texall script: bugfixes and improved performance

1 files changed, 82 insertions, 49 deletions

diff --git a/texall b/texall
index a6f98d4..03d6d9c 100755
--- a/texall
+++ b/texall
@@ -44,10 +44,10 @@ class TexallOptionParser(OptionParser):
                         dest="interactionmode", choices=("batchmode", "nonstopmode", "scrollmode", "errorstopmode"),
                         help="control the behaviour of latex (on of batchmode (default), nonstopmode (default for -o), scrollmode, errorstopmode)")
         self.add_option("-I", "--ignore",
-                        action="store", dest="ignorepattern", default=".*vorlage.*|\.\#.*|\..*\.rmligs\.tex",
+                        action="store", dest="ignorepattern", default=".*vorlage|\.\#|.*-rmligs\.tex$",
                         help="regular expression of filenames to ignore")
         self.add_option("-P", "--required-pattern",
-                        action="store", dest="requiredpattern", default='^\\\\documentclass|%%% TeX-master: t',
+                        action="store", dest="requiredpattern", default='^\\\\documentclass|^%%% TeX-master: t',
                         help="regular expression that must match the file content")
         self.add_option("-T", "--preserve-tempfiles",
                         action="store_false", dest="deletetempfiles", default=True,
@@ -110,10 +110,10 @@ class AnalyseTexRequired(Analyser):
         d.required = False
         Analyser.__init__(self,d)
     def parse(self, text):
-        if re_required.match(text):
+        if re_required.search(text):
             self.d.required = True
     def merge(self, d):
-        pass # do not take over master file status
+        self.d.required |= d.required
 
 class AnalyseTexDependencies(Analyser):
     def __init__(self, d):
@@ -123,7 +123,7 @@ class AnalyseTexDependencies(Analyser):
     def parse(self, text):
         for m in re_inputinclude.finditer(text):
             self.d.deps.append((m.group(2), (".tex", "")))
-        m = re_documentclass.match(text)
+        m = re_documentclass.search(text)
         if m:
             cls = m.group(2)
             self.d.deps.append((cls, (".cls", "")))
@@ -141,7 +141,7 @@ class AnalyseTexDependencies(Analyser):
     def merge(self, d):
         self.d.deps.extend(d.deps)
         self.d.grfdeps.extend(d.grfdeps)
-    def finish(self, _):
+    def finish(self, dirname):
         if hasattr(self.d, "needsPS"):
             if self.d.needsPS:
                 for f in self.d.grfdeps:
@@ -149,6 +149,17 @@ class AnalyseTexDependencies(Analyser):
             else:
                 for f in self.d.grfdeps:
                     self.d.deps.append((f, (".pdf", ".jpg", ".png", "")))
+        expandeddeps = []
+        for dep, extlist in self.d.deps:
+            if extlist == None:
+                expandeddeps.append((dep, extlist))
+            else:
+                try:
+                    expandeddeps.append((texpath(dirname, dep, extlist=extlist), None))
+                except ValueError, e:
+                    if opts.verbosity > 2:
+                        error(self.d.name, "ignoring dependency: %s"%e, warning=True)
+        self.d.deps = expandeddeps
 
 class AnalyseTexBibfiles(Analyser):
     def __init__(self, d):
@@ -168,7 +179,7 @@ class AnalyseTexBibfiles(Analyser):
             try:
                 bibpath = texpath(dirname, bib, pathtype="bib",progname="bibtex")
                 bibpath = strippath(bibpath, dirname)
-                self.d.deps.append((bibpath, ("",)))
+                self.d.deps.append((bibpath, None))
                 bibpaths.append(bibpath)
             except ValueError, e:
                 error(self.d.name, str(e), warning=True)
@@ -181,7 +192,7 @@ class AnalyseTexNeedsPs(Analyser):
         d.needsPS = False
         Analyser.__init__(self, d)
     def parse(self, text):
-        if re_needsps.match(text):
+        if re_needsps.search(text):
             self.d.needsPS = True
     def merge(self, d):
         self.d.needsPS |= d.needsPS
@@ -191,7 +202,7 @@ class AnalyseTexRmligs(Analyser):
         d.rmligs = False
         Analyser.__init__(self, d)
     def parse(self, text):
-        if re_rmligs.match(text):
+        if re_rmligs.search(text):
             self.d.rmligs = True
     def merge(self, d):
         self.d.rmligs |= d.rmligs
@@ -203,25 +214,32 @@ class AnalyseLogErrors(Analyser):
         d.errors = False
         Analyser.__init__(self, d)
     def parse(self, text):
-        if re_error.match(text):
+        if re_error.search(text):
             self.d.errors = True
 
