sgjesse@chromium.org | c514574 | 2009-10-07 09:00:33 +0000 | [diff] [blame] | 1 | #!/usr/bin/python2.4 |
| 2 | |
| 3 | # Copyright 2009 the V8 project authors. All rights reserved. |
| 4 | # Redistribution and use in source and binary forms, with or without |
| 5 | # modification, are permitted provided that the following conditions are |
| 6 | # met: |
| 7 | # |
| 8 | # * Redistributions of source code must retain the above copyright |
| 9 | # notice, this list of conditions and the following disclaimer. |
| 10 | # * Redistributions in binary form must reproduce the above |
| 11 | # copyright notice, this list of conditions and the following |
| 12 | # disclaimer in the documentation and/or other materials provided |
| 13 | # with the distribution. |
| 14 | # * Neither the name of Google Inc. nor the names of its |
| 15 | # contributors may be used to endorse or promote products derived |
| 16 | # from this software without specific prior written permission. |
| 17 | # |
| 18 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
| 19 | # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
| 20 | # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
| 21 | # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
| 22 | # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
| 23 | # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
| 24 | # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
| 25 | # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
| 26 | # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| 27 | # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
| 28 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| 29 | |
| 30 | """A JavaScript minifier. |
| 31 | |
| 32 | It is far from being a complete JS parser, so there are many valid |
| 33 | JavaScript programs that will be ruined by it. Another strangeness is that |
| 34 | it accepts $ and % as parts of identifiers. It doesn't merge lines or strip |
| 35 | out blank lines in order to ease debugging. Variables at the top scope are |
| 36 | properties of the global object so we can't rename them. It is assumed that |
| 37 | you introduce variables with var as if JavaScript followed C++ scope rules |
| 38 | around curly braces, so the declaration must be above the first use. |
| 39 | |
| 40 | Use as: |
| 41 | import jsmin |
| 42 | minifier = JavaScriptMinifier() |
| 43 | program1 = minifier.JSMinify(program1) |
| 44 | program2 = minifier.JSMinify(program2) |
| 45 | """ |
| 46 | |
| 47 | import re |
| 48 | |
| 49 | |
| 50 | class JavaScriptMinifier(object): |
| 51 | """An object that you can feed code snippets to to get them minified.""" |
| 52 | |
| 53 | def __init__(self): |
| 54 | # We prepopulate the list of identifiers that shouldn't be used. These |
| 55 | # short language keywords could otherwise be used by the script as variable |
| 56 | # names. |
| 57 | self.seen_identifiers = {"do": True, "in": True} |
| 58 | self.identifier_counter = 0 |
| 59 | self.in_comment = False |
| 60 | self.map = {} |
| 61 | self.nesting = 0 |
| 62 | |
| 63 | def LookAtIdentifier(self, m): |
| 64 | """Records identifiers or keywords that we see in use. |
| 65 | |
| 66 | (So we can avoid renaming variables to these strings.) |
| 67 | Args: |
| 68 | m: The match object returned by re.search. |
| 69 | |
| 70 | Returns: |
| 71 | Nothing. |
| 72 | """ |
| 73 | identifier = m.group(1) |
| 74 | self.seen_identifiers[identifier] = True |
| 75 | |
| 76 | def Push(self): |
| 77 | """Called when we encounter a '{'.""" |
| 78 | self.nesting += 1 |
| 79 | |
| 80 | def Pop(self): |
| 81 | """Called when we encounter a '}'.""" |
| 82 | self.nesting -= 1 |
| 83 | # We treat each top-level opening brace as a single scope that can span |
| 84 | # several sets of nested braces. |
| 85 | if self.nesting == 0: |
| 86 | self.map = {} |
| 87 | self.identifier_counter = 0 |
| 88 | |
| 89 | def Declaration(self, m): |
| 90 | """Rewrites bits of the program selected by a regexp. |
| 91 | |
| 92 | These can be curly braces, literal strings, function declarations and var |
| 93 | declarations. (These last two must be on one line including the opening |
| 94 | curly brace of the function for their variables to be renamed). |
| 95 | |
| 96 | Args: |
| 97 | m: The match object returned by re.search. |
| 98 | |
| 99 | Returns: |
| 100 | The string that should replace the match in the rewritten program. |
| 101 | """ |
| 102 | matched_text = m.group(0) |
| 103 | if matched_text == "{": |
| 104 | self.Push() |
| 105 | return matched_text |
| 106 | if matched_text == "}": |
| 107 | self.Pop() |
| 108 | return matched_text |
| 109 | if re.match("[\"'/]", matched_text): |
| 110 | return matched_text |
| 111 | m = re.match(r"var ", matched_text) |
| 112 | if m: |
| 113 | var_names = matched_text[m.end():] |
| 114 | var_names = re.split(r",", var_names) |
| 115 | return "var " + ",".join(map(self.FindNewName, var_names)) |
| 116 | m = re.match(r"(function\b[^(]*)\((.*)\)\{$", matched_text) |
| 117 | if m: |
| 118 | up_to_args = m.group(1) |
| 119 | args = m.group(2) |
| 120 | args = re.split(r",", args) |
| 121 | self.Push() |
| 122 | return up_to_args + "(" + ",".join(map(self.FindNewName, args)) + "){" |
| 123 | |
| 124 | if matched_text in self.map: |
| 125 | return self.map[matched_text] |
| 126 | |
| 127 | return matched_text |
| 128 | |
| 129 | def CharFromNumber(self, number): |
| 130 | """A single-digit base-52 encoding using a-zA-Z.""" |
| 131 | if number < 26: |
| 132 | return chr(number + 97) |
| 133 | number -= 26 |
| 134 | return chr(number + 65) |
| 135 | |
| 136 | def FindNewName(self, var_name): |
| 137 | """Finds a new 1-character or 2-character name for a variable. |
| 138 | |
| 139 | Enters it into the mapping table for this scope. |
| 140 | |
| 141 | Args: |
| 142 | var_name: The name of the variable before renaming. |
| 143 | |
| 144 | Returns: |
| 145 | The new name of the variable. |
| 146 | """ |
| 147 | new_identifier = "" |
| 148 | # Variable names that end in _ are member variables of the global object, |
| 149 | # so they can be visible from code in a different scope. We leave them |
| 150 | # alone. |
| 151 | if var_name in self.map: |
| 152 | return self.map[var_name] |
| 153 | if self.nesting == 0: |
| 154 | return var_name |
| 155 | while True: |
| 156 | identifier_first_char = self.identifier_counter % 52 |
| 157 | identifier_second_char = self.identifier_counter / 52 |
| 158 | new_identifier = self.CharFromNumber(identifier_first_char) |
| 159 | if identifier_second_char != 0: |
| 160 | new_identifier = ( |
| 161 | self.CharFromNumber(identifier_second_char - 1) + new_identifier) |
| 162 | self.identifier_counter += 1 |
| 163 | if not new_identifier in self.seen_identifiers: |
| 164 | break |
| 165 | |
| 166 | self.map[var_name] = new_identifier |
| 167 | return new_identifier |
| 168 | |
| 169 | def RemoveSpaces(self, m): |
| 170 | """Returns literal strings unchanged, replaces other inputs with group 2. |
| 171 | |
| 172 | Other inputs are replaced with the contents of capture 1. This is either |
| 173 | a single space or an empty string. |
| 174 | |
| 175 | Args: |
| 176 | m: The match object returned by re.search. |
| 177 | |
| 178 | Returns: |
| 179 | The string that should be inserted instead of the matched text. |
| 180 | """ |
| 181 | entire_match = m.group(0) |
| 182 | replacement = m.group(1) |
| 183 | if re.match(r"'.*'$", entire_match): |
| 184 | return entire_match |
| 185 | if re.match(r'".*"$', entire_match): |
| 186 | return entire_match |
| 187 | if re.match(r"/.+/$", entire_match): |
| 188 | return entire_match |
| 189 | return replacement |
| 190 | |
| 191 | def JSMinify(self, text): |
| 192 | """The main entry point. Takes a text and returns a compressed version. |
| 193 | |
| 194 | The compressed version hopefully does the same thing. Line breaks are |
| 195 | preserved. |
| 196 | |
| 197 | Args: |
| 198 | text: The text of the code snippet as a multiline string. |
