133,7 +134,7 @@ var ( if err != nil { return emptyKV, err } - linkname, err := Vis(str, DefaultVisFlags) + linkname, err := govis.Vis(str, DefaultVisFlags) if err != nil { return emptyKV, err } diff --git a/keywords.go b/keywords.go index b139d6d..ed121b5 100644 --- a/keywords.go +++ b/keywords.go @@ -3,8 +3,14 @@ package mtree import ( "fmt" "strings" + + "github.com/vbatts/go-mtree/pkg/govis" ) +// DefaultVisFlags is the set of Vis flags used when encoding filenames and +// other similar entries. +const DefaultVisFlags govis.VisFlag = govis.VisWhite | govis.VisOctal | govis.VisGlob + // Keyword is the string name of a keyword, with some convenience functions for // determining whether it is a default or bsd standard keyword. type Keyword string diff --git a/keywords_linux.go b/keywords_linux.go index a14108e..bab7a8d 100644 --- a/keywords_linux.go +++ b/keywords_linux.go @@ -12,6 +12,7 @@ import ( "strings" "syscall" + "github.com/vbatts/go-mtree/pkg/govis" "github.com/vbatts/go-mtree/xattr" ) @@ -62,7 +63,7 @@ var ( } klist := []KeyVal{} for k, v := range hdr.Xattrs { - encKey, err := Vis(k, DefaultVisFlags) + encKey, err := govis.Vis(k, DefaultVisFlags) if err != nil { return emptyKV, err } @@ -84,7 +85,7 @@ var ( if err != nil { return emptyKV, err } - encKey, err := Vis(xlist[i], DefaultVisFlags) + encKey, err := govis.Vis(xlist[i], DefaultVisFlags) if err != nil { return emptyKV, err } diff --git a/tar.go b/tar.go index 800a484..7f7318e 100644 --- a/tar.go +++ b/tar.go @@ -9,6 +9,8 @@ import ( "os" "path/filepath" "strings" + + "github.com/vbatts/go-mtree/pkg/govis" ) // Streamer creates a file hierarchy out of a tar stream @@ -128,7 +130,7 @@ hdrloop: return } // Alright, it's either file or directory - encodedName, err := Vis(filepath.Base(hdr.Name), DefaultVisFlags) + encodedName, err := govis.Vis(filepath.Base(hdr.Name), DefaultVisFlags) if err != nil { tmpFile.Close() os.Remove(tmpFile.Name()) @@ -148,7 +150,7 @@ hdrloop: log.Println(err) break } - linkname, err := Unvis(KeyVal(kv).Value()) + linkname, err := govis.Unvis(KeyVal(kv).Value(), DefaultVisFlags) if err != nil { log.Println(err) break @@ -248,7 +250,7 @@ func populateTree(root, e *Entry, hdr *tar.Header) error { dirNames := strings.Split(wd, "/") parent := root for _, name := range dirNames[:] { - encoded, err := Vis(name, DefaultVisFlags) + encoded, err := govis.Vis(name, DefaultVisFlags) if err != nil { return err } diff --git a/unvis.go b/unvis.go deleted file mode 100644 index 7c63666..0000000 --- a/unvis.go +++ /dev/null @@ -1,7 +0,0 @@ -package mtree - -// 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) { - return unvis(src) -} diff --git a/unvis_c.go b/unvis_c.go deleted file mode 100644 index ac421f1..0000000 --- a/unvis_c.go +++ /dev/null @@ -1,11 +0,0 @@ -// +build cgo,cvis - -package mtree - -import ( - "github.com/vbatts/go-mtree/cvis" -) - -func unvis(src string) (string, error) { - return cvis.Unvis(src) -} diff --git a/unvis_go.go b/unvis_go.go deleted file mode 100644 index 0f2a848..0000000 --- a/unvis_go.go +++ /dev/null @@ -1,233 +0,0 @@ -// +build !cvis - -package mtree - -import "unicode" - -func unvis(src string) (string, error) { - dst := []rune{} - var s state - for i, r := range src { - again: - err := unvisRune(&dst, r, &s, 0) - switch