webui fixes and tidy ups.
[pyTivo/wmcbrine/lucasnz.git] / Cheetah / SourceReader.py
blob025cba530783d967d5b79b3799b9cc619e2b1e8d
1 #!/usr/bin/env python
2 # $Id: SourceReader.py,v 1.15 2007/04/03 01:57:42 tavis_rudd Exp $
3 """SourceReader class for Cheetah's Parser and CodeGenerator
5 Meta-Data
6 ================================================================================
7 Author: Tavis Rudd <tavis@damnsimple.com>
8 License: This software is released for unlimited distribution under the
9 terms of the MIT license. See the LICENSE file.
10 Version: $Revision: 1.15 $
11 Start Date: 2001/09/19
12 Last Revision Date: $Date: 2007/04/03 01:57:42 $
13 """
14 __author__ = "Tavis Rudd <tavis@damnsimple.com>"
15 __revision__ = "$Revision: 1.15 $"[11:-2]
17 import re
18 import sys
20 EOLre = re.compile(r'[ \f\t]*(?:\r\n|\r|\n)')
21 EOLZre = re.compile(r'(?:\r\n|\r|\n|\Z)')
22 ENCODINGsearch = re.compile("coding[=:]\s*([-\w.]+)").search
24 class Error(Exception):
25 pass
27 class SourceReader:
28 def __init__(self, src, filename=None, breakPoint=None, encoding=None):
30 ## @@TR 2005-01-17: the following comes from a patch Terrel Shumway
31 ## contributed to add unicode support to the reading of Cheetah source
32 ## files with dynamically compiled templates. All the existing unit
33 ## tests pass but, it needs more testing and some test cases of its
34 ## own. My instinct is to move this up into the code that passes in the
35 ## src string rather than leaving it here. As implemented here it
36 ## forces all src strings to unicode, which IMO is not what we want.
37 # if encoding is None:
38 # # peek at the encoding in the first two lines
39 # m = EOLZre.search(src)
40 # pos = m.end()
41 # if pos<len(src):
42 # m = EOLZre.search(src,pos)
43 # pos = m.end()
44 # m = ENCODINGsearch(src,0,pos)
45 # if m:
46 # encoding = m.group(1)
47 # else:
48 # encoding = sys.getfilesystemencoding()
49 # self._encoding = encoding
50 # if type(src) is not unicode:
51 # src = src.decode(encoding)
52 ## end of Terrel's patch
54 self._src = src
55 self._filename = filename
57 self._srcLen = len(src)
58 if breakPoint == None:
59 self._breakPoint = self._srcLen
60 else:
61 self.setBreakPoint(breakPoint)
62 self._pos = 0
63 self._bookmarks = {}
64 self._posTobookmarkMap = {}
66 ## collect some meta-information
67 self._EOLs = []
68 pos = 0
69 while pos < len(self):
70 EOLmatch = EOLZre.search(src, pos)
71 self._EOLs.append(EOLmatch.start())
72 pos = EOLmatch.end()
74 self._BOLs = []
75 for pos in self._EOLs:
76 BOLpos = self.findBOL(pos)
77 self._BOLs.append(BOLpos)
79 def src(self):
80 return self._src
82 def filename(self):
83 return self._filename
85 def __len__(self):
86 return self._breakPoint
88 def __getitem__(self, i):
89 self.checkPos(i)
90 return self._src[i]
92 def __getslice__(self, i, j):
93 i = max(i, 0); j = max(j, 0)
94 return self._src[i:j]
96 def splitlines(self):
97 if not hasattr(self, '_srcLines'):
98 self._srcLines = self._src.splitlines()
99 return self._srcLines
101 def lineNum(self, pos=None):
102 if pos == None:
103 pos = self._pos
105 for i in range(len(self._BOLs)):
106 if pos >= self._BOLs[i] and pos <= self._EOLs[i]:
107 return i
109 def getRowCol(self, pos=None):
110 if pos == None:
111 pos = self._pos
112 lineNum = self.lineNum(pos)
113 BOL, EOL = self._BOLs[lineNum], self._EOLs[lineNum]
114 return lineNum+1, pos-BOL+1
116 def getRowColLine(self, pos=None):
117 if pos == None:
118 pos = self._pos
119 row, col = self.getRowCol(pos)
120 return row, col, self.splitlines()[row-1]
122 def getLine(self, pos):
123 if pos == None:
124 pos = self._pos
125 lineNum = self.lineNum(pos)
126 return self.splitlines()[lineNum]
128 def pos(self):
129 return self._pos
131 def setPos(self, pos):
132 self.checkPos(pos)
133 self._pos = pos
136 def validPos(self, pos):
137 return pos <= self._breakPoint and pos >=0
139 def checkPos(self, pos):
140 if not pos <= self._breakPoint:
141 raise Error("pos (" + str(pos) + ") is invalid: beyond the stream's end (" +
142 str(self._breakPoint-1) + ")" )
143 elif not pos >=0:
144 raise Error("pos (" + str(pos) + ") is invalid: less than 0" )
