Plex 로 Wiki Page Parser 를 만들ë�˜ì¤‘. Plex 는 아주 훌ë¥í•œ readability ì�˜ lexical analyzer code 를 만들ë�„ë¡� ë�„와준다.
Plex Example : Wiki Parser ¶
현재 PyKië�¼ëŠ”, 1002ê°€ ê°œì�¸ì �으로 만들어서 사용중ì�¸ 위키ì—�서ì�˜ parser í�´ëž˜ìФ 중 ì�¼ë¶€ 코드ì�´ë‹¤.
{{{~cpp
from Plex import *
import cStringIO, StringIO
class Parser:
def __init__(self, anInterWikiMap={}, aScriptName='', aMacros={}):
self.interWikiMap = anInterWikiMap
self.scriptName = aScriptName
self.macros = aMacros
def makeToStream(self, aText):
return cStringIO.StringIO(aText)
def parse(self, aText):
stream = self.makeToStream(aText)
return WikiParser(stream, self.interWikiMap,self.scriptName, self.macros).linkedLine()
class WikiParser(Scanner):
def repl_bold(self, aText):
self.bold = not self.bold
return ("<B>","</B>")[not self.bold]
def repl_italic(self, aText):
self.italic = not self.italic
return ("<I>","</I>")[not self.italic]
def repl_boldAndItalic(self, aText):
self.italic = not self.italic
return ("<I><B>","</B></I>")[not self.italic]
def repl_ruler(self, aText):
return "<HR>\n"
def repl_enter(self, aText):
return "<BR>\n"
def getLinkStr(self, anUrl):
return "<a href='%(url)s'>%(url)s</a>" % {'url':anUrl}
def getImageStr(self, anUrl):
return "<img src='%(url)s'>" % {'url':anUrl}
def isImage(self, aFileUrl):
imgExtensions = ("jpg", "png", "gif", "jpeg")
extension = aFileUrl[aFileUrl.rfind(".")+1:]
return extension in imgExtensions
def repl_url(self, aText):
if self.isImage(aText):
return self.getImageStr(aText)
return self.getLinkStr(aText)
def repl_interwiki(self, aText):
wikiMapName,pageName = self.text.split(":")
if self.interWikiMap.has_key(wikiMapName):
url = self.interWikiMap[wikiMapName]
imageTag = "<IMG SRC=../intertag.gif BORDER=0>"
formatText = "<A HREF=%(url)s>%(imageTag)s</A><A HREF=%(url)s%(pageName)s>%(pageName)s</A>"
else:
url = self.scriptName+"/InterMap"
imageTag = "<IMG SRC=../intertag.gif BORDER=1 ALT='Bad:%s'>"%(wikiMapName)
formatText="<A HREF=%(url)s>%(imageTag)s</A><A HREF=%(url)s>%(pageName)s</A>"
return formatText % {'url':url, 'pageName':pageName, 'imageTag':imageTag}
def repl_pagelink(self, aText):
if aText == "NonExistPage":
return "<a class=nonexist href='%(scriptName)s/%(pageName)s'>%(pageName)s</a>" % {'scriptName':self.scriptName, 'pageName':aText}
return "<a href='%(scriptName)s/%(pageName)s'>%(pageName)s</a>" \
% {'scriptName':self.scriptName, 'pageName':aText}
def repl_pagelinkUsingUnderbar(self, aText):
pageName = aText[:-1]
return "<a href='%(scriptName)s/%(pageName)s'>%(pageName)s</a>" % {'scriptName':self.scriptName, 'pageName':pageName}
def repl_macro(self, aText):
class NoneMacro:
def execute(self):
return ''
macroName = aText[2:-2]
macro=self.macros.get(macroName) or NoneMacro()
return macro.execute()
def repl_normalString(self, aText):
return aText
# for auto link
urlHeader = Str("http://")
stringUntilSpace = Rep(AnyBut(" "))
url = urlHeader + stringUntilSpace
upperCase = Range('AZ')
lowerCase = Range('az')
upperCaseSequence = Rep1(upperCase)
lowerCaseSequence = Rep1(lowerCase)
alphabetSequence = Rep1(upperCaseSequence | lowerCaseSequence)