err { - case unvisValid: - break - case unvisValidPush: - goto again - case unvisNone: - fallthrough - case unvisNochar: - break - default: - return "", err - } - if i == len(src)-1 { - unvisRune(&dst, r, &s, unvisEnd) - } - } - - str := "" - for _, ch := range dst { - str += string(ch) - } - return str, nil -} - -func unvisRune(dst *[]rune, r rune, s *state, flags VisFlag) error { - if (flags & unvisEnd) != 0 { - if *s == stateOctal2 || *s == stateOctal3 { - *s = stateGround - return unvisValid - } - if *s == stateGround { - return unvisNochar - } - return unvisErrSynbad - } - - switch *s & ^stateHTTP { - case stateGround: - if r == '\\' { - *s = stateStart - return unvisNone - } - if flags&VisHttpstyle != 0 && r == '%' { - *s = stateStart | stateHTTP - return unvisNone - } - *dst = append(*dst, r) - return unvisValid - case stateStart: - if *s&stateHTTP != 0 && ishex(unicode.ToLower(r)) { - if unicode.IsNumber(r) { - *dst = append(*dst, r-'0') - } else { - *dst = append(*dst, unicode.ToLower(r)-'a') - } - *s = stateHex2 - return unvisNone - } - switch r { - case '\\': - *s = stateGround - *dst = append(*dst, r) - return unvisValid - case '0': - fallthrough - case '1': - fallthrough - case '2': - fallthrough - case '3': - fallthrough - case '4': - fallthrough - case '5': - fallthrough - case '6': - fallthrough - case '7': - *s = stateOctal2 - *dst = append(*dst, r-'0') - return unvisNone - case 'M': - *s = stateMeta - *dst = append(*dst, rune(0200)) - return unvisNone - case '^': - *s = stateCtrl - return unvisNone - case 'n': - *s = stateGround - *dst = append(*dst, '\n') - return unvisValid - case 'r': - *s = stateGround - *dst = append(*dst, '\r') - return unvisValid - case 'b': - *s = stateGround - *dst = append(*dst, '\b') - return unvisValid - case 'a': - *s = stateGround - *dst = append(*dst, '\007') - return unvisValid - case 'v': - *s = stateGround - *dst = append(*dst, '\v') - return unvisValid - case 't': - *s = stateGround - *dst = append(*dst, '\t') - return unvisValid - case 'f': - *s = stateGround - *dst = append(*dst, '\f') - return unvisValid - case 's': - *s = stateGround - *dst = append(*dst, ' ') - return unvisValid - case 'E': - *s = stateGround - *dst = append(*dst, '\033') - return unvisValid - case '\n': - // hidden newline - *s = stateGround - return unvisNochar - case '$': - // hidden marker - *s = stateGround - return unvisNochar - } - *s = stateGround - return unvisErrSynbad - case stateMeta: - if r == '-' { - *s = stateMeta1 - } else if r == '^' { - *s = stateCtrl - } else { - *s = stateGround - return unvisErrSynbad - } - return unvisNone - case stateMeta1: - *s = stateGround - dp := *dst - dp[len(dp)-1] |= r - return unvisValid - case stateCtrl: - dp := *dst - if r == '?' { - dp[len(dp)-1] |= rune(0177) - } else { - dp[len(dp)-1] |= r & 037 - } - *s = stateGround - return unvisValid - case stateOctal2: - if isoctal(r) { - dp := *dst - if len(dp) > 0 { - last := dp[len(dp)-1] - dp[len(dp)-1] = (last << 3) + (r - '0') - } else { - dp = append(dp, (0<<3)+(r-'0')) - } - *s = stateOctal3 - return unvisNone - } - *s = stateGround - return unvisValidPush - case stateOctal3: - *s = stateGround - if isoctal(r) { - dp := *dst - if len(dp) > 0 { - last := dp[len(dp)-1] - dp[len(dp)-1] = (last << 3) + (r - '0') - } else { - dp = append(dp, (0<<3)+(r-'0')) - } - return unvisValid - } - return unvisValidPush - case stateHex2: - if ishex(unicode.ToLower(r)) { - last := rune(0) - dp := *dst - if len(dp) > 0 { - last = dp[len(dp)-1] - } - if unicode.IsNumber(r) { - dp = append(dp, (last<<4)+(r-'0')) - } else { - dp = append(dp, (last<<4)+(unicode.ToLower(r)-'a'+10)) - } - } - *s = stateGround - return unvisValid - } - - *s = stateGround - return unvisErrSynbad -} - -type state int - -const ( - stateGround state = iota /* haven't seen escape char */ - stateStart /* start decoding special sequence */ - stateMeta /* metachar started (M) */ - stateMeta1 /* metachar more, regular char (-) */ - stateCtrl /* control char started (^) */ - stateOctal2 /* octal digit 2 */ - stateOctal3 /* octal digit 3 */ - stateHex2 /* hex digit 2 */ - - stateHTTP state = 0x080 /* %HEXHEX escape */ -) diff --git a/unvis_go_test.go b/unvis_go_test.go deleted file mode 100644 index 0289330..0000000 --- a/unvis_go_test.go +++ /dev/null @@ -1,93 +0,0 @@ -package mtree - -import "testing" - -type runeCheck func(rune) bool - -func TestUnvisHelpers(t *testing.T) { - testset := []struct { - R rune - Check runeCheck - Expect bool - }{ - {'a', ishex, true}, - {'A', ishex, true}, - {'z', ishex, false}, - {'Z', ishex, false}, - {'G', ishex, false}, - {'1', ishex, true}, - {'0', ishex, true}, - {'9', ishex, true}, - {'0', isoctal, true}, - {'3', isoctal, true}, - {'7', isoctal, true}, - {'9', isoctal, false}, - {'a', isoctal, false}, - {'z', isoctal, false}, - {'3', isalnum, true}, - {'a', isalnum, true}, - {';', isalnum, false}, - {'!', isalnum, false}, - {' ', isalnum, false}, - {'3', isgraph, true}, - {'a', isgraph, true}, - {';', isgraph, true}, - {'!', isgraph, true}, - {' ', isgraph, false}, - } - - for i, ts := range testset { - got := ts.Check(ts.R) - if got != ts.Expect { - t.Errorf("%d: %q expected: %t; got %t", i, string(ts.R), ts.Expect, got) - } - } -} - -func TestUnvisUnicode(t *testing.T) { - // Ensure that unicode strings are not messed up by Unvis. - for _, test := range []string{ - "", - "this.is.a.normal_string", - "AC_Raíz_Certicámara_S.A..pem", - "NetLock_Arany_=Class_Gold=_Főtanúsítvány.pem", - "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem", - } { - got, err := Unvis(test) - if err != nil { - t.Errorf("unexpected error doing unvis(%q): %s", test, err) - continue - } - if got != test { - t.Errorf("expected %q to be unchanged, got %q", test, got) - } - } -} - -func TestVisUnvis(t *testing.T) { - // Round-trip testing. - for _, test := range []string{ - "", - "this.is.a.normal_string", - "AC_Raíz_Certicámara_S.A..pem", - "NetLock_Arany_=Class_Gold=_Főtanúsítvány.pem", - "TÜBİTAK_UEKAE_Kök_Sertifika_Hizmet_Sağlayıcısı_-_Sürüm_3.pem", - "hello world [ this string needs=enco ding! ]", - "even \n more encoding necessary\a\a ", - "\024 <-- some more weird characters --> 你好，世界", - } { - enc, err := Vis(test, DefaultVisFlags) - if err != nil { - t.Errorf("unexpected error doing vis(%q): %s", test, err) - continue - } - dec, err := Unvis(enc) - if err != nil { - t.Errorf("unexpected error doing unvis(%q): %s", enc, err) - continue - } - if dec != test { - t.Errorf("roundtrip failed: unvis(vis(%q) = %q) = %q", test, enc, dec) - } - } -} diff --git a/vis.go b/vis.go deleted file mode 