def get_includes(self):
"""Return a list of all files that are included (with \include
or \input) in the texfile """
# TODO: move includepattern and inputtpattern to self._patterns
includepattern = re.compile(r'\\include\{(?P<filename>.*?)\}')
inputpattern = re.compile(r'\\input(TikZ)?\{(?P<filename>.*?)\}')
result = []
try:
texfile = open(self._basename + ".tex")
for line in texfile:
includematch = includepattern.search(line)
inputmatch = inputpattern.search(line)
if includematch:
filename = includematch.group('filename') + ".tex"
if os.path.isfile(filename):
result.append(filename)
if inputmatch:
filename = inputmatch.group('filename')
if os.path.isfile(filename):
result.append(filename)
texfile.close()
except IOError, data:
Out.write("Couldn't get included files from %s:\n" \
% self._basename + ".tex", VERB_WARN)
Out.write(data + "\n", VERB_WARN)
def run_bibtex(self):
""" Run bibtex on the texfile """
Out.write("Running bibtex %s\n" % self._basename)
bibtex_command = \
self.options['bibtexbin'].replace('%', self._basename)
try:
bibtexprocess = subprocess.Popen( \
bibtex_command + " 2>&1", \
shell=True, \
cwd=os.getcwd(), \
env=os.environ, \
stdout=subprocess.PIPE, \
stdin=open(os.devnull)
)
parser = CompilerOutputPrinter(bibtexprocess.stdout)
fatal, error, warning = parser.parseStream()
# TODO: print out fatal, error, warning (for all the parsers, not
# just this one)
while True:
time.sleep(1)
exitcode = bibtexprocess.poll()
if exitcode is not None:
break
Out.write("Waiting for bibtex to finish.\n", VERB_DEBUG)
if exitcode != 0:
Out.write("bibtex returned with error (exit code %s).\n" \
% exitcode, VERB_WARN)
return False #Failure
return True
except OSError, data:
Out.write("bibtex failed to run:\n", VERB_WARN)
Out.write(data + "\n", VERB_WARN)
def launch_viewer(self):
""" Launch the pdf viewer for the preview pdf
"""
if self.options['viewer'] != '' \
and self.options['viewer'] is not None:
Out.write("Launching viewer '" + self.options['viewer']+ "' for " \
+ self._basename + ".preview.pdf\n")
os.system(self.options['viewer'] + " " + self._basename \
+ ".preview.pdf & ")
def run_makeindex(self):
""" Run makeindex on the texfile """
Out.write("Running makeindex %s\n" % self._basename)
makeindex_command = \
self.options['makeindexbin'].replace('%', self._basename)
try:
makeindexprocess = subprocess.Popen( \
makeindex_command + " 2>&1" , \
shell=True, \
cwd=os.getcwd(), \
env=os.environ, \
stdout=subprocess.PIPE, \
stdin=open(os.devnull)
)
parser = CompilerOutputPrinter(makeindexprocess.stdout)
fatal, error, warning = parser.parseStream()
while True:
time.sleep(1)
exitcode = makeindexprocess.poll()
if exitcode is not None:
break
print "Waiting for makeindex to finish"
if exitcode != 0:
Out.write("'%s' returned with error (exit code %s).\n" \
% (makeindex_command, exitcode), VERB_WARN)
return False #Failure
return True
except OSError, data:
Out.write("makeindex failed to run:\n", VERB_WARN)
Out.write(data + "\n", VERB_WARN)
def firstcompile(self):
""" Make the first complete compilation of the texfile """
Out.write("Start Initial Compilation\n")
if os.path.isfile(self._basename + ".pdf"):
Out.write("There was an old pdf file %s. It will be deleted.\n" \
% (self._basename + ".pdf"))
os.remove(self._basename + ".pdf")
if self.options['smart']:
for watchfile in self.watchfilelist():
elements = self._get_elements_from_file(watchfile)
self._references[watchfile] = elements['references']
self._labels[watchfile] = elements['labels']
self._citations[watchfile] = elements['citations']
self._indexitems[watchfile] = elements['index']
return self.fullcompile()
def create_previewfile(self):
""" Copy the file.pdf resulting from a compilation to
file.preview.pdf
"""
compiledpdf = self._basename + ".pdf"
previewpdf = self._basename + ".preview.pdf"
if not os.path.isfile(compiledpdf):
Out.write("pdf file %s does not exist.\n" % compiledpdf, VERB_ERR)
if self.options['texcompiler'] != 'pdflatex':
self.cleanup()
Out.write("Did you forget to select --dvi?\n", VERB_ERR)
sys.exit(2)
try:
Out.write("Copying %s to %s\n" % (compiledpdf, previewpdf))
shutil.copy(compiledpdf, previewpdf)
except IOError, data:
Out.write(data + "\n", VERB_WARN)
Out.write("Could not copy %s to %s.\n" \
% (compiledpdf, previewpdf), VERB_WARN)
return False
def extract_element(fullstring, position):
""" Extract the contents of the first {...} block found after