| 199 | |
| 200 | Returns: |
| 201 | The compressed text of the code snippet as a multiline string. |
| 202 | """ |
| 203 | new_lines = [] |
| 204 | for line in re.split(r"\n", text): |
| 205 | line = line.replace("\t", " ") |
| 206 | if self.in_comment: |
| 207 | m = re.search(r"\*/", line) |
| 208 | if m: |
| 209 | line = line[m.end():] |
| 210 | self.in_comment = False |
| 211 | else: |
| 212 | new_lines.append("") |
| 213 | continue |
| 214 | |
| 215 | if not self.in_comment: |
| 216 | line = re.sub(r"/\*.*?\*/", " ", line) |
| 217 | line = re.sub(r"//.*", "", line) |
| 218 | m = re.search(r"/\*", line) |
| 219 | if m: |
| 220 | line = line[:m.start()] |
| 221 | self.in_comment = True |
| 222 | |
| 223 | # Strip leading and trailing spaces. |
| 224 | line = re.sub(r"^ +", "", line) |
| 225 | line = re.sub(r" +$", "", line) |
| 226 | # A regexp that matches a literal string surrounded by "double quotes". |
| 227 | # This regexp can handle embedded backslash-escaped characters including |
| 228 | # embedded backslash-escaped double quotes. |
| 229 | double_quoted_string = r'"(?:[^"\\]|\\.)*"' |
| 230 | # A regexp that matches a literal string surrounded by 'double quotes'. |
| 231 | single_quoted_string = r"'(?:[^'\\]|\\.)*'" |
| 232 | # A regexp that matches a regexp literal surrounded by /slashes/. |
fschneider@chromium.org | 0c20e67 | 2010-01-14 15:28:53 +0000 | [diff] [blame] | 233 | # Don't allow a regexp to have a ) before the first ( since that's a |
| 234 | # syntax error and it's probably just two unrelated slashes. |
erik.corry@gmail.com | bbceb57 | 2012-03-09 10:52:05 +0000 | [diff] [blame^] | 235 | # Also don't allow it to come after anything that can only be the |
| 236 | # end of a primary expression. |
| 237 | slash_quoted_regexp = r"(?<![\w$'\")\]])/(?:(?=\()|(?:[^()/\\]|\\.)+)(?:\([^/\\]|\\.)*/" |
sgjesse@chromium.org | c514574 | 2009-10-07 09:00:33 +0000 | [diff] [blame] | 238 | # Replace multiple spaces with a single space. |
| 239 | line = re.sub("|".join([double_quoted_string, |
| 240 | single_quoted_string, |
| 241 | slash_quoted_regexp, |
| 242 | "( )+"]), |
| 243 | self.RemoveSpaces, |
| 244 | line) |
| 245 | # Strip single spaces unless they have an identifier character both before |
| 246 | # and after the space. % and $ are counted as identifier characters. |
| 247 | line = re.sub("|".join([double_quoted_string, |
| 248 | single_quoted_string, |
| 249 | slash_quoted_regexp, |
| 250 | r"(?<![a-zA-Z_0-9$%]) | (?![a-zA-Z_0-9$%])()"]), |
| 251 | self.RemoveSpaces, |
| 252 | line) |
| 253 | # Collect keywords and identifiers that are already in use. |
| 254 | if self.nesting == 0: |
| 255 | re.sub(r"([a-zA-Z0-9_$%]+)", self.LookAtIdentifier, line) |
| 256 | function_declaration_regexp = ( |
| 257 | r"\bfunction" # Function definition keyword... |
| 258 | r"( [\w$%]+)?" # ...optional function name... |
| 259 | r"\([\w$%,]+\)\{") # ...argument declarations. |
| 260 | # Unfortunately the keyword-value syntax { key:value } makes the key look |
| 261 | # like a variable where in fact it is a literal string. We use the |
| 262 | # presence or absence of a question mark to try to distinguish between |
| 263 | # this case and the ternary operator: "condition ? iftrue : iffalse". |
| 264 | if re.search(r"\?", line): |
| 265 | block_trailing_colon = r"" |
| 266 | else: |
| 267 | block_trailing_colon = r"(?![:\w$%])" |
| 268 | # Variable use. Cannot follow a period precede a colon. |
| 269 | variable_use_regexp = r"(?<![.\w$%])[\w$%]+" + block_trailing_colon |
| 270 | line = re.sub("|".join([double_quoted_string, |
| 271 | single_quoted_string, |
| 272 | slash_quoted_regexp, |
| 273 | r"\{", # Curly braces. |
| 274 | r"\}", |
| 275 | r"\bvar [\w$%,]+", # var declarations. |
| 276 | function_declaration_regexp, |
| 277 | variable_use_regexp]), |
| 278 | self.Declaration, |
| 279 | line) |
| 280 | new_lines.append(line) |
| 281 | |
| 282 | return "\n".join(new_lines) + "\n" |