100644 index 440283c..0000000 --- a/vis.go +++ /dev/null @@ -1,89 +0,0 @@ -package mtree - -import "unicode" - -// Vis is a wrapper of the C implementation of the function vis, which encodes -// a character with a particular format/style. -// For most use-cases use DefaultVisFlags. -func Vis(src string, flags VisFlag) (string, error) { - return vis(src, flags) -} - -// DefaultVisFlags are the typical flags used for encoding strings in mtree -// manifests. -var DefaultVisFlags = VisWhite | VisOctal | VisGlob - -// VisFlag sets the extent of charactures to be encoded -type VisFlag int - -// flags for encoding -const ( - // to select alternate encoding format - VisOctal VisFlag = 0x01 // use octal \ddd format - VisCstyle VisFlag = 0x02 // use \[nrft0..] where appropriate - - // to alter set of characters encoded (default is to encode all non-graphic - // except space, tab, and newline). - VisSp VisFlag = 0x04 // also encode space - VisTab VisFlag = 0x08 // also encode tab - VisNl VisFlag = 0x10 // also encode newline - VisWhite VisFlag = (VisSp | VisTab | VisNl) - VisSafe VisFlag = 0x20 // only encode "unsafe" characters - - // other - VisNoSlash VisFlag = 0x40 // inhibit printing '\' - VisHttpstyle VisFlag = 0x80 // http-style escape % HEX HEX - VisGlob VisFlag = 0x100 // encode glob(3) magics - -) - -// errors used in the tokenized decoding strings -const ( - // unvis return codes - unvisValid unvisErr = 1 // character valid - unvisValidPush unvisErr = 2 // character valid, push back passed char - unvisNochar unvisErr = 3 // valid sequence, no character produced - unvisErrSynbad unvisErr = -1 // unrecognized escape sequence - unvisErrUnrecoverable unvisErr = -2 // decoder in unknown state (unrecoverable) - unvisNone unvisErr = 0 - - // unvisEnd means there are no more characters - unvisEnd VisFlag = 1 // no more characters -) - -// unvisErr are the return conditions for Unvis -type unvisErr int - -func (ue unvisErr) Error() string { - switch ue { - case unvisValid: - return "character valid" - case unvisValidPush: - return "character valid, push back passed char" - case unvisNochar: - return "valid sequence, no character produced" - case unvisErrSynbad: - return "unrecognized escape sequence" - case unvisErrUnrecoverable: - return "decoder in unknown state (unrecoverable)" - } - return "Unknown Error" -} - -func ishex(r rune) bool { - lr := unicode.ToLower(r) - return (lr >= '0' && lr <= '9') || (lr >= 'a' && lr <= 'f') -} - -func isoctal(r rune) bool { - return r <= '7' && r >= '0' -} - -// the ctype isgraph is "any printable character except space" -func isgraph(r rune) bool { - return unicode.IsPrint(r) && !unicode.IsSpace(r) -} - -func isalnum(r rune) bool { - return unicode.IsNumber(r) || unicode.IsLetter(r) -} diff --git a/vis_c.go b/vis_c.go deleted file mode 100644 index 5024356..0000000 --- a/vis_c.go +++ /dev/null @@ -1,11 +0,0 @@ -// +build cgo,cvis - -package mtree - -import ( - "github.com/vbatts/go-mtree/cvis" -) - -func vis(src string, flags VisFlag) (string, error) { - return cvis.Vis(src, int(flags)) -} diff --git a/vis_go.go b/vis_go.go deleted file mode 100644 index a914693..0000000 --- a/vis_go.go +++ /dev/null @@ -1,107 +0,0 @@ -// +build !cvis - -package mtree - -import ( - "fmt" - "unicode" -) - -func vis(src string, flags