def test_defaults():
"""Test default conversions."""
data = [
('1m', [
C(1, 'meter', 100, 'centimeter', '[length]'),
C(1, 'meter', 0.001, 'kilometer', '[length]')
]),
('100g', [
C(100, 'gram', 0.1, 'kilogram', '[mass]'),
C(100, 'gram', 100000, 'milligram', '[mass]')
]),
]
wf = Workflow3()
if 'default_units' not in wf.settings:
wf.settings['default_units'] = {}
wf.settings['default_units']['[length]'] = ['centimeter', 'kilometer']
wf.settings['default_units']['[mass]'] = ['kilogram', 'milligram']
c = convert.Converter(Defaults(wf))
for t in data:
i = c.parse(t[0])
res = c.convert(i)
assert len(res) == len(t[1])
for j, r in enumerate(res):
log.debug(r)
verify_conversion(t[1][j], r)
def setup():
if CONFIG["converter"] == None:
CONFIG["converter"] = convert.Converter()
if CONFIG["account"] == None:
CONFIG["account"] = account.Account()
if CONFIG["environment"] == None:
CONFIG["environment"] = Environment(loader=FileSystemLoader(\
searchpath="./WebPages"))
def test_conversion():
"""Test conversions."""
data = [
('1km m', C(1, 'kilometer', 1000, 'meter', '[length]')),
]
c = convert.Converter(None)
for t in data:
i = c.parse(t[0])
res = c.convert(i)
verify_conversion(t[1], res[0])
def test_valid():
"""Test valid input."""
data = [
('1.3 km', T(1.3, '[length]', 'kilometer', None)),
('1.3 km miles', T(1.3, '[length]', 'kilometer', 'mile')),
('5 m/s kph', T(5.0, '[length] / [time]', 'meter/second', 'kph')),
('21.3 m^2 acres', T(21.3, '[length] ** 2', u'meter²', 'acre')),
]
c = convert.Converter(None)
for t in data:
i = c.parse(t[0])
verify_parsed(t[1], i)
def test_invalid():
"""Test invalid input."""
queries = [
'dave', # doesn't start with a number
'1.3', # no unit
'5 daves', # invalid units
'10 km m cm', # too many units
]
c = convert.Converter(None)
for query in queries:
with pytest.raises(ValueError):
c.parse(query)
def processTestSet():
testPieces = {} #{composer name: piece object}
currdir = os.path.dirname(__file__) #the absolute filepath to the directory containing main.py
folderpath = "test_set" #the subdirectory containing our training data
finalPath = os.path.join(currdir, folderpath) #concatenate the two paths for use
for composer in os.listdir(finalPath): #test_set is full of folders corresponding to composers; inside is all pieces of theirs to train with
testPieces[composer] = []
if (composer != ".DS_Store"):
nextPath = os.path.join(finalPath, composer) #the path to this composer folder
if (os.path.isdir(nextPath)): #if this is a folder
for testPiece in os.listdir(nextPath): #loop through all pieces by the corresponding composer
if testPiece.endswith(".mxl"): #if the file is .mxl format (it should be)
piecePath = os.path.join(nextPath, testPiece)
C = convert.Converter(piecePath)
C.convert() #convert the file
toAdd = piece.Piece(C.getBass(), C.getChords())
testPieces[composer].append(toAdd)
return testPieces
def convert(self, path):
path = str(path)
dbpath, idprove = path.split('|')
outdir = os.path.dirname(dbpath)
outdir = os.path.join(outdir, 'm4')
if not os.path.exists(outdir):
os.mkdir(outdir)
self.outdir = outdir
self.force = self.doForceData.isChecked()
fi = self.doForceImages.currentIndex()
if fi == 0:
self.keep_img = True
self.img = True
elif fi == 1:
self.img = False
self.keep_img = True
elif fi == 2:
self.img = True
self.keep_img = False
self.pid = 'Converting to misura format: ' + idprove
self.converter = convert.Converter(dbpath, self.outdir)
self.outpath = self.converter.get_outpath(idprove,
img=self.img,
keep_img=self.keep_img)
self.connect(registry.tasks, QtCore.SIGNAL('sig_done(QString)'),
self.done)
run = widgets.RunMethod(self.converter.convert, frm=self.format)
run.step = 100
run.pid = self.pid
QtCore.QThreadPool.globalInstance().start(run)
self.progress_timer = QtCore.QTimer(self)
self.progress_timer.setInterval(300)
self.connect(self.progress_timer, QtCore.SIGNAL('timeout()'),
self.update_progress)
self.progress_timer.start()
def train(n):
learners = {} #{composer name: NGLearner object}
