def make_bom():
"""Generate a bill of materials and return the tt object."""
finder = ModuleFinder()
finder.run_script(sys.argv[0])
tt = ttable()
tt.add_option(LONGTABLE)
tt.add_head(['name', 'ver', 'path', 'lines', 'bytes', 'sha-1'])
tt.set_col_alignments("lllrrl")
for inc in [False, True]:
for name, mod in sorted(finder.modules.items()):
if system_module(mod) != inc:
continue # don't use this one
stats = file_stats(mod.__file__)
ver = mod.globalnames.get('__version__', '---')
if ver == 1 and name in sys.modules:
# It has a version number; get it.
try:
ver = sys.modules[name].__version__
except AttributeError as e:
ver = '---'
fname = mod.__file__
if type(name) != str:
name = ""
if type(fname) != str:
fname = ""
tt.add_data([
latex_escape(name), ver,
latex_escape(fname), stats[0], stats[1], stats[2]
])
tt.add_data(ttable.HR)
return tt
def analyze_file(*,filename,mode=ttable.TEXT,verbose=False):
"""Analyze the named excel spreadsheet and return a string in the specified format."""
NL = tytable.line_end(mode)
ret = [NL] # an array of strings to be joined and returned
if mode==ttable.TEXT:
ret += [filename,'\n',"="*len(filename),NL]
elif mode==ttable.LATEX:
ret += [textbf(latex_escape(filename+":")),NL]
if filename.lower().endswith(".csv"):
wb = load_csv_workbook(filename=filename)
elif filename.lower().endswith(".xlsx"):
try:
wb = load_workbook(filename=filename, read_only=True)
except zipfile.BadZipFile:
ret += ['Cannot read; badZipFile',NL]
return "".join(ret)
except TypeError:
ret += ['Cannot read; bad Excel File',NL]
return "".join(ret)
else:
return "Cannot read file type: {}".format(filename)
tt = ttable()
tt.add_head(['Worksheet Name','rows with data','columns with data',
'total numeric values','cells with $>4$ sigfigs'])
ret_worksheets = []
for ws in wb.worksheets:
if verbose:
print(f" Analyzing {os.path.basename(filename)} worksheet {ws.title}")
sf_ws = SigFigStats() # for this worksheet
empty_cells = 0
for row in ws.iter_rows():
for cell in row:
if isinstance(cell,EmptyCell):
empty_cells += 1
continue
val = get_number(cell.value)
if val!=None:
sf_ws.add(val=val, row=cell.row, col=cell.column)
ret_worksheets += ["Worksheet "+latex_escape(ws.title)]
ret_worksheets += [sf_ws.typeset(mode=mode)]
ret_worksheets += ["rows: {} columns: {} numbers: {}".format(sf_ws.max_row,sf_ws.max_col,sf_ws.count)]
ret_worksheets += [""]
tt.add_data([latex_escape(ws.title),sf_ws.max_row,sf_ws.max_col,sf_ws.count, sf_ws.improper_count()])
# End of worksheet processing
tt.set_col_totals([3,4])
ret += [tt.typeset(mode=mode)] + [NL] + ret_worksheets
# count the number of numbers and their precision
return NL.join(ret)
def make_runtime():
"""Generate information about the runtime and return the tt object"""
tt = ttable()
tt.set_col_alignment(0, ttable.LEFT)
tt.set_col_alignment(1, ttable.LEFT)
tt.add_data(ttable.HR)
for (k, v) in [["hostname", socket.gethostname()],
["uptime", shell("uptime")],
["time", datetime.datetime.now().isoformat()[0:19]]]:
tt.add_data([k, v])
tt.add_data(ttable.HR)
return tt
def analyze_xlsx(*,filename,mode=TEXT):
"""Analyze the named excel spreadsheet and return a string in the specified format."""
