opensource/galaxy
3
0
Fork 0
mirror of https://github.com/galaxyproject/galaxy.git synced 2026-09-24 16:30:27 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fixes and cleanup to summary statistics tool and commented out tests for grouping tool since they require mouse clicks (not handled by twill).

Browse Source
This commit is contained in:
Greg Von Kuster
2007-08-20 20:33:15 +00:00
parent 5a7e2c3c6c
commit a3e9cb73dd
3 changed files with 44 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/stats/grouping.xml
+5
View File
@@ -33,19 +33,24 @@
</outputs>
<tests>
<!-- Test valid data -->
<!-- Testing this tool will not be possible unless / until the twill team inplements an
enhancement to allow for mouse clicks in the test cases...
<test>
<param name="input1" value="1.bed"/>
<param name="groupcol" value="1"/>
<param name="operations" value="mean 2"/>
<output name="out_file1" file="groupby_out1.dat"/>
</test>
-->
<!-- Test data with an invalid value in a column -->
<!--
<test>
<param name="input1" value="1.tabular"/>
<param name="groupcol" value="1"/>
<param name="operations" value="mean 2,c 3"/>
<output name="out_file1" file="groupby_out2.dat"/>
</test>
-->
</tests>
<help>
tools/stats/gsummary.py
+35 -25
View File
@@ -9,6 +9,10 @@ def stop_err(msg):
sys.stderr.write(msg)
sys.exit()
def stop_out(msg):
sys.stdout.write(msg)
sys.exit()
def mode_func(c):
try:
check = float(c)
@@ -85,15 +89,8 @@ def S3_METHODS(all="key"):
"lgamma", "gamma", "gammaCody",
"digamma", "trigamma",
"cumsum", "cumprod", "cummax", "cummin"]
Group_Ops = [ "+", "-", "*", "/", "^", "%%", "%/%",
"&", "|", "!",
"==", "!=", "<", "<=", ">=", ">"]
Group_Summary = [ "all", "any"
"sum", "prod",
"min", "max",
"range" ]
Group_Ops = [ "+", "-", "*", "/", "^", "%%", "%/%", "&", "|", "!", "==", "!=", "<", "<=", ">=", ">"]
Group_Summary = [ "all", "any", "sum", "prod", "min", "max", "range" ]
if all is "key": return { 'Math' : Group_Math, 'Ops' : Group_Ops, 'Summary' : Group_Summary }
@@ -104,7 +101,7 @@ def read_table(datafile, cols):
first_invalid_line = 0
for i, line in enumerate(file(datafile)):
valid = True
line = line.strip()
line = line.rstrip('\r\n')
if line and not line.startswith( '#' ):
f = line.split()
for col in cols:
@@ -139,7 +136,7 @@ def read_table(datafile, cols):
width = len(f)
if len(table) > 0:
"""
terms will look like this: r dataframe: c1=table['c1'], c2=table['c2'], ...
terms will look like this: c7 = r.as_numeric(table["c7"]), c8 = r.as_numeric(table["c8"]), ...