position in fullstring """
Out.write("Extracting element from fullstring at position %s\n" \
% position, VERB_DEBUG)
try:
startposition = fullstring.index("{", position) + 1
endposition = startposition
open_brackets = 1
while open_brackets > 0:
endposition += 1
if fullstring[endposition] == '{':
open_brackets += 1
elif fullstring[endposition] == '}':
open_brackets -= 1
return fullstring[startposition:endposition]
except (ValueError, IndexError):
Out.write("Internal Error extracting element: " \
+ "string has invalid brackets.\n", VERB_WARN)
return None
def _get_elements_from_file(self, filename):
""" Return a dict with the following four elements:
- a list of 'labels' defined in the file
- a list of 'references' defined in the file
- a list of 'citations' defined in the file
- a list of 'index' items defined in the file
"""
result = {'labels' :[],
'references':[],
'citations' :[],
'index' :[]}
for extension in BINARYEXTENSIONS:
if (filename.lower()).endswith(extension):
return result
try:
afile = open(filename)
filecontents = afile.read()
afile.close()
except IOError, data:
Out.write("Couldn't read %s for analysis:\n" % filename, VERB_WARN)
Out.write(data + "\n", VERB_WARN)
return None
def parseStream(self):
"""docstring for parseStream"""
line = self.input_stream.readline()
VERB_CURR = VERB_STATUS
keep_verb = False
while line and not self.done:
line = line
if (line.rstrip() == ''):
keep_verb = False
VERB_CURR = VERB_STATUS
# process matching patterns until we find one
for (pat, pat_keep_verb) in self.warn_patterns:
myMatch = pat.search(line)
if myMatch:
self.numWarns += 1
keep_verb = pat_keep_verb
VERB_CURR = VERB_WARN
break
for (pat, pat_keep_verb) in self.err_patterns:
myMatch = pat.search(line)
if myMatch:
self.numErrs += 1
keep_verb = pat_keep_verb
VERB_CURR = VERB_ERR
break
Out.write("DEBUG: writing %s line\n" % VERB_CURR, VERB_DEBUG, stream='sub')
Out.write(line, VERB_CURR, stream='sub')
if not keep_verb:
VERB_CURR = VERB_STATUS
line = self.input_stream.readline()
return self.isFatal, self.numErrs, self.numWarns
def add_watchfile(self, watchfile_wc):
""" Add a watchfile, wildcard expression, or shell backquote to the
list of watchfiles.
For example watchfile_wc could be any of the following:
images/fig1.pdf
images/*.tikz
`find . | grep tikz`
"""
class WatchFileExistsException(Exception):
"""Raised internally if a watchfile is already on the watchlist"""
pass
watchfilelist = []
if watchfile_wc.startswith('`'):
watchfile_wc = watchfile_wc[1:-1]
pipe = subprocess.Popen(watchfile_wc, shell=True, bufsize=0,
stdout=subprocess.PIPE).stdout
for file in pipe.readlines():
file = file.rstrip('\r\n')
watchfilelist.append(file)
else:
watchfilelist = glob(watchfile_wc)
for watchfile in watchfilelist:
try:
if os.path.isfile(watchfile):
for existing_file in self._watchfiletimes.keys():
if os.path.samefile(existing_file, watchfile):
raise WatchFileExistsException(watchfile)
self._watchfiletimes[watchfile] = \
os.path.getmtime(watchfile)
else:
Out.write("The file %s that you want " % watchfile \
+ "to watch does not exist.\n", VERB_ERR)
sys.exit(2)
except WatchFileExistsException, data:
Out.write("The file %s is already being watched" \
% data, VERB_DEBUG)
def __init__(self, filename):
""" Create a Texfile object wrapping the filename"""
self.filename = filename
self.changed = {'citations':False, 'labels':False,
'references':False, 'index':False}
self.options = {}
self.options['smart'] = True
self.options['makeindex'] = True
self.options['texcompiler'] = 'pdflatex'
self.options['compileroptions'] = ''
self.options['bibtex'] = True
self.options['bibtexbin'] = 'bibtex %'
self.options['makeindexbin'] = 'makeindex %'
self.options['extracompiler'] = ''
self.options['color'] = False
self.options['dvi'] = False
self.options['dvipdf'] = 'dvipdf %.dvi'
self.options['no_cleanup'] = False
self.options['viewer'] = 'kpdf'
self.options['cleanup'] = []
self._basename = self.filename # filename without ending
if self._basename.endswith('.tex'):
self._basename = self._basename.replace('.tex', '')
if not os.path.isfile(self._basename + ".tex"):
Out.write("The file %s that " % (self.options['filename'])\
+ "you want to compile doesn't exist.\n", VERB_ERR)
self._references = {} # These four dicts map filenames to
self._labels = {} # lists of references, labels,
self._citations = {} # citations, and index items that
self._indexitems = {} # occur in these files.
