# Copyright 2014 DWARF Debugging Information Format Committee
#
-# Looks at dwarfnamecmds.tex to find all the commands
+# Looks at dwarfnamecmds.tex(+) to find all the commands
# And sees what is actually used and whether references
# have definitions.
# The initial implemenation just *assumes* it is run from the tools
# Run as (for example)
# python refer.py
+# This is the simplest 'parse' of the .tex that we can manage
+# while still finding what we want to find.
+# One would hardly call it a parser, really.
+
+# There are essentially three namespaces at present in the document.
+# The hyperlink/hypertarget namespace.
+# The label vref ref namespace
+# which also involves our refersec referfol.
+# The index namespace, which we are not presently filling in very much.
import sys
import fileio
+# These two hold the commands we care about so we can
+# bypass most lines easily in phase two.
# All the newdwfnamecommands
global dwfnamecommsdict
# All the newcommand instances.
global newcommsdict
-# LINK suffix
-global linksdict
-
-# TARG suffix hyperlink of 'chap:#1' Index #2
-global targsdict
-# INDX suffix index #2
-global indxdict
-# MARK suffix hypertarget chap:#1 and index #2
-global markdict
-
# NAME suffix. No index, just text shows.
+# Not very useful so far
global namedict
-# This is the 'normal, most common case in running text.
-# Default implies index #2 and hyperlink of chap:#1
-global defaultdict
-# Targets for \hypertarget and livetarg[i]
+# Targets for \hypertarget and \hyperlink
+#targhyperdict is all the \hyperarget instances.
+#linkhyperdict is all the \hyperlink instances.
global targhyperdict
-# Links using hyperlink.
global linkhyperdict
# label targets (labels)
global labelrefdict
# The index content for named things.
+# The strings here are the links, the targets are
+# built by the latex index software.
+# So this dictionary is not really needed here (yet)
+# and is not yet fully built up.
global indexdict
dwfnamecommsdict = {}
# Links meaning \livelink \livetarg \livetargi macros
-linksdict = {}
-targsdict = {}
-indxdict = {}
-markdict = {}
+#linksdict = {}
+#targsdict = {}
namedict = {}
-defaultdict = {}
targhyperdict= {}
linkhyperdict= {}
labeldict = {}
return ''.join(t._tex)
-def pickup(linetoks,tnumin,pattern,myfile,linenum):
+def pickup(linetoks,tnumin,pattern,myfile,linenum,suppresserr):
""" The token pattern characters are
i meaning identifier
e meaning identifier, but ifnext token is }
for c in pattern:
patterncharnum = patterncharnum + 1
if curnum >= inlen:
- print "ERROR line ended surprisingly, pattern ", pattern," line ",linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line ended surprisingly, pattern ", pattern," line ",linenum," file ",myfile._name
return outtoks,numabsorbed
curtok = linetoks[curnum]
if c == " ":
while dwspace(curtok) == "y":
curnum = curnum + 1
if curnum >= inlen:
- print "ERROR line ended surprisingly in space, pattern ", pattern, " line ",linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line ended surprisingly in space, pattern ", pattern, " line ",linenum," file ",myfile._name
return outtoks,numabsorbed
numabsorbed = numabsorbed + 1
curtok = linetoks[curnum]
continue
elif c == "i":
if curtok._class != "id":
- print "ERROR line expected identifier got ",curtok._tex, "pattern" , pattern, " line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected identifier got ",curtok._tex, "pattern" , pattern, " line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
numabsorbed = numabsorbed + 1
outtoks += [curtok]
