Re: Spyce compiler newline sensitivity?

Hi Philip,

the culprit was a long-standing bug in the handling of multiline string
constants.  Change the findMultiLineQuote function in spyceCompile.py to
the following:

  def findMultiLineQuote(s):
    quotelist = []
    def eatToken(type, string, begin, end, _, quotelist=quotelist):
      if type == token.STRING:
        if string.startswith('r"""') or string.startswith("r'''"):
          quotelist.append((string, begin, end))
    tokenize.tokenize(StringIO(s).readline, eatToken)
    return quotelist

(I'll commit this to subversion as soon as sourceforge gets their server
back up.  Grr.)

-J

On Wed, 31 May 2006 23:30:12 -0400, "Philip Semanchuk"
<[email protected]> said:
> Hi all,
> It's Philip with another weird problem. I use Spyce to generate Spyce  
> code which has worked well up until this point. Today I found some code  
> that causes the compilation step to exit ungracefully  
> ("spyceSyntaxError: <unprintable instance object>"). What's really odd  
> is that the code compiles OK if I save it in TextEdit (I'm on a Mac)  
> and only fails to compile if I save it in BBEdit, which I use to save  
> *all* of my Spyce, Python and HTML files and have not had a problem  
> with up until this point. The files only differ in their newlines.
> 
> Here's the code which has been stripped down to a minimal test case:
> [[\
> OpenBrackets = "[" * 2
> CloseBrackets = "]" * 2
> 
> s = r"""
> 
> %(OpenBrackets)s\
> 
> if True == False:
>      pass
> 
> %(CloseBrackets)s
> 
> """ % vars()
> 
> ]]
> 
> 
> A hex dump of the files (compiling and non-compiling versions) shows  
> that they're the same size, but the TextEdit version uses 0x0D as  
> newlines and ends the file with 0x0D 0x0A, while the BBEdit version  
> uses 0x0A as newlines and ends the file with 0x0A 0x0A.
> 
> I created a tarball and a zip file (same contents in each) that contain  
> the two versions of the code as well as a hex dump of each. Those files  
> are here:
> http://semanchuk.com/philip/temp/SpyceNewlineCode.tar.gz
> http://semanchuk.com/philip/temp/SpyceNewlineCode.zip
> 
> I'd appreciate it if anyone else would be willing to try to compile  
> these files and report the results.
> 
> Thanks
> Philip
> 
> 
> PS - here's the error I get with the version that fails to compile:
> $ /usr/local/spyce-2.0.3/spyceCmd.py -c CompileFails.spy
> cache miss for ('/usr/local/spyce-2.0.3/modules/error.py',  
> 'defaultErrorTemplate')
> cache miss for ('/usr/local/spyce-2.0.3/tags/core.py', 'core')
> cache miss for ('/usr/local/spyce-2.0.3/tags/form.py', 'form')
> generating new spyceCode for test4.spy
> cache hit (1116524720) for core in /usr/local/spyce-2.0.3/tags/core.py
> cache hit (1115998066) for form in /usr/local/spyce-2.0.3/tags/form.py
> Traceback (most recent call last):
>    File "/usr/local/spyce-2.0.3/spyceCmd.py", line 297, in spyceMain
>      s = server.spyce_cache['file', script]
>    File "/usr/local/spyce-2.0.3/spyceCache.py", line 145, in __getitem__
>      return self.get(key)
>    File "/usr/local/spyce-2.0.3/spyceCache.py", line 133, in get
>      self.cache[key] = self.generate(key)
>    File "/usr/local/spyce-2.0.3/spyce.py", line 819, in  
> spyceCacheGenerate
>      return {
>    File "/usr/local/spyce-2.0.3/spyce.py", line 798, in  
> spyceFileCacheGenerate
>      s = spyceCode(code, filename=filename, sig=sig)
>    File "/usr/local/spyce-2.0.3/spyce.py", line 492, in __init__
>      self._code, self._coderefs, self._modrefs = \
>    File "/usr/local/spyce-2.0.3/spyceCompile.py", line 1437, in  
> spyceCompile
>      refs = emitPython(out, bracedPythonCode, refs)
>    File "/usr/local/spyce-2.0.3/spyceCompile.py", line 1384, in  
> emitPython
>      tokenize.tokenize(StringIO(bracedPythonCode).readline, eatToken)
>    File  
> "/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
> python2.3/tokenize.py", line 153, in tokenize
>      tokenize_loop(readline, tokeneater)
>    File  
> "/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
> python2.3/tokenize.py", line 160, in tokenize_loop
>      tokeneater(*token_info)
>    File "/usr/local/spyce-2.0.3/spyceCompile.py", line 1382, in eatToken
>      raise spyceException.spyceSyntaxError(sys.exc_info()[0])
> spyceSyntaxError: <unprintable instance object>
> 
> 
> 
> 
> -------------------------------------------------------
> All the advantages of Linux Managed Hosting--Without the Cost and Risk!
> Fully trained technicians. The highest number of Red Hat certifications
> in
> the hosting industry. Fanatical Support. Click to learn more
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=107521&bid=248729&dat=121642
> _______________________________________________
> Spyce-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/spyce-users
-- 
C++ is history repeated as tragedy. Java is history repeated as farce.  --Scott McKay



-------------------------------------------------------
All the advantages of Linux Managed Hosting--Without the Cost and Risk!
Fully trained technicians. The highest number of Red Hat certifications in
the hosting industry. Fanatical Support. Click to learn more
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=107521&bid=248729&dat=121642