VisFlag) (string, error) { - var ret string - for _, r := range src { - vStr, err := visRune(r, flags) - if err != nil { - return "", err - } - ret = ret + vStr - } - return ret, nil -} - -func visRune(r rune, flags VisFlag) (string, error) { - if flags&VisHttpstyle != 0 { - // Described in RFC 1808 - if !isalnum(r) || - /* safe */ - r == '$' || r == '-' || r == '_' || r == '.' || r == '+' || - /* extra */ - r == '!' || r == '*' || r == '\'' || r == '(' || - r == ')' || r == ',' { - if r < 16 { - return fmt.Sprintf("%%0%X", r), nil - } - return fmt.Sprintf("%%%X", r), nil - } - } - - if (flags&VisGlob) != 0 && (r == '*' || r == '?' || r == '[' || r == '#') { - // ... ? - } else if isgraph(r) || - ((flags&VisSp) == 0 && r == ' ') || - ((flags&VisTab) == 0 && r == '\t') || - ((flags&VisNl) == 0 && r == '\n') || - ((flags&VisSafe) != 0 && (r == '\b' || r == '\007' || r == '\r')) { - if r == '\\' && (flags&VisNoSlash) == 0 { - return fmt.Sprintf("%s\\", string(r)), nil - } - return string(r), nil - } - - if (flags & VisCstyle) != 0 { - switch r { - case '\n': - return "\\n", nil - case '\r': - return "\\r", nil - case '\b': - return "\\b", nil - case '\a': - return "\\a", nil - case '\v': - return "\\v", nil - case '\t': - return "\\t", nil - case '\f': - return "\\f", nil - case ' ': - return "\\s", nil - case rune(0x0): - return "\\0", nil - /* - if isoctal(nextr) { - dst = append(dst, '0') - dst = append(dst, '0') - } - */ - } - } - if ((r & 0177) == ' ') || isgraph(r) || (flags&VisOctal) != 0 { - dst := make([]rune, 4) - dst[0] = '\\' - dst[1] = (r >> 6 & 07) + '0' - dst[2] = (r >> 3 & 07) + '0' - dst[3] = (r & 07) + '0' - return string(dst), nil - } - var dst []rune - if (flags & VisNoSlash) == 0 { - dst = append(dst, '\\') - } - if (r & 0200) != 0 { - r &= 0177 - dst = append(dst, 'M') - } - if unicode.IsControl(r) { - dst = append(dst, '^') - if r == 0177 { - dst = append(dst, '?') - } else { - dst = append(dst, r+'@') - } - } else { - dst = append(dst, '-') - dst = append(dst, r) - } - return string(dst), nil -} diff --git a/vis_test.go b/vis_test.go deleted file mode 100644 index 03aba51..0000000 --- a/vis_test.go +++ /dev/null @@ -1,39 +0,0 @@ -package mtree - -import "testing" - -func TestVisBasic(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, DefaultVisFlags) - if err != nil { - t.Errorf("working with %q: %s", testset[i].Src, err) - } - if got != testset[i].Dest { - t.Errorf("%q: expected %#v; got %#v", testset[i].Src, testset[i].Dest, got) - continue - } - - got, err = Unvis(got) - if err != nil { - t.Errorf("working with %q: %s: %q", testset[i].Src, err, got) - continue - } - if got != testset[i].Src { - t.Errorf("%q: expected %#v; got %#v", testset[i].Dest, testset[i].Src, got) - continue - } - } -} diff --git a/walk.go b/walk.go index b1dfe87..3af4110 100644 --- a/walk.go +++ b/walk.go @@ -9,6 +9,8 @@ import ( "sort" "strings" "time" + + "github.com/vbatts/go-mtree/pkg/govis" ) // ExcludeFunc is the type of function called on each path walked to determine @@ -165,7 +167,7 @@ func Walk(root string, excludes []ExcludeFunc, keywords []Keyword, fsEval FsEval } } } - encodedEntryName, err := Vis(entryPathName, DefaultVisFlags) + encodedEntryName, err := govis.Vis(entryPathName, DefaultVisFlags) if err != nil { return err }