This is an old revision of this page, as edited by Tsirel(talk | contribs) at 19:28, 6 January 2009(←Created page with '<pre> ###################################################################### # # # parse...'). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.Revision as of 19:28, 6 January 2009 by Tsirel(talk | contribs)(←Created page with '<pre> ###################################################################### # # # parse...')
######################################################################
# #
# parse.py #
# #
# Main infile "all.in"; its lines: "[[article]] tags". #
# Tags are space-separated, #
# each either "(rvnumber:rvtype)" #
# or a name (corename or aroundname) and maybe suffix "1" or "2". #
# Parameters from four infiles. #
# Outfile: all.dat (for format.py). #
# #
######################################################################
# read parameters
corenames = eval(open("corenames.dat").read())
aroundnames = eval(open("aroundnames.dat").read())
rvnumber = eval(open("rvnumber.dat").read())
rvtype = eval(open("rvtype.dat").read())
# only abbreviations will be used; full texts are for "format.py"
corenames = map ( (lambda(x,y):x), corenames )
aroundnames = map ( (lambda(x,y):x), aroundnames )
rvnumber = map ( (lambda(x,y):x), rvnumber )
rvtype = map ( (lambda(x,y):x), rvtype )
# distinction of "core" and "around" is for "format.py"
names = corenames + aroundnames
# parameters are ready; the main loop follows
biglist = [] # for parsed lines
notags = [] # for lines without tags (if any)
everything = {} # for articles (duplication check)
for line in open("all.in"): # main loop: lines of the main input file
line=line.strip()
if line=="": continue # ignore empty lines (if any)
if not line.startswith("[["):
print "BAD LINE: "+line # syntax error in the main input file
continue
article=line[2:line.index("]]")] # extract article from "[[article]] tags"
article=article[0].upper()+article[1:] # article -> Article
if article in everything: # duplication?
print "AGAIN "+article
everything[article]=0 # the 0 is of no use
rest=line[line.index("]]")+2:].split() # list of tags
if rest==[]: notags.append(article); continue # no tags: report, do not process
tags0=[] # for "(A:B)" tags converted to (A,B)
tags1=[] # for other tags
for tag in rest: # loop over all tags of the given line
if tag.startswith("("): # "(A:B)" tag?
maintag=tag[1:]
if not maintag.endswith(")"): print "!!! "+maintag; break # syntax error
maintag=maintag[:-1]
maintag1=maintag[:maintag.index(":")] # A from "(A:B)"
maintag2=maintag[maintag.index(":")+1:] # B from "(A:B)"
for x in maintag1: # syntax check
if x not in rvnumber:
print "?? "+x+" "+whole_line
raise "oops-1" # syntax error
for x in maintag2: # syntax check
if x not in rvtype:
print "?? "+x+" "+whole_line
raise "oops-2" # syntax error
tags0.append((maintag1,maintag2)) # store (A,B)
else: # other tag, not "A:B)"
assert len(tag)==3 or len(tag)==4 # 3 chars, and maybe 1 char suffix
if tag[0:3] not in names:
print "BAD TAG: "+tag
continue # syntax error
if len(tag)==4: assert tag[3] in ["1","2"] # suffix syntax check
tags1.append(tag) # store the other tag
biglist.append( (tags0,tags1,article) ) # store the parsed line
# the main loop is finished
print >>open("all.dat","w"), biglist # write main out file
if notags != []: # report lines with no tags (if any)
print "*** notags ***"
print notags
######################## the end ###################################