 class AnalyseLogRequests(Analyser):
     def __init__(self, d):
         d.requests = {}
+        d.requestreason = {}
         Analyser.__init__(self, d)
         self.inrequest = False
-    def addrequest(self, name, priority):
+    def addrequest(self, name, priority, reason=None):
         if self.d.requests.has_key(name):
-            self.d.requests[name] = max(self.d.requests[name], priority)
+            if self.d.requests[name] < priority:
+                self.d.requests[name] = priority
+                if reason:
+                    self.d.reqeustreason[name] = reason
+            elif not self.d.requestreason[name]:
+                self.d.requestreason[name] = reason
         else:
             self.d.requests[name] = priority
+            self.d.requestreason[name] = reason
     def parse(self, text):
         for m in re_logmatcher.finditer(text):
             if m.group() == 'LaTeX Warning: There were undefined references.':
-                self.addrequest("latex",0)
+                self.addrequest("latex", 0, " (because of undefined references)")
             elif m.group() == 'LaTeX Warning: Label(s) may have changed. Rerun to get cross-references right.':
-                self.addrequest("latex",0)
+                self.addrequest("latex", 0, " (because of changed labels)")
             elif m.group(1)!=None:
                 self.addrequest(m.group(2), int(m.group(1)))
             elif m.group(3)!=None:
@@ -246,6 +264,9 @@ class AnalyseLogRequests(Analyser):
                 if infile != None:
                     options.append(infile)
                 self.addrequest(tuple(options), pri)
+    def finish(self, _):
+        if self.d.errors:
+            self.addrequest("latex", 0, " (because of errors in .log file)")
 
 loganalysers = (AnalyseLogErrors, AnalyseLogRequests)
 
@@ -282,18 +303,21 @@ def analyseFile(dirname, filename, analyserclasses, parsetype, fulltext=False, c
         f.close()
     if hasattr(data, "deps") and (recurse
                                   or (recurse==None
-                                      and hasattr(data, "requrired")
+                                      and hasattr(data, "required")
                                       and data.required)):
         # recurse:
         for dep, extlist in data.deps:
-            deppath = texpath(dirname, dep, extlist=extlist)
+            try:
+                deppath = texpath(dirname, dep, extlist=extlist)
+            except ValueError, e:
+                continue
             if (not recursesystemfiles) and re_systemfile.match(deppath): # skip system files
                 continue
             depname = strippath(deppath, dirname)
-            depdata = analyseFile(dirname, depname, fulltext, cache, recurse=True)
+            depdata = analyseFile(dirname, depname, analyserclasses, parsetype, fulltext, cache, True, recursesystemfiles)
             for a in analysers:
                 a.merge(depdata)
-    if hasattr(data, "required") and data.required:
+    if (hasattr(data, "required") and data.required) or parsetype == "log":
         for a in analysers:
             a.finish(dirname)
     if cache:
@@ -309,6 +333,8 @@ def analyseLog(dirname, filename):
 kpsecache = {}
 kpseproc = {}
 def texpath(dirname,filename,pathtype="tex",progname="latex",extlist=("",)):
+    if extlist == None:
+        return filename
     filename = os.path.expanduser(filename)
     # first try files in current directory
     if pathtype=="bib" and extlist==("",):
@@ -360,7 +386,7 @@ def texpath(dirname,filename,pathtype="tex",progname="latex",extlist=("",)):
                                             stdin=slave,
                                             cwd="/")
                     kpseproc[(pathtype, progname)] = (proc, master)
-                if select.select([master], [], [], .2)[0]:
+                if select.select([master], [], [], .05)[0]:
                     pathname = os.read(master, 1024)
                 else:
                     pathname = None
@@ -458,6 +484,10 @@ def alltexfiles(arglist):
             for directory, subdirs, files in os.walk(a):
                 if not opts.recurse:
                     subdirs[:] = []
+                else:
+                    for skipdir in (".svn", ".bzr"):
+                        if skipdir in subdirs:
+                            subdirs.remove(skipdir)
                 for name in files:
                     if re_texfile.match(name):
                         if re_ignore.match(name):
@@ -507,14 +537,10 @@ def needsupdate(dirname, texname, data=None):
     outdate = os.path.getmtime(outpath)
     if outdate < os.path.getmtime(os.path.join(dirname, texname)):
         return " (because .tex is newer than .pdf)"
-    for filename, extlist in data.deps:
-        try:
-            deppath = texpath(dirname, filename, extlist=extlist)
-            if outdate < os.path.getmtime(os.path.join(dirname, deppath)):
-                return " (because %s is newer than .pdf)"%deppath
-        except ValueError, e:
-            if opts.verbosity > 2:
-                error(os.path.join(dirname, texname), "ignored dependency: " + str(e), warning=True)
+    for deppath, extlist in data.deps:
+        assert extlist == None
+        if outdate < os.path.getmtime(os.path.join(dirname, deppath)):
+            return " (because %s is newer than .pdf)"%deppath
 