interwikiDelim = Str(":")
stringUntilSpaceOrCommaOrRest = Rep1(AnyBut(" .,"))
interwiki = alphabetSequence + interwikiDelim + stringUntilSpaceOrCommaOrRest
pagelinkUsingCamelWord = upperCase + Rep(lowerCase) + Rep1(upperCaseSequence + lowerCaseSequence)
underbarForLink = Str('_')
pagelinkUsingUnderbar = Rep1(AnyBut('{\n_ ')) + underbarForLink
macro = Str('[[') + Rep1(AnyBut(']]')) + Str(']]')
space = Str(" ")
# for tag
bold = Str("'''")
italic = Str("''")
boldanditalic = bold + italic
htmllefttag = Str("<")
htmlrighttag = Str(">")
htmlandtag = Str("&")
enterCode = Str("\n")
ruler = Str("----") + enterCode
rawTextStart=Str("{{{~cpp ")
rawTextEnd=Str("} } }")
def repl_rawTextStart(self, aText):
self.begin("rawtext")
return "<PRE>"
def repl_rawTextEnd(self, aText):
self.begin("")
return "</PRE>"
lexicon = Lexicon([
(rawTextStart, repl_rawTextStart),
State('rawtext', [
(rawTextEnd, repl_rawTextEnd),
(AnyChar, repl_normalString),
]),
(italic, repl_italic),
(bold, repl_bold),
(boldanditalic, repl_boldAndItalic),
(htmllefttag, "<"),
(htmlrighttag, ">"),
(htmlandtag, "&"),
(url, repl_url),
(interwiki, repl_interwiki),
(pagelinkUsingCamelWord, repl_pagelink),
(ruler, repl_ruler),
(enterCode, repl_enter),
(pagelinkUsingUnderbar, repl_pagelinkUsingUnderbar),
(macro, repl_macro),
(AnyChar | space, repl_normalString),])
def __init__(self, aStream, anInterWikiMap={}, aScriptName='pyki.cgi', aMacroList={}):
Scanner.__init__(self, self.lexicon, aStream)
self.interWikiMap=anInterWikiMap
self.scriptName = aScriptName
self.macros = aMacroList
self.bold = False
self.italic = False
def linkedLine(self):
writer = StringIO.StringIO("")
while True:
token = self.read()
if token[0] is None:
break
writer.write(token[0])
return writer.getvalue()
}}}
=== 개발중 �슈 ===
처ì�Œì—�는 Wiki ì—�서 Tag ì—� 대해 Tagger í�´ëž˜ìŠ¤ë¥¼ ë§Œë“¤ê³ , link 를 걸어주는 부분ì—� 대해 AutoLinker 를, Macro ì—�는 MacroApplyer 를 ê°�ê°� 만들어주었다. 그러다가 ë¬¸ì œê°€ ìƒ�겼는ë�°, 태그중ì—� ê·¸ ì˜�í–¥ë ¥ì�´ 겹치는 부분ì�´ ìƒ�겨나게 ë�œ 것ì�´ë‹¤. 즉, 예를 ë“ ë‹¤ë©´ Macro ì�˜ 경우 CamelWord 로 ì�´ë¦„지어지기ë�„ 하는ë�°, ì�´ëŠ” AutoLinker ì�˜ apply 를 거치면서 archor 태그가 붙어버리는 것ì�´ë‹¤.
해결방법 : ë‘�가지ì�¸ë�°, 하나는 AutoLinker ì—�서 Macro ê´€ë ¨ 태그시 ë¬´ì‹œí•˜ê³ ì§€ë‚˜ê°€ëŠ” 방법ì�´ê³ 하나는 AutoLinker 와 MacroApplyer를 통합하는 방법ì�´ë‹¤.
ì „ìž�ì�˜ 경우 ê°�ê°�ì�˜ Class Responsibility 들ì�„ ìœ ì§€í•œë‹¤ëŠ” 장ì �ì�´ 있지만, AutoLinker ì—�서 ì›�래 ìƒ�ê°�치 않았ë�˜ 한가지 ì�¼ì�„ ë�” 해야 한다는 ì �ì�´ ìžˆê² ë‹¤.
후ìž�ì�˜ 경우 í�´ëž˜ìŠ¤ê°€ 커진다는 단ì �ì�´ 있지만, ì�˜ë�„한 lexical 들만 표현ë�œë‹¤ëŠ” ì �ê³¼ 1 pass 로 파싱ì�´ ê°™ì�´ ì�´ë£¨ì–´ì§ˆ 수 있다는 장ì �ì�´ 있다.
ê²°êµì�€ 후ìž�를 ì„ íƒ�하였다. ê·¼ë�°, 그러면서 ì�´ë²ˆì—” Tagger 와 AutoLinker 양쪽ì—� ì˜�í–¥ë ¥ì�„ ë¯¸ì¹ ê±°ë¦¬ê°€ ìƒ�겼는ë�°, 바로 í…�스트를 그대로 보여주는 태그부분ì�´ë‹¤.
그러나......~ 후ìž�로 ìˆ˜ì •í•˜ëŠ”ë�° 40ë¶„ë�„ 안걸리다.; 작업으로 본다면 Parser ë‘�ê°œì�˜ lexicon ì�„ 합치는 작업임ì—�ë�„ ë¶ˆêµ¬í•˜ê³ , ê·¸ ì•ˆì •ì„±ë�„ 보장받으면서 parser ì—�서 line 단위 ìž�르기 부분까지 ìˆ˜ì •í•˜ì˜€ë‹¤. 매 번 ìˆ˜ì •í• ë•Œë§ˆë‹¤ 테스트를 ë�Œë¦¬ë©´ì„œ 확ì�¸í–ˆê¸° 때문ì—� ê·¸ 결과가 보장ì�´ ë�˜ì—ˆë‹¤. Text Processing ì—�서 ì�´ëŸ¬í•œ 부분ì—� 대한 TDDì�˜ 파워는 ì •ë§� í�¬ë‹¤ëž€ ìƒ�ê°�ì�´ ë“ ë‹¤.
----
[Plex]