146 def breakPoint(self):
147 return self._breakPoint
149 def setBreakPoint(self, pos):
150 if pos > self._srcLen:
151 raise Error("New breakpoint (" + str(pos) +
152 ") is invalid: beyond the end of stream's source string (" +
153 str(self._srcLen) + ")" )
154 elif not pos >= 0:
155 raise Error("New breakpoint (" + str(pos) + ") is invalid: less than 0" )
157 self._breakPoint = pos
159 def setBookmark(self, name):
160 self._bookmarks[name] = self._pos
161 self._posTobookmarkMap[self._pos] = name
163 def hasBookmark(self, name):
164 return self._bookmarks.has_key(name)
166 def gotoBookmark(self, name):
167 if not self.hasBookmark(name):
168 raise Error("Invalid bookmark (" + name + ") is invalid: does not exist")
169 pos = self._bookmarks[name]
170 if not self.validPos(pos):
171 raise Error("Invalid bookmark (" + name + ', '+
172 str(pos) + ") is invalid: pos is out of range" )
173 self._pos = pos
175 def atEnd(self):
176 return self._pos >= self._breakPoint
178 def atStart(self):
179 return self._pos == 0
181 def peek(self, offset=0):
182 self.checkPos(self._pos+offset)
183 pos = self._pos + offset
184 return self._src[pos]
186 def getc(self):
187 pos = self._pos
188 if self.validPos(pos+1):
189 self._pos += 1
190 return self._src[pos]
192 def ungetc(self, c=None):
193 if not self.atStart():
194 raise Error('Already at beginning of stream')
196 self._pos -= 1
197 if not c==None:
198 self._src[self._pos] = c
200 def advance(self, offset=1):
201 self.checkPos(self._pos + offset)
202 self._pos += offset
204 def rev(self, offset=1):
205 self.checkPos(self._pos - offset)
206 self._pos -= offset
208 def read(self, offset):
209 self.checkPos(self._pos + offset)
210 start = self._pos
211 self._pos += offset
212 return self._src[start:self._pos]
214 def readTo(self, to, start=None):
215 self.checkPos(to)
216 if start == None:
217 start = self._pos
218 self._pos = to
219 return self._src[start:to]
222 def readToEOL(self, start=None, gobble=True):
223 EOLmatch = EOLZre.search(self.src(), self.pos())
224 if gobble:
225 pos = EOLmatch.end()
226 else:
227 pos = EOLmatch.start()
228 return self.readTo(to=pos, start=start)
231 def find(self, it, pos=None):
232 if pos == None:
233 pos = self._pos
234 return self._src.find(it, pos )
236 def startswith(self, it, pos=None):
237 if self.find(it, pos) == self.pos():
238 return True
239 else:
240 return False
242 def rfind(self, it, pos):
243 if pos == None:
244 pos = self._pos
245 return self._src.rfind(it, pos)
247 def findBOL(self, pos=None):
248 if pos == None:
249 pos = self._pos
250 src = self.src()
251 return max(src.rfind('\n',0,pos)+1, src.rfind('\r',0,pos)+1, 0)
253 def findEOL(self, pos=None, gobble=False):
254 if pos == None:
255 pos = self._pos
257 match = EOLZre.search(self.src(), pos)
258 if gobble:
259 return match.end()
260 else:
261 return match.start()
263 def isLineClearToPos(self, pos=None):
264 if pos == None:
265 pos = self.pos()
266 self.checkPos(pos)
267 src = self.src()
268 BOL = self.findBOL()
269 return BOL == pos or src[BOL:pos].isspace()
271 def matches(self, strOrRE):
272 if isinstance(strOrRE, (str, unicode)):
273 return self.startswith(strOrRE, pos=self.pos())
274 else: # assume an re object
275 return strOrRE.match(self.src(), self.pos())
277 def matchWhiteSpace(self, WSchars=' \f\t'):
278 return (not self.atEnd()) and self.peek() in WSchars
280 def getWhiteSpace(self, max=None, WSchars=' \f\t'):
281 if not self.matchWhiteSpace(WSchars):
282 return ''
283 start = self.pos()
284 breakPoint = self.breakPoint()
285 if max is not None:
286 breakPoint = min(breakPoint, self.pos()+max)
287 while self.pos() < breakPoint:
288 self.advance()
289 if not self.matchWhiteSpace(WSchars):
290 break
291 return self.src()[start:self.pos()]
293 def matchNonWhiteSpace(self, WSchars=' \f\t\n\r'):
294 return self.atEnd() or not self.peek() in WSchars
296 def getNonWhiteSpace(self, WSchars=' \f\t\n\r'):
297 if not self.matchNonWhiteSpace(WSchars):
298 return ''
299 start = self.pos()
300 while self.pos() < self.breakPoint():
301 self.advance()
302 if not self.matchNonWhiteSpace(WSchars):
303 break
304 return self.src()[start:self.pos()]