def __init__(self, fn=None, text=None, encoding='UTF-8', **args):
File.__init__(self, fn=fn, encoding=encoding, **args)
if text is not None:
self.text = text
elif fn is not None and os.path.exists(fn):
self.text = String(self.read().decode(encoding))
else:
self.text = String("")
def __init__(self, fn=None, text=None, encoding='UTF-8', **args):
File.__init__(self, fn=fn, encoding=encoding, **args)
if text is not None:
self.text = text
elif fn is not None and os.path.exists(fn):
self.text = String(self.read().decode(encoding))
else:
self.text = String("")
def __init__(self, fn=None, styles=None, text=None, encoding='UTF-8', **args):
File.__init__(self, fn=fn, encoding=encoding, **args)
if styles is not None:
self.styles = styles
elif text is not None:
self.styles = Styles.from_css(text)
elif fn is not None and os.path.exists(fn):
self.styles = Styles.from_css(open(fn, 'rb').read().decode(encoding))
else:
self.styles = Styles()
def __init__(
self,
fn=None,
root=None,
tree=None,
parser=None,
encoding='UTF-8',
schemas=None,
NS=None,
**args
):
# parent init of course
File.__init__(self, fn=fn, root=root, parser=parser, schemas=schemas, **args)
# assign the current class's values to self, because XML's values might have been overridden in the child class.
self.ROOT_TAG = self.__class__.ROOT_TAG
self.DEFAULT_NS = self.__class__.DEFAULT_NS
if NS is not None:
self.NS = NS
else:
self.NS = self.__class__.NS
# set up the root element
if root is None and tree is not None:
self.root = self.tree.getroot()
elif isinstance(root, str) or isinstance(root, bytes):
self.root = XML.fromstring(root, parser=parser)
elif isinstance(root, dict):
self.root = self.from_dict(root)
elif root is not None:
self.root = root
elif type(fn) in [str, bytes] and os.path.isfile(fn): # read from fn
tree = etree.parse(fn, parser=parser)
self.root = tree.getroot()
self.info = self.get_info(tree=tree)
elif self.ROOT_TAG is not None:
from .builder import Builder
B = Builder(default=self.DEFAULT_NS, **self.NS)
tag = self.tag_name(self.ROOT_TAG)
tagns = self.tag_namespace(self.ROOT_TAG) # None if no namespace
if tagns is not None:
nstag = list(self.NS.keys())[list(self.NS.values()).index(tagns)]
else:
nstag = '_' # this is the "default" nstag for the Builder
self.root = B[nstag](tag)
else:
self.root = etree.Element(
String(self.__class__.__name__).identifier(camelsplit=True).lower(), nsmap=self.NS
)
# set up document info (based on tree.docinfo)
if self.info is None:
self.info = self.get_info(tree=tree)
def __init__(self,
fn=None,
styles=None,
text=None,
encoding='UTF-8',
**args):
File.__init__(self, fn=fn, encoding=encoding, **args)
if styles is not None:
self.styles = styles
elif text is not None:
self.styles = Styles.from_css(text)
elif fn is not None and os.path.exists(fn):
self.styles = Styles.from_css(
open(fn, 'rb').read().decode(encoding))
else:
self.styles = Styles()
def convert(self, outfn=None, **params):
if outfn is not None and File(fn=outfn).fn != self.fn:
self.write(fn=outfn)
else:
outfn = self.fn
res = Image(fn=outfn).mogrify(**params)
return outfn + ': ' + res
def write(
self,
fn=None,
root=None,
encoding='UTF-8',
doctype=None,
canonicalized=False,
xml_declaration=True,
pretty_print=True,
with_comments=True,
):
data = self.tobytes(
root=root or self.root,
xml_declaration=xml_declaration,
pretty_print=pretty_print,
encoding=encoding or self.info.encoding,
doctype=doctype or self.info.doctype,
canonicalized=canonicalized,
with_comments=with_comments,
)
if canonicalized == True and xml_declaration == True: # add the xml declaration
data = ("<?xml version='1.0' encoding='%s'?>\n" % encoding).encode(encoding) + data
File.write(self, fn=fn or self.fn, data=data)
def insert(self,
text=None,
data=None,
prefix="",
suffix="",
ext='.txt',
duplicate=False):
"""insert the given queue entry into the queue.
text = the text of the queue entry (REQUIRED)
prefix = a prefix for each queue entry filename (default '')
suffix = a suffix for each queue entry filename (before extension, default '')
ext = the extension of the queue entry filename (default '.json')
"""
# only insert the entry if there is no other entry containing the same data
# -- no need to duplicate an existing entry.
if duplicate == False:
qfns = self.list()
for qfn in qfns:
try:
with open(qfn, 'rb') as f:
qft = f.read().decode('utf-8')
if qft == text:
return qfn
except:
# queue managed to get to the file before we read it -- no need to keep reading,
# since they are processed in the same order by self.list()
break
qfn = os.path.join(
self.path,
"%s%s-%.5f%s%s" % (prefix, time.strftime("%Y%m%d.%H%M%S"),
random.random(), suffix, ext))
if text is not None:
qf = Text(fn=qfn, text=text)
elif data is not None:
qf = File(fn=qfn, data=data)
qf.write()
return qfn
def write(self, fn=None, text=None, encoding='UTF-8', **args):
try:
data = (text or self.text or '').encode(encoding)
except:
data = (text or self.text or '').encode()
File.write(self, fn=fn, data=data, **args)
def write(self, fn=None, text=None, encoding='UTF-8', **args):
try:
data = (text or self.text or '').encode(encoding)
except:
data = (text or self.text or '').encode()
File.write(self, fn=fn, data=data, **args)
def write(self, fn=None, encoding='UTF-8', **args):
text = self.render_styles()
File.write(self, fn=fn, data=text.encode(encoding))
def gswrite(self,
fn=None,
device='jpeg',
res=600,
alpha=4,
quality=90,
gs=None):
"use ghostscript to create output file(s) from the PDF"
gs = (gs or self.gs or os.environ.get('gs') or 'gs')
# count the number of pages
if fn is None:
fn = os.path.splitext(self.fn)[0] + DEVICE_EXTENSIONS[device]
fn = File(fn=fn).fn # normalize path
if not os.path.exists(os.path.dirname(fn)):
os.makedirs(os.path.dirname(fn))
log.debug("PDF.gswrite():\n\tself.fn = %s\n\tout fn = %s" %
(self.fn, fn))
if os.path.splitext(self.fn)[-1].lower() == '.pdf':
cmd = [
gs, '-q', '-dNODISPLAY', '-c',
"(%s) (r) file runpdfbegin pdfpagecount = quit" % self.fn
]
log.debug(cmd)
out = subprocess.check_output(cmd).decode('utf-8').strip()
if out == '': out = '1'
pages = int(out)
log.debug("%d page(s)" % pages)
else:
pages = 1
if pages > 1:
# add a counter to the filename, which tells gs to create a file for every page in the input
fb, ext = os.path.splitext(fn)
n = len(re.split('.', str(pages))) - 1
counter = "-%%0%dd" % n
fn = fb + counter + ext
# remove any existing output filenames that match the pattern
for existingfn in glob(fb + '*' + ext):
log.debug("REMOVING %s" % existingfn)
os.remove(existingfn)
callargs = [
gs, '-q', '-dSAFER', '-dBATCH', '-dNOPAUSE', '-dUseCropBox',
'-sDEVICE=%s' % device,
'-r%d' % res
]
if device == 'jpeg':
callargs += ['-dJPEGQ=%d' % quality]
if 'png' in device or 'jpeg' in device or 'tiff' in device:
callargs += [
'-dTextAlphaBits=%d' % alpha,
'-dGraphicsAlphaBits=%d' % alpha
]
callargs += ['-sOutputFile=%s' % fn, self.fn]
try:
log.debug(callargs)
subprocess.check_output(callargs)
except subprocess.CalledProcessError as e:
log.error(callargs)
log.error(str(e.output, 'utf-8'))
fns = sorted(glob(re.sub(r'%\d+d', '*', fn)))
log.debug('\n\t' + '\n\t'.join(fns))
return fns
def write(self, fn=None, encoding='UTF-8', **args):
text = self.render_styles()
File.write(self, fn=fn, data=text.encode(encoding))