"""
terms = ["%s = %s(table[\"%s\"])" % (x, mode_func(table[x][0]), x) for x in ["c" + str(col_i + 1) for col_i in range(0, width)]]
code = "d = r.data_frame(%s)" % ",".join(terms)
@@ -156,10 +153,10 @@ def main():
if len(sys.argv) >= 4:
datafile = sys.argv[1]
outfile = sys.argv[2]
formula = sys.argv[3]
expression = sys.argv[3]
else:
print sys.argv
stop_err('Usage: python gsummary.py input_file ouput_file formula')
stop_err('Usage: python gsummary.py input_file ouput_file expression')
if len(sys.argv) == 5:
if sys.argv[4].find('none') < 0:
@@ -169,7 +166,7 @@ def main():
if len(group_terms) > 0:
dep_var = group_terms[0]
tmp_rhs = "|".join([ dep_var, tmp_rhs])
formula = '~'.join([formula,tmp_rhs])
expression = '~'.join([expression,tmp_rhs])
else:
stop_err("%s unrecognized for groups" % tmp_rhs)
elif sys.argv[4] is 'none':
@@ -192,20 +189,32 @@ def main():
of = open(outfile,'w')
if formula.find("~") > 0:
lhs,rhs = formula.split('~')
if expression.find("~") > 0:
lhs,rhs = expression.split('~')
else:
lhs = formula
lhs = expression
for word in re.compile('[a-zA-Z]+').findall(formula):
for word in re.compile('[a-zA-Z]+').findall(expression):
if word and not word in lhs_allowed:
of.close()
stop_err("Invalid formula '%s': Function/term '%s' is not recognized or allowed" % (formula, word))
stop_out("Invalid expression '%s': term '%s' is not recognized or allowed" % (expression, word))
for symbol in re.compile('[^a-z0-9\s]+').findall(formula):
"""
Users sometimes want statistics for more than 1 column, so they enter a comma-separated
string of columns in the free text field. This tool only handles a single column or an
expression (computed for 1 or more columns), so we'll use the following hack to provide a
useful response for multiple column entries.
"""
symbols = sets.Set()
for symbol in re.compile('[^a-z0-9\s]+').findall(expression):
if symbol and not symbol in ops_allowed:
of.close()
stop_err("Invalid formula '%s': Operator '%s' is not recognized or allowed" % (formula, symbol))
stop_out("Invalid expression '%s': operator '%s' is not recognized or allowed" % (expression, symbol))
else:
symbols.add(symbol)
if len(symbols) == 1 and ',' in symbols:
of.close()
stop_out( "Invalid columns '%s': this tool requires a single column or expression" %expression )
cols = []
if lhs:
@@ -235,7 +244,8 @@ def main():
summary = f(r(lhs))
except RException, s:
# Due to previous checking, this should not occur
stop_err("Computation attempted on invalid data in column %s. Exception:\n\t%s" %(lhs, s))
of.close()
stop_err("Computation attempted on invalid data in column %s. Exception: %s" %(lhs, s))
summary = summary.as_py(BASIC_CONVERSION)
print >>of,"#%s" % "\t".join(returns)
@@ -249,10 +259,10 @@ def main():
set_default_mode(NO_CONVERSION)
try:
df_g = r.groupedData(r.formula(formula), df)
df_r = r.groupedData(r.formula(formula), r.data_frame(df_g, response=r.getResponse(df_g)))
df_g = r.groupedData(r.expression(expression), df)
df_r = r.groupedData(r.expression(expression), r.data_frame(df_g, response=r.getResponse(df_g)))
except RException, s:
stop-err("Computation attempted on invalid data in column on the left hand side of expression. Excpetion:\n\t%s" % s)
stop-err("Computation attempted on invalid data in column on the left hand side of expression. Exception:\n\t%s" % s)
"""
Try some plotting stuff
tools/stats/gsummary.xml
+4 -7
View File
@@ -1,21 +1,18 @@
<tool id="Summary Statistics1" name="Summary Statistics">
<tool id="Summary_Statistics1" name="Summary Statistics">
<description>for any numerical column</description>
<command interpreter="python">gsummary.py $input $out_file1 "$cond" "none"</command>
<inputs>
<param name="cond" size="40" type="text" value="c5" label="expression" help="Columns are specified as c1, c2, c3 ... First column is c1"/>
<!-- <param name="groups" size="40" type="text" value="none" label="group terms (c1,c4,etc.)"/>
-->
<param format="tabular" name="input" type="data" label="summary statistics on" help="Query missing? See TIP below"/>
<param format="tabular" name="input" type="data" label="Summary statistics on" help="Query missing? See TIP below"/>
<param name="cond" size="40" type="text" value="c5" label="Column or expression" help="See tool syntax below" />
</inputs>
<outputs>
<data format="tabular" name="out_file1" />
</outputs>
<tests>
<test>
<param name="cond" value="c2"/>
<param name="input" value="1.bed"/>
<output name="out_file1" file="sta_summary.dat"/>
<param name="cond" value="c2"/>
</test>
</tests>
<help>