currdir = os.path.dirname(__file__) #the absolute filepath to the directory containing main.py
folderpath = "mxl_files" #the subdirectory containing our training data
finalPath = os.path.join(currdir, folderpath) #concatenate the two paths for use
for composer in os.listdir(finalPath): #mxl_files is full of folders corresponding to composers; inside is all pieces of theirs to train with
if (composer != ".DS_Store"):
print("checking composer ", composer)
nextPath = os.path.join(finalPath, composer) #the path to this composer folder
composerFB = NGramModel.N_Gram_Model(n, True) #the learner for figured bass
composerChords = NGramModel.N_Gram_Model(n, False) #the learner for chords
if (os.path.isdir(nextPath)): #if this is a folder
print("in folder")
for piece in os.listdir(nextPath): #loop through all pieces by the corresponding composer
print("checking piece", piece)
if piece.endswith(".mxl"): #if the file is .mxl format (it should be)
piecePath = os.path.join(nextPath, piece)
C = convert.Converter(piecePath)
try:
C.convert() #convert the file
except music21.musicxml.xmlToM21.MusicXMLImportException:
print("Error: %s could not be parsed.\n" % (piece))
continue
composerFB.train(C.getBass()) #train on this piece's figured bass for this composer
composerChords.train(C.getChords()) #train on this piece's chords for this composer
print("creating learner object for composer", composer)
toAdd = NGLearner.learner(composerFB, composerChords) #create the learner object
print("adding learner")
learners[composer] = toAdd #add it to the learners dictionary
return learners
def testCheckAccessOk(self, mock_access, mock_isfile):
converter = convert.Converter([self.INPUT_PY])
AssertThat(converter._Check()).IsTrue()
AssertThat(mock_isfile).WasCalled().Once().With(self.INPUT_PY)
AssertThat(mock_access).WasCalled().With(self.INPUT_PY, os.R_OK)
AssertThat(mock_access).WasCalled().LastWith(self.INPUT_PY, os.W_OK)
def testCheckUnwritableInputFile(self, mock_access, mock_isfile):
converter = convert.Converter([self.INPUT_PY])
AssertThat(converter._Check()).IsFalse()
AssertThat(mock_isfile).WasCalled().Once().With(self.INPUT_PY)
AssertThat(mock_access).WasCalled().With(self.INPUT_PY, os.R_OK)
AssertThat(mock_access).WasCalled().LastWith(self.INPUT_PY, os.W_OK)
def testCheckNonexistentInputFile(self, mock_isfile):
converter = convert.Converter([self.INPUT_PY])
AssertThat(converter._Check()).IsFalse()
AssertThat(mock_isfile).WasCalled().Once().With(self.INPUT_PY)
def main():
print("Online Novel to PDF")
title = input("Title of PDF: ")
address = input("Address of first page: ")
index = int(input("Incrementing index: "))
#parse Address
parts = address.split(str(index))
prefix = parts[0]
suffix = parts[1]
titleKeyword = input("Keyword in Title \
(To make sure only scraping what is needed): ")
address = prefix + str(index) + suffix
scraper = scrape.Scraper(address)
tag = ''
attrs = ''
if input("Do you know which HTML container the content resides in? "
) == 'yes':
tag = input("Element type: ")
attrName = input("Element Atrribute Name: ")
attrValue = input("Element Attribute Value: ")
attrs = {attrName: attrValue}
else:
print(
"You will be shown the text of each of the elements on the page. If \
the text matches the content you are looking to scrape, type in yes. \
If not, Just press enter.")
elems = scraper.getAllElements()
for i in range(2, len(elems)):
content = elems[i].text
if content != "":
print(elems[i].text)
if input("Is this correct?") == 'yes':
#Find element attributes
elem = elems[i]
tag = elem.name
attrs = elem.attrs
break
if tag == '':
print("No element selected. Exiting.")
exit()
pdf = convert.Converter()
text = scraper.getText(tag, attrs)
pdf.addText(text)
print(text)
index += 1
address = prefix + str(index) + suffix
scraper.goTo(address)
while titleKeyword in scraper.getText('title'):
text = scraper.getText(tag, attrs)
pdf.addText(text)
print(text)
index += 1
address = prefix + str(index) + suffix
scraper.goTo(address)
pdf.printPDF(title)
print("done!")