     # check for errors and requests
     logname = jobname + ".log"
@@ -579,7 +605,7 @@ def processTex(dirname, texname, data=None, reason=""):
                         reason = " (because %s is newer than %s)"%(bib,bbl)
                     else:
                         reason = " (because of --force)"
-                rmtempfile(bbl, dirname, force=True, reason=reason)
+                rmtempfile(bblpath, force=True, reason=reason)
                 nextbbls.remove(bbl)
         bbls = nextbbls
 
@@ -589,8 +615,10 @@ def processTex(dirname, texname, data=None, reason=""):
         # check for undefined references and run requests
         numrun=0
         reqs=opts.act # ignore requests if programs are not run
+        logname = jobname + ".log"
         while reqs:
-            reqs = analyseLog(dirname, jobname+".log").requests
+            logdata = analyseLog(dirname, logname)
+            reqs = logdata.requests
             if not reqs: break
             pri = reqs.values()
             pri.sort(reverse=True)
@@ -603,7 +631,9 @@ def processTex(dirname, texname, data=None, reason=""):
                     if rp == p:
                         reason = ""
                         if opts.verbosity:
-                            reason = " (because of request in .log file, priority %d)"%rp
+                            reason = logdata.requestreason[req]
+                            if not reason:
+                                reason = " (because of request in .log file, priority %d)"%rp
                         if req == "latex":
                             runtex(tex, texname, realname, dirname, reason=reason)
                         elif isinstance(req, tuple):
@@ -621,16 +651,16 @@ def processTex(dirname, texname, data=None, reason=""):
             runtex(tex, texname, realname, dirname, reason=" (because .ind file might have changed)")
 
     if data.needsPS:
-        run(["dvips", jobname+".dvi"], dirname)
+        run(["dvips", jobname+".dvi"], dirname, stderr=subprocess.STDOUT)
         run(["ps2pdf", jobname+".ps"], dirname)
 
-    rmtempfile(tmplist, dirname)
+    rmtempfile(tmplist)
 
 def bblfiles(dirname, jobname):
     files = dircache.listdir(dirname)
     return fnmatch.filter(files, "%s*.bbl"%jobname)
 