self._watchfiletimes = {} # dict of filenames to change times
self._patterns = {
'citations' : re.compile(r'\\cite[a-z*]{,3}\{'),
'labels' : re.compile(r'\\label\{'),
'references' : re.compile(r'\\ref\{'),
'index' : re.compile(r'\\index\{')
}
self.add_watchfile(self._basename + '.tex')
def convert_dvi(self):
""" Convert file.dvi to file.pdf """
# TODO: check if dvi file actually exists
dvipdf_command = \
self.options['dvipdf'].replace('%', self._basename)
Out.write("Running '%s' to convert %s to %s\n" \
% (dvipdf_command, \
self._basename + ".dvi", \
self._basename + ".pdf"))
try:
if subprocess.call( \
dvipdf_command + " 2>&1", \
cwd=os.getcwd(), \
env=os.environ, \
shell=True, \
) != 0:
Out.write("Failed to convert " + self._basename \
+ ".dvi to pdf.\n", VERB_WARN)
self.cleanup()
Out.write("Is '%s' available?\n" % dvipdf_command, VERB_ERR)
except OSError, data:
Out.write("'%s' failed:\n" % dvipdf_command, VERB_WARN)
Out.write(data + "\n", VERB_WARN)
return False
def run_latex(self):
""" This runs pdflatex (or whatever is given as
texcompiler). If dvi is set, it is assumed that the compiler
produced a dvi file, which is then converted to pdf via
'dvipdf'.
"""
Out.write("Running %s %s on %s\n" % (self.options['texcompiler'],
self.options['compileroptions'],
self._basename + ".tex"))
try:
latexprocess = subprocess.Popen( \
self.options['texcompiler'] + " " \
+ self.options['compileroptions'] + " " \
+ self._basename + " 2>&1", \
shell=True, \
cwd=os.getcwd(), \
env=os.environ, \
stdout=subprocess.PIPE, \
stdin=open(os.devnull)
)
parser = CompilerOutputPrinter(latexprocess.stdout)
fatal, error, warning = parser.parseStream()
while True:
time.sleep(1)
exitcode = latexprocess.poll()
if exitcode is not None:
break
print "Waiting for %s to finish" \
% self.options['texcompiler']
if exitcode != 0:
Out.write(self._basename + \
".tex failed to compile (exit code %s).\n" \
% exitcode, VERB_WARN)
return False #Failure
except OSError, data:
Out.write(self._basename + ".tex failed to compile:\n", VERB_WARN)
Out.write(data + "\n", VERB_WARN)
return False # Failure
def run_extracompiler(self):
""" Run the compiler set in the extracompiler attribute """
if self.options['extracompiler'] is not None:
self.options['extracompiler'] = \
self.options['extracompiler'].strip()
else:
self.options['extracompiler'] = ''
extracompiler = self.options['extracompiler']
if extracompiler != '':
extracompiler = extracompiler.replace("%", self._basename)
try:
Out.write("Running extracompiler '%s'\n" % extracompiler)
extracompilerprocess = subprocess.Popen( \
extracompiler + " 2>&1" , \
shell=True, \
cwd=os.getcwd(), \
env=os.environ, \
stdout=subprocess.PIPE, \
stdin=open(os.devnull)
)
parser = \
CompilerOutputPrinter(extracompilerprocess.stdout)
fatal, error, warning = parser.parseStream()
while True:
time.sleep(1)
exitcode = extracompilerprocess.poll()
if exitcode is not None:
break
print "Waiting for '%s' to finish" % extracompiler
if exitcode != 0:
Out.write("'%s' returned with error (exit code %s).\n" \
% (extracompiler, exitcode), VERB_WARN)
return False #Failure
return True
except OSError, data:
Out.write("'%s' failed to run:\n" % extracompiler, VERB_WARN)
Out.write(data + "\n", VERB_WARN)
def fullcompile(self):
""" Make a complete unconditional compilation of the texfile
This is guaranteed to produce a working pdf with all
references, bibliographies, etc. complete. The file is
processed several times, the steps are:
- compile (just pdflatex, or whatever is set as texcompiler)
- bibtex (if bibtex attribute is set)
- recompile (like compile)
- makeindex (if the makeindex attribute is set)
- extracompiler (if set)
- recompile
"""
Out.write("Start Full Compilation.\n")
if not self.run_latex():
return False # Failure
if self.options['bibtex']:
if not self.run_bibtex():
Out.write("bibtex failed.\n", VERB_WARN)
if not self.run_latex():
return False # Failure
if self.options['makeindex']:
if not self.run_makeindex():
Out.write("makeindex failed.\n", VERB_WARN)
self.run_latex()
if not self.run_extracompiler():
Out.write("'%s' failed.\n" % self.options['extracompiler'], \
VERB_WARN)
if not self.run_latex():
return False # Failure
if self.options['dvi']:
if not self.convert_dvi():
return False #Failure
if not self.create_previewfile():
return False # Failure
return True # Success
def cleanup(self):
""" Delete the temporary files that are generated during the
compilation of texfilebase.tex.
"""
if not self.options['no_cleanup']:
Out.write("Deleting temporary files for %s" \
% self._basename + '.tex\n')
files_to_delete = []
for element in self.options['cleanup']:
element = element.replace('%', self._basename)
files_to_delete += glob(element)
for filename in files_to_delete:
if os.path.isfile(filename):
try:
Out.write(" Deleting %s" % filename, VERB_DEBUG)
os.remove(filename)
except OSError, data:
Out.write(data + "\n", VERB_WARN)
def smartcompile(self):
""" Run whatever compilers are necessary to create a complete
pdf with all references etc. resolved.
At a minimum, the texcompiler is run once. If an
extracompiler is set, texompiler -> extracompiler
-> texcompiler is run at minimum.
Bibtex and Makeindex are skipped if they are set to
False in the options.
"""
Out.write("Start Smart Compilation.\n")
if not self.run_latex():
return False # Failure
if self.changed['citations']:
if self.options['bibtex']:
if not self.run_bibtex():
Out.write("bibtex failed.\n", VERB_WARN)
else:
Out.write("There were changes in the citations, but bibtex is "\
+ "disabled. You should enable bibtex.\n", VERB_WARN)
if not self.run_latex():
return False # Failure
if self.changed['index']:
if self.options['makeindex']:
if not self.run_makeindex():
Out.write("makeindex failed\n", VERB_WARN)
self.run_latex()
else:
Out.write("There were changes in the index, but makeindex is "\
+ "disabled. You should enable makeindex.\n", VERB_WARN)
if not self.run_extracompiler():
Out.write("'%s' failed.\n" % self.options['extracompiler'], \
VERB_WARN)
if self.changed['citations'] or self.changed['labels'] \
or self.changed['references'] or self.changed['index'] \
or self.options['extracompiler'] != '':
if not self.run_latex():
return False # Failure
if self.options['dvi']:
self.convert_dvi()
self.create_previewfile()
return True # Success
def has_changed(self):
""" Check if the texfile or any of the watchfiles have
changed
"""
changed = False
for watchfile in self._watchfiletimes.keys():
trials = 0
# check if file has been renewed: some editors delete the file
# temporarily while it is being saved, so we make up to 10 trials
# to get the current modification time.
while True:
trials += 1
try:
newer = ( self._watchfiletimes[watchfile] < \
os.path.getmtime(watchfile) )
break
except OSError, data:
Out.write(str(data) + "\n", VERB_WARN)
time.sleep(1)
if trials > 10:
raise
if newer:
changed = True
Out.write("%s has changed.\n" % watchfile)
if self.options['smart']:
elements = self._get_elements_from_file(watchfile)
# references
if self._references[watchfile] != elements['references']:
Out.write("Changed references in %s\n" % watchfile)
self.changed['references'] = True
self._references[watchfile] = elements['references']
else:
self.changed['references'] = False
# labels
if self._labels[watchfile] != elements['labels']:
Out.write("Changed labels in %s\n" % watchfile)
self.changed['labels'] = True
self._labels[watchfile] = elements['labels']
else:
self.changed['labels'] = False
# citations
if self._citations[watchfile] != elements['citations']:
Out.write("Changed citations in %s\n" % watchfile)
self.changed['citations'] = True
self._citations[watchfile] = elements['citations']
else:
self.changed['citations'] = False
for extension in BIBEXTENSIONS:
if (watchfile.lower()).endswith(extension):
Out.write("Bibliography file %s has changed\n" \
% watchfile)
self.changed['citations'] = False
# index
if self._indexitems[watchfile] != elements['index']:
Out.write("Changed index in %s\n" % watchfile)
self.changed['index'] = True
self._indexitems[watchfile] = elements['index']
else:
self.changed['index'] = False
self._watchfiletimes[watchfile] = os.path.getmtime(watchfile)