# Do not update location.
continue
else:
- print "ERROR line expected identifier got ",curtok._tex, "pattern" , pattern, " line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected identifier got ",curtok._tex, "pattern" , pattern, " line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
else:
numabsorbed = numabsorbed + 1
curnum = curnum + 1
numabsorbed = numabsorbed + 1
else:
- print "ERROR line expected { got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected { got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
elif c == "}":
if isbrace(curtok,"}") == "y":
curnum = curnum + 1
numabsorbed = numabsorbed + 1
else:
- print "ERROR line expected } got ",curtok._tex,"pattern",pattern," line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected } got ",curtok._tex,"pattern",pattern," line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
elif c == "[":
if isbrace(curtok,"[") == "y":
curnum = curnum + 1
numabsorbed = numabsorbed + 1
else:
- print "ERROR line expected [ got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected [ got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
elif c == "]":
if isbrace(curtok,"]") == "y":
curnum = curnum + 1
numabsorbed = numabsorbed + 1
else:
- print "ERROR line expected ] got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR line expected ] got ",curtok._tex," pattern ",pattern," line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
elif c == "*":
outlist = []
if curnum >= inlen:
outtoks += [outlist]
if patterncharnum < (len(pattern) -1):
- print "ERROR insufficient tokens on line for pattern ", pattern," line " ,linenum," file ",myfile._name
+ if suppresserr == "n":
+ print "ERROR insufficient tokens on line for pattern ", pattern," line " ,linenum," file ",myfile._name
return outtoks,numabsorbed
numabsorbed = numabsorbed + 1
curtok = linetoks[curnum]
# Found a right brace, so done here.
outtoks += [outlist]
else:
- print "ERROR pattern had unexpected character ",pattern
- sys.exit(1)
+ if suppresserr == "n":
+ print "ERROR pattern had unexpected character ",pattern
return outtoks,numabsorbed
def printbadcommand(name,myfile,myline):
+# Here we try two different parses, the [] is optional
+# with simplenametable.
def processbegin(linetoks,tnumin,myfile,linenum):
global targhyperdict
global indexdict
lbracecount = countbraces(linetoks,tnumin)
if lbracecount < 3:
return 1
- ourtoks,inlen = pickup(linetoks,tnumin," i { i } [ * ] { * } { i }",myfile,linenum)
- if len(ourtoks) < 12:
- return 1
+ ourtoks,inlen = pickup(linetoks,tnumin," i { i } [ * ] { * } { i }",myfile,linenum,"y")
+ if len(ourtoks) == 13:
+ lcom = ourtoks[2]
+ lcomname = toknamestring(lcom)
+ if lcomname != "simplenametable":
+ return inlen
+ targ = ourtoks[11]
+ hypstr = toknamestring(targ)
+ hypmen = tokmention(targ,myfile,linenum)
+ applytodict(labeldict,hypstr,hypmen)
+ return inlen
+ ourtoks,inlen = pickup(linetoks,tnumin," i { i } { * } { i }",myfile,linenum,"y")
+ if len(ourtoks) < 10:
+ return inlen
lcom = ourtoks[2]
lcomname = toknamestring(lcom)
if lcomname != "simplenametable":
- return inlen
- targ = ourtoks[11]
+ return inlen
+ targ = ourtoks[8]
hypstr = toknamestring(targ)
hypmen = tokmention(targ,myfile,linenum)
applytodict(labeldict,hypstr,hypmen)
global targhyperdict
global indexdict
t = linetoks[tnumin]
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { * }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { * }",myfile,linenum,"n")
if len(ourtoks) > 5:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
global targhyperdict
global indexdict
t = linetoks[tnumin]
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { e } { * }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { e } { * }",myfile,linenum,"n")
if len(ourtoks) > 5:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
global linkhyperdict
global indexdict
t = linetoks[tnumin]
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { * }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i } { * }",myfile,linenum,"n")
if len(ourtoks) > 5:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
""" \label{alabel} """
global labeldict
t = linetoks[tnumin]
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum,"n")
if len(ourtoks) > 2:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
def addtoindexprocess(linetoks,tnumin,myfile,linenum):
""" \addtoindex{strings} """
global indexdict
- ourtoks,inlen = pickup(linetoks,tnumin,"i { * }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { * }",myfile,linenum,"n")
if len(ourtoks) > 2:
# The * means a list of tokens.
fake = ""
def hyperlinkprocess(linetoks,tnumin,myfile,linenum):
""" \hyperlink{entryname} """
global linkhyperdict
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum,"n")
if len(ourtoks) > 2:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
def indexprocess(linetoks,tnumin,myfile,linenum):
""" \index{indexentryname} """
global indexdict
- ourtoks,inlen = pickup(linetoks,tnumin,"i { * }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { * }",myfile,linenum,"n")
if len(ourtoks) > 2:
fake = ""
# For now not bothering with index strings
""" \refersec{label} """
global labelrefdict
t = linetoks[tnumin]
- ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum)
+ ourtoks,inlen = pickup(linetoks,tnumin,"i { i }",myfile,linenum,"n")
if len(ourtoks) > 2:
t2 = ourtoks[2];
index = tokmention(t2,myfile,linenum)
global newcommsdict
# Link naming target
- global linksdict
+ global linkhyperdict
# TARG suffix
- global targsdict
+ global targhyperdict
# INDX suffix
- global indxdict
- # MARK suffix
- global markdict
+ global indexdict
# NAME suffix
global namedict
- # This is the 'normal, most common case in running text.
- global defaultdict
if len(linetoks) < 1:
return linetoks
# index the DWname
# Link is to chap:DWname
tm = tokmention(t,myfile,linenum)
- applytodict(defaultdict,rawname,tm)
linkname = makelinkname(rawname)
indxname = deloptionalprefix(commandname,"\\")
- applytodict(linkhyperdict,linkname,tm);
applytodict(indexdict,indxname,tm);
+
+ applytodict(linkhyperdict,linkname,tm);
tnumin = tnumin + 1
continue
if newcommsdict.has_key(rawname):
# Suffixes are LINK TARG INDX MARK NAME
commandname =rawnameiscommand(rawname,"LINK")
if len(commandname) > 0:
- # We do not expect to see this in text, LINK is really how
- # the defaultdict entries are created so should
- # be hidden from us here.
# index the DWname
# Link is to chap:DWname
if dwfnamecommsdict.has_key(commandname):
tm = tokmention(t,myfile,linenum)
targname = makelinkname(commandname)
indxname = deloptionalprefix(commandname,"\\")
- applytodict(targsdict,targname,tm)
+ applytodict(targhyperdict,targname,tm)
applytodict(indexdict,indxname,tm);
else:
printodderr(rawname,commandname,myfile,linenum)
if dwfnamecommsdict.has_key(commandname):
tm = tokmention(t,myfile,linenum)
indexname = deloptionalprefix(commandname,"\\")
- applytodict(indxdict,indexname,tm)
+ applytodict(indexdict,indexname,tm)
else:
printodderr(rawname,commandname,myfile,linenum)
tnumin = tnumin + 1
# index DWname
if dwfnamecommsdict.has_key(commandname):
tm = tokmention(t,myfile,linenum)
- applytodict(markdict,commandname,tm)
+ applytodict(targhyperdict,commandname,tm)
indexname = deloptionalprefix(commandname,"\\")
- applytodict(indxdict,indexname,tm)
- targname = makelinkname(commandname)
- applytodict(markdict,targname,tm)
+ applytodict(indexdict,indexname,tm)
else:
printodderr(rawname,commandname,myfile,linenum)
tnumin = tnumin + 1
l = v._line
print " [%2d] %s in file %s line %d" %(ct,name,f._name,l)
+# Are the lines close? If so
+def closetogether(l1,l2):
+ d = abs(l1 - l2)
+ if d < 4:
+ return "y"
+ return "n"
+
+def shouldprintalldups(v):
+ if len(v) != 2:
+ return "y"
+ if v[0]._file != v[1]._file:
+ return "y"
+ if closetogether(v[0]._line,v[1]._line) == "y":
+ # Are the lines near one another?
+ # If so a harmless duplication
+ return "n"
+ return "y"
+
def printtoomany(name,k,vlist):
print "Duplicate in %s: %s:" %(name,k)
ct = 0
for v in vlist:
- printtokmention(ct,v)
- ct = ct + 1
+ printtokmention(ct,v)
+ ct = ct + 1
def checkduplicatetargs(dname,d):
targs = d.items()
for vi in targs:
(k,v) = vi
if len(v) > 1:
- printtoomany(dname,k,v)
+ if shouldprintalldups(v) == "y":
+ printtoomany(dname,k,v)
def checkmissingtarg(name,targ,refs):
rlist = refs.items()
print "Unused target from",name,":",k
def print_stats():
- # All the newdwfnamecommands
global dwfnamecommsdict
- # All the newcommand instances.
global newcommsdict
-
- # LINK suffix
- global linksdict
-
- # TARG suffix hyperlink of 'chap:#1' Index #2
- global targsdict
- # INDX suffix index #2
- global indxdict
- # MARK suffix hypertarget chap:#1 and index #2
- global markdict
- # NAME suffix. No index, just text shows.
- global namedict
- # This is the 'normal, most common case in running text.
- # Default implies index #2 and hyperlink of chap:#1
- global defaultdict
-
- # Targets for \hypertarget and livetarg[i]
global targhyperdict
- # Links using hyperlink.
global linkhyperdict
- # label targets (labels)
+ global indexdict
global labeldict
- # \refersec \ref \vref to labels
global labelrefdict
- # The index content for named things.
- global indexdict
- checkduplicatetargs("targets",targsdict)
- checkduplicatetargs("hypertargets",targhyperdict)
- checkduplicatetargs("MARKs",markdict)
checkduplicatetargs("newdwfname commands",dwfnamecommsdict)
checkduplicatetargs("commands",newcommsdict)
+ checkduplicatetargs("hypertargets",targhyperdict)
checkduplicatetargs("labels",labeldict)
checkmissingtarg("hyperlinks",targhyperdict,linkhyperdict)
checkmissingtarg("labels",labeldict,labelrefdict)
checkmissingref("labels",labeldict,labelrefdict)
- checkmissingref("names",dwfnamecommsdict,namedict)
-
-debug = "n"
-
+# Perhaps these should be controlled by
+# the command line.
+debug = "n"
+winpath = "n"
+def buildfilepaths(files,basetarg):
+ outlist = []
+ prefix = ""
+ for f in files:
+ prefix = ""
+ if len(basetarg) > 0:
+ prefix = basetarg
+ elif winpath == "y":
+ prefix = "..\\latexdoc\\"
+ else:
+ prefix = "../latexdoc/"
+ outlist += [prefix + f]
+ return outlist
def read_all_args():
filelist1 = []
filelist2 = []
+ baselist1 = []
+ baselist2 = []
+ basetarg = ""
fileio.setkeepordeletecomments("d")
- filelist1 = {"./testrefer.tex"}
- filelist2 = {"./testrefer.tex"}
- if debug == "n":
- filelist1 = {"../latexdoc/dwarfnamecmds.tex",
- "../latexdoc/dwarf5.tex",
- "../latexdoc/generaldescription.tex"}
-
- filelist2 = {"../latexdoc/dwarf5.tex",
- "../latexdoc/attributesbytag.tex",
- "../latexdoc/changesummary.tex",
- "../latexdoc/compression.tex",
- "../latexdoc/copyright.tex",
- "../latexdoc/dataobject.tex",
- "../latexdoc/datarepresentation.tex",
- "../latexdoc/debugsectionrelationships.tex",
- "../latexdoc/encodingdecoding.tex",
- "../latexdoc/examples.tex",
- "../latexdoc/foreword.tex",
- "../latexdoc/generaldescription.tex",
- "../latexdoc/gnulicense.tex",
- "../latexdoc/introduction.tex",
- "../latexdoc/otherdebugginginformation.tex",
- "../latexdoc/programscope.tex",
- "../latexdoc/sectionversionnumbers.tex",
- "../latexdoc/splitobjects.tex",
- "../latexdoc/typeentries.tex"}
+ if debug == "y":
+ baselist1 = ["testrefer.tex"]
+ baselist2 = ["testrefer.tex"]
+ basetarg = "./"
+ else:
+ baselist1 = ["dwarfnamecmds.tex",
+ "dwarf5.tex",
+ "generaldescription.tex"]
+
+ baselist2 = ["dwarf5.tex",
+ "attributesbytag.tex",
+ "changesummary.tex",
+ "compression.tex",
+ "copyright.tex",
+ "dataobject.tex",
+ "datarepresentation.tex",
+ "debugsectionrelationships.tex",
+ "encodingdecoding.tex",
+ "examples.tex",
+ "foreword.tex",
+ "generaldescription.tex",
+ "gnulicense.tex",
+ "introduction.tex",
+ "otherdebugginginformation.tex",
+ "programscope.tex",
+ "sectionversionnumbers.tex",
+ "splitobjects.tex",
+ "typeentries.tex"]
+ filelist1 = buildfilepaths(baselist1,basetarg)
+ filelist2 = buildfilepaths(baselist2,basetarg)
if (len(filelist1) < 1) or (len(filelist2) < 1):
print >> sys.stderr , "No files specified to refer.py, internal error."
DWARF Git Projects / dwarf-doc.git / commitdiff