ret = [] # an array of strings to be joined and returned
NL = tytable.line_end(mode)
if mode==TEXT:
ret += [filename,'\n',"="*len(filename),NL]
elif mode==LATEX:
ret += [textbf(latex_escape(filename+":")),NL]
try:
wb = load_workbook(filename=filename, read_only=True)
except zipfile.BadZipFile:
ret += ['Cannot read; badZipFile',NL]
return "".join(ret)
except TypeError:
ret += ['Cannot read; bad Excel File',NL]
return "".join(ret)
tt = ttable()
tt.add_head(['','Worksheet Name','rows with data','columns with data',
'total numeric values','cells with $>4$ sigfigs'])
sb = SigFigStats() # for all the worksheets
for i in range(len(wb.worksheets)):
ws = wb.worksheets[i]
wssb = SigFigStats() # for this worksheet
improper_count = 0
for row in ws.iter_rows():
for cell in row:
if isinstance(cell,EmptyCell):
continue
try:
if cell.value==None:
continue
val = float(cell.value)
except (ValueError,TypeError,OverflowError) as e:
continue
if improperly_rounded(val):
sb.add(val=val, row=cell.row+1, col=cell.column+1)
wssb.add(val=val, row=cell.row+1, col=cell.column+1)
improper_count += 1
tt.add_data([i+1,latex_escape(ws.title),wssb.max_row,wssb.max_col,wssb.count, improper_count])
# End of worksheet processing
tt.add_data(tt.HR)
tt.add_data(['total','--','--','--',sb.count])
ret += [tt.typeset(mode=mode)]
ret += [sb.typeset(mode=mode)]
# count the number of numbers and their precision
ret.append(NL)
return "\n".join(ret)
def typeset(self,*,mode):
if not self.count:
return ""
t = ttable()
t.add_head(['Significant Digits','Cell count','Problem cells'])
t.set_latex_colspec("rrl")
t.set_col_alignment(1,ttable.CENTER)
for (key,val) in sorted(self.sig_digits_histogram.items()):
alerts = " ".join( str(alert) for alert in self.sig_digits_alerts[key][0:ALERT_VALUES])
if len(self.sig_digits_alerts[key]) >= ALERT_VALUES:
alerts += r" \dots"
t.add_data( (key,val,alerts) )
return t.typeset(mode=mode)
def typeset(self,*,mode):
if not self.count:
return ""
t = ttable()
t.add_head(['Significant Digits','Cell count','Problem cells'])
t.set_latex_colspec("rrl")
t.set_col_alignment(1,ttable.CENTER)
for (sig_digs,vals) in sorted(self.sig_digits_alerts.items()):
alert_cells = sorted(vals)[0:ALERT_VALUES]
alerts = " ".join( [str(a) for a in alert_cells])
if len(alert_cells) >= ALERT_VALUES:
alerts += r" . . . "
t.add_data( (sig_digs,len(vals),alerts) )
if t.data:
return t.typeset(mode=mode)
else:
return f"Cells: {self.count}; nothing improperly rounded."
count = 0
considered = 0
rows = []
for a in range(1,115):
for b in range(a,115):
for c in range(b,115):
mean = statistics.mean([a,b,c])
median = statistics.median([a,b,c])
considered += 1
if median==MEDIAN and mean==MEAN:
rows.append([a,b,c])
count += 1
# Now build a multicolumn table
COLUMNS = 3
tt = tytable.ttable()
tt.add_variable("mymedian",MEDIAN)
tt.add_variable("mymean",MEAN)
tt.add_head(['a','b','c'] * COLUMNS)
total_rows = len(rows)
for i in range(total_rows//COLUMNS):
row = []
for col in range(0,3):
offset = (total_rows * col)//COLUMNS + i
if offset < len(rows):
row += rows[offset]
else:
row += ['']*len(rows[0])
tt.add_data(row)
tt.set_latex_colspec('@{\hspace{1pc}}|@{\hspace{1pc}}'.join(['rrr'] * COLUMNS))