-def rmtempfile(filenames, dirname, force=False, reason=""):
+def rmtempfile(filenames, force=False, reason=""):
     if (opts.deletetempfiles or force) and filenames:
         if not isinstance(filenames, list):
             filenames = [filenames]
@@ -638,11 +668,10 @@ def rmtempfile(filenames, dirname, force=False, reason=""):
             print "  removing %s%s"%(", ".join(filenames), reason)
         if opts.act:
             for f in filenames:
-                fullname = os.path.join(dirname, f)
-                if os.path.exists(fullname):
-                    os.remove(fullname)
+                if os.path.exists(f):
+                    os.remove(f)
                 else:
-                    error(fullname, "could not be removed (does not exist)", warning=True)
+                    error(f, "could not be removed (does not exist)", warning=True)
 
 def run(arglist, dirname, reason="", inf=None, outf=None, stderr=None):
     """Run a command with optional io redirections.
@@ -761,7 +790,7 @@ def preparermligs(texname, dirname, texpath, tmplist):
     else:
         try:
             (subdir, texbasename) = os.path.split(texname)
-            rmligsbasename = ".%s.rmligs.tex"%texbasename[:-4]
+            rmligsbasename = "%s-rmligs.tex"%texbasename[:-4]
             rmligsname = os.path.join(subdir, rmligsbasename)
             rmligspath = os.path.join(dirname, rmligsname)
             if opts.act:
@@ -773,17 +802,18 @@ def preparermligs(texname, dirname, texpath, tmplist):
             return texname
         else:
             tmplist.append(rmligspath)
-
             def rewriteinputrmligs(m):
-                name=m.group(2)
+                name = m.group(2)
                 if not os.path.exists(os.path.join(dirname,name)):
                     name+=".tex"
-                if os.path.exists(os.path.join(dirname, name)):
-                    name = preparermligs(name, dirname, tmplist)
+                path = os.path.join(dirname, name)
+                if os.path.exists(path):
+                    name = preparermligs(name, dirname, path, tmplist)
                 else:
                     error(texpath, "%s: File could not be located. Not using rmligs."%name, warning=True)
                 return "\\%s{%s}"%(m.group(1), name[:-4])
             def filteroutput(d):
+                d = re_commentinclude.sub("", d)
                 rmligsfile.write(re_inputinclude.sub(rewriteinputrmligs, d))
                 return None
             run([rmligs_name, "-f"], dirname, inf=texfile, outf=filteroutput)
@@ -835,14 +865,17 @@ try:
     re_texfile = re.compile(".*\.tex$", re.I)
     if iswindows:
         re_systemfile = re.compile("^?:\\\\Program", re.I)
+        re_nonlocal = re.compile("^?:\\\\")
     else:
         re_systemfile = re.compile("^/usr/")
+        re_nonlocal = re.compile("^/")
 
     re_texcomment = re.compile("(?<!\\\\)%.*?$", re.M)
 
     re_required = re.compile(opts.requiredpattern)
 
     re_inputinclude = re.compile('\\\\(input|include|@input)\\{([^}]*)\\}')
+    re_commentinclude = re.compile('(?<!\\\\)%[^%\n\r]*\\\\(input|include|@input)\\{([^}]*?)\\}[^%\n\r]*?$', re.M)
     re_documentclass = re.compile('\\\\documentclass(\\[.*?\\])?\\{([^}]*)\\}')
     re_usepackage = re.compile('\\\\usepackage(\\[.*?\\])?\\{([^}]*)\\}')
     re_usetheme = re.compile('\\\\use(|color|font|inner|outer)theme(\\[.*?\\])?\\{([^}]*)\\}')
@@ -859,15 +892,15 @@ try:
     re_longreq = re.compile("^(binary|option|infile)=(.*?)$", re.M)
 
     re_error = re.compile('^! ', re.M)
-    re_rmligsfile = re.compile("\\.([^/ ]+)\\.rmligs\\.tex")
+    re_rmligsfile = re.compile("([^/ ]+)-rmligs\\.tex")
 
     if __name__ == "__main__":
-#         import cProfile
-#         import pstats
-#         cProfile.run("main(programargs)", "mystats")
-#         p = pstats.Stats("mystats")
-#         p.sort_stats('time').print_stats()
-        
+        #import cProfile
+        #import pstats
+        #cProfile.run("main(programargs)", "mystats")
+        #p = pstats.Stats("mystats")
+        #p.sort_stats('time').print_stats()
+         
         main(programargs)
 
 except KeyboardInterrupt: