Mercurial > repos > iuc > table_compute
changeset 1:dddadbbac949 draft
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/table_compute commit 6820ec9431a22576f3716c40feeb27f0b8cf5e83"
| author | iuc |
|---|---|
| date | Fri, 30 Aug 2019 05:28:18 -0400 |
| parents | 1b0f96ed73f2 |
| children | 02c3e335a695 |
| files | scripts/safety.py scripts/table_compute.py table_compute.xml test-data/skiplines.tsv |
| diffstat | 4 files changed, 503 insertions(+), 109 deletions(-) [+] |
line wrap: on
line diff
--- a/scripts/safety.py Sat Aug 17 16:25:37 2019 -0400 +++ b/scripts/safety.py Fri Aug 30 05:28:18 2019 -0400 @@ -11,6 +11,7 @@ '(', ')', 'if', 'else', 'or', 'and', 'not', 'in', '+', '-', '*', '/', '%', ',', '!=', '==', '>', '>=', '<', '<=', 'min', 'max', 'sum', + 'str', 'int', 'float' ) __allowed_ref_types = { 'pd.DataFrame': { @@ -163,26 +164,25 @@ safe = True # examples of user-expressions - # '-math.log(1 - elem/4096) * 4096 if elem != bn else elem - 0.5' + # '-math.log(1 - elem/4096) * 4096 if elem != 1 else elem - 0.5' # 'vec.median() + vec.sum()' # 1. Break expressions into tokens # e.g., # [ # '-', 'math.log', '(', '1', '-', 'elem', '/', '4096', ')', '*', - # '4096', 'if', 'elem', '!=', 'bn', 'else', 'elem', '-', '0.5' + # '4096', 'if', 'elem', '!=', '1', 'else', 'elem', '-', '0.5' # ] # or # ['vec.median', '(', ')', '+', 'vec.sum', '(', ')'] tokens = [ e for e in re.split( - r'([a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr + r'("[a-zA-Z%0-9_.]+"|[a-zA-Z0-9_.]+|[^a-zA-Z0-9_.() ]+|[()])', self.expr ) if e.strip() ] # 2. Subtract allowed standard tokens rem = [e for e in tokens if e not in self.__allowed_tokens] - # 3. Subtract allowed qualified objects from allowed modules # and whitelisted references and their attributes rem2 = [] @@ -194,18 +194,32 @@ if len(parts) == 2: if parts[0] in self.these: parts[0] = '_this' + elif parts[0] == "": + # e.g. '.T' gives ['','.T'] + # Here we assume that the blank part[0] refers to the + # self.ref_type (e.g. "pd.DataFrame"), and that + # the second part is a function of that type. + if parts[1] in self.allowed_qualified['_this']: + continue + if parts[0] in self.allowed_qualified: if parts[1] in self.allowed_qualified[parts[0]]: continue + rem2.append(e) - # 4. Assert that rest are real numbers + # Debug + # for x in (tokens, rem, rem2):print(x) + + # 4. Assert that rest are real numbers or strings e = '' for e in rem2: try: _ = float(e) except ValueError: - safe = False - break + # e.g. '"TEXT"' is okay. + if not(e[0] == '"' and e[-1] == '"'): + safe = False + break return safe, e
--- a/scripts/table_compute.py Sat Aug 17 16:25:37 2019 -0400 +++ b/scripts/table_compute.py Fri Aug 30 05:28:18 2019 -0400 @@ -3,7 +3,8 @@ Table Compute tool - a wrapper around pandas with parameter input validation. """ -__version__ = "0.8" + +__version__ = "0.9.1" import csv import math @@ -11,15 +12,17 @@ import numpy as np import pandas as pd -import userconfig as uc from safety import Safety -# This should be generated in the same directory -# Version command should not need to copy the config if len(argv) == 2 and argv[1] == "--version": print(__version__) exit(-1) +# The import below should be generated in the same directory as +# the table_compute.py script. +# It is placed here so that the --version switch does not fail +import userconfig as uc # noqa: I100,I202 + class Utils: @staticmethod @@ -37,12 +40,74 @@ "Returns a valid two value DataFrame or Series operator" return getattr(type(pd_obj), "__" + op_name + "__") + @staticmethod + def readcsv(filedict, narm): + data = pd.read_csv( + filedict["file"], + header=filedict["header"], + index_col=filedict["row_names"], + keep_default_na=narm, + nrows=filedict["nrows"], + skipfooter=filedict["skipfooter"], + skip_blank_lines=filedict["skip_blank_lines"], + sep='\t' + ) + # Fix whitespace issues in index or column names + data.columns = [col.strip() if type(col) is str else col + for col in data.columns] + data.index = [row.strip() if type(row) is str else row + for row in data.index] + return(data) -# Math is imported but not directly used because users -# may specify a "math.<function>" when inserting a custom -# function. To remove linting errors, which break CI testing -# we will just use an arbitrary math statement here. -__ = math.log + @staticmethod + def rangemaker(tab): + # e.g. "1:3,2:-2" specifies "1,2,3,2,1,0,-1,-2" to give [0,1,2,1,0,-1,-2] + # Positive indices are decremented by 1 to reference 0-base numbering + # Negative indices are unaltered, so that -1 refers to the last column + out = [] + err_mess = None + for ranges in tab.split(","): + nums = ranges.split(":") + if len(nums) == 1: + numb = int(nums[0]) + # Positive numbers get decremented. + # i.e. column "3" refers to index 2 + # column "-1" still refers to index -1 + if numb != 0: + out.append(numb if (numb < 0) else (numb - 1)) + else: + err_mess = "Please do not use 0 as an index" + elif len(nums) == 2: + left, right = map(int, nums) + if 0 in (left, right): + err_mess = "Please do not use 0 as an index" + elif left < right: + if left > 0: # and right > 0 too + # 1:3 to 0,1,2 + out.extend(range(left - 1, right)) + elif right < 0: # and left < 0 too + # -3:-1 to -3,-2,-1 + out.extend(range(left, right + 1)) + elif left < 0 and right > 0: + # -2:2 to -2,-1,0,1 + out.extend(range(left, 0)) + out.extend(range(0, right)) + elif right < left: + if right > 0: # and left > 0 + # 3:1 to 2,1,0 + out.extend(range(left - 1, right - 2, -1)) + elif left < 0: # and right < 0 + # -1:-3 to -1,-2,-3 + out.extend(range(left, right - 1, -1)) + elif right < 0 and left > 0: + # 2:-2 to 1,0,-1,-2 + out.extend(range(left - 1, right - 1, -1)) + else: + err_mess = "%s should not be equal or contain a zero" % nums + if err_mess: + print(err_mess) + return(None) + return(out) # Set decimal precision @@ -55,19 +120,7 @@ if user_mode == "single": # Read in TSV file - data = pd.read_csv( - uc.Data["tables"][0]["reader_file"], - header=uc.Data["tables"][0]["reader_header"], - index_col=uc.Data["tables"][0]["reader_row_col"], - keep_default_na=uc.Default["narm"], - sep='\t' - ) - # Fix whitespace issues in index or column names - data.columns = [col.strip() if type(col) is str else col - for col in data.columns] - data.index = [row.strip() if type(row) is str else row - for row in data.index] - + data = Utils.readcsv(uc.Data["tables"][0], uc.Default["narm"]) user_mode_single = params["user_mode_single"] if user_mode_single == "precision": @@ -79,9 +132,13 @@ rows_specified = params["select_rows_wanted"] # Select all indexes if empty array of values - if not cols_specified: + if cols_specified: + cols_specified = Utils.rangemaker(cols_specified) + else: cols_specified = range(len(data.columns)) - if not rows_specified: + if rows_specified: + rows_specified = Utils.rangemaker(rows_specified) + else: rows_specified = range(len(data)) # do not use duplicate indexes @@ -161,16 +218,44 @@ elif user_mode_single == "element": # lt, gt, ge, etc. operation = params["element_op"] + bool_mat = None if operation is not None: - op = Utils.getTwoValuePandaOp(operation, data) - value = params["element_value"] - try: - # Could be numeric - value = float(value) - except ValueError: - pass - # generate filter matrix of True/False values - bool_mat = op(data, value) + if operation == "rowcol": + # Select all indexes if empty array of values + if "element_cols" in params: + cols_specified = Utils.rangemaker(params["element_cols"]) + else: + cols_specified = range(len(data.columns)) + if "element_rows" in params: + rows_specified = Utils.rangemaker(params["element_rows"]) + else: + rows_specified = range(len(data)) + + # Inclusive selection: + # - True: Giving a row or column will match all elements in that row or column + # - False: Give a row or column will match only elements in both those rows or columns + inclusive = params["element_inclusive"] + + # Create a bool matrix (intialised to False) with selected + # rows and columns set to True + bool_mat = data.copy() + bool_mat[:] = False + if inclusive: + bool_mat.iloc[rows_specified, :] = True + bool_mat.iloc[:, cols_specified] = True + else: + bool_mat.iloc[rows_specified, cols_specified] = True + + else: + op = Utils.getTwoValuePandaOp(operation, data) + value = params["element_value"] + try: + # Could be numeric + value = float(value) + except ValueError: + pass + # generate filter matrix of True/False values + bool_mat = op(data, value) else: # implement no filtering through a filter matrix filled with # True values. @@ -265,13 +350,7 @@ # Read and populate tables for x, t_sect in enumerate(table_sections): - tmp = pd.read_csv( - t_sect["file"], - header=t_sect["header"], - index_col=t_sect["row_names"], - keep_default_na=uc.Default["narm"], - sep="\t" - ) + tmp = Utils.readcsv(t_sect, uc.Default["narm"]) table.append(tmp) table_names.append("table" + str(x + 1)) table_names_real.append("table[" + str(x) + "]")
--- a/table_compute.xml Sat Aug 17 16:25:37 2019 -0400 +++ b/table_compute.xml Fri Aug 30 05:28:18 2019 -0400 @@ -1,7 +1,7 @@ <tool id="table_compute" name="Table Compute" version="@VERSION@"> <description>computes operations on table data</description> <macros> - <token name="@VERSION@">0.8</token> + <token name="@VERSION@">0.9.1</token> <token name="@COPEN@"><![CDATA[<code>]]></token> <token name="@CCLOSE@"><![CDATA[</code>]]></token> <import>allowed_functions.xml</import> @@ -19,11 +19,11 @@ <sanitizer sanitize="false" /> </macro> <macro name="validator_index_ranges"> - <validator type="regex" message="Specify a comma-separated list index numbers or ranges">^(?:\d+(?::\d)*(?:, *\d+(?::\d)*)*)?$</validator> + <validator type="regex" message="Specify a comma-separated list index numbers or ranges">^(?:-?\d+(?::-?\d+)*(?:, *-?\d+(?::-?\d+)*)*)?$</validator> <sanitizer sanitize="false" /> </macro> <macro name="validator_functiondef"> - <validator type="regex" message="An expression is required and is allowed to contain only letters, numbers and the characters '_ !-+=/*%.<>()'">^[\w !\-+=/*%,.<>()]+$</validator> + <validator type="regex" message="An expression is required and is allowed to contain only letters, numbers and the characters '_ !-+=/*%.<>()'">^['"\w !\-+=/*%,.<>()]+$</validator> <sanitizer sanitize="false" /> </macro> <!-- macro for main input tests --> @@ -36,6 +36,16 @@ </conditional> </conditional> </macro> + <macro name="test_inputs_ranges" > + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.1.tsv" /> + <conditional name="user" > + <param name="mode" value="select" /> + <yield /> + </conditional> + </conditional> + </macro> <!-- macro for umi to transcript tests --> <macro name="umi2trans" > <yield /> @@ -53,14 +63,31 @@ <macro name="file_opts"> <param name="input" type="data" format="tsv,tabular" label="Table" /> <param name="col_row_names" type="select" display="checkboxes" multiple="true" optional="true" - label="This input data has"> + label="Input data has"> <option value="has_col_names" selected="true">Column names on the first row</option> - <option value="has_row_names" selected="true">Row names on the first column"</option> + <option value="has_row_names" selected="true">Row names on the first column</option> </param> + <section name="adv" title="Advanced File Options " expanded="false" > + <param name="header" type="integer" min="0" optional="true" label="Header begins at line N" help="All lines before line N will be skipped. If a value is set, this will override the above 'Column names on the first row' parameter." /> + <param name="nrows" type="integer" min="0" optional="true" label="Read N lines only" help="Parses only N lines after the header line." /> + <param name="skipfooter" type="integer" min="0" optional="true" label="Skip N lines from bottom" help="Do not use this in conjunction with the 'Read N lines only' parameter." /> + <param name="skip_blank_lines" type="boolean" checked="true" falsevalue="False" truevalue="True" label="Skip blank lines" help="Otherwise it will insert NaN values for every blank line detected." /> + </section> </macro> <!-- element value macro --> - <macro name="elem_val_macro" > - <param name="element_value" type="text" optional="true" label="Filter value" help="This value is converted to numeric if possible, otherwise it is treated as a string" /> + <macro name="elem_val_macro" token_when="@WHEN@" > + <when value="@WHEN@" > + <param name="element_value" type="text" optional="true" label="Filter value" help="This value is converted to numeric if possible, otherwise it is treated as a string" /> + </when> + </macro> + <!-- Row/Col macro --> + <macro name="select_rowcols" > + <param name="select_cols_wanted" type="text" optional="true" label="List of columns to select" help="Comma separated. (e.g. @COPEN@3:5,10,-1,2:-2@CCLOSE@ will select columns @COPEN@3,4,5,10,-1,2,1,0,-1,-2@CCLOSE@, where negative indices refer to the offset from the last index). Columns can be duplicated by specifying them multiple times. Leave blank to retain all columns. See Example #1 for an example of using this mode." > + <expand macro="validator_index_ranges" /> + </param> + <param name="select_rows_wanted" type="text" optional="true" label="List of rows to select" help="The same rules apply as above" > + <expand macro="validator_index_ranges" /> + </param> </macro> </macros> <requirements> @@ -69,7 +96,7 @@ </requirements> <version_command><![CDATA[ - touch '$__tool_directory__/scripts/userconfig.py' && python '$__tool_directory__/scripts/table_compute.py' --version + python '$__tool_directory__/scripts/table_compute.py' --version ]]></version_command> <command detect_errors="exit_code"><![CDATA[ @@ -83,15 +110,6 @@ ]]></command> <configfiles> <configfile name="userconf"><![CDATA[ - -## Range Maker -## "2:5,11,1:2" specifies "2,3,4,5,11,1,2" which in python is "1,2,3,4,10,0,1" -#def rangemake(tab): -#echo [(n-1) for r in map(lambda y: range(int(y[0]),int(y[-1])+1), map(lambda x: x.split(':'), tab.split(','))) for n in r] -#end def - -## Strip leading and trailing whitespace off custom functions - Default = { "reader_skip": 0, "precision": $precision, @@ -118,16 +136,25 @@ Data = { "tables": [ { - "reader_file": '$singtabop.input', - #if 'has_col_names' in str($singtabop.col_row_names): - "reader_header": 0, + "file": '$singtabop.input', + #if $singtabop.adv.header: + "header": int('$singtabop.adv.header'), + #elif 'has_col_names' in str($singtabop.col_row_names): + "header": 0, #else: - "reader_header": None, + "header": None, #end if - #if 'has_row_names' in str($singtabop.col_row_names): - "reader_row_col": 0, - #else: - "reader_row_col": False, + "row_names": #if ('has_row_names' in str($singtabop.col_row_names)) then 0 else False#, + "skipfooter": #if $singtabop.adv.skipfooter then int($singtabop.adv.skipfooter) else 0#, + #if $singtabop.adv.skip_blank_lines + "skip_blank_lines": $singtabop.adv.skip_blank_lines, + #else + "skip_blank_lines": None, + #end if + #if $singtabop.adv.nrows + "nrows": int('$singtabop.adv.nrows'), + #else + "nrows": None, #end if } ], @@ -141,12 +168,12 @@ "select_cols_unique": #echo 'select_cols_keepdupe' in str($singtabop.user.select_keepdupe)#, "select_rows_unique": #echo 'select_rows_keepdupe' in str($singtabop.user.select_keepdupe)#, #if $singtabop.user.select_cols_wanted: - "select_cols_wanted": $rangemake(str($singtabop.user.select_cols_wanted)), + "select_cols_wanted": '$singtabop.user.select_cols_wanted', #else "select_cols_wanted": None, #end if #if $singtabop.user.select_rows_wanted: - "select_rows_wanted": $rangemake(str($singtabop.user.select_rows_wanted)), + "select_rows_wanted": '$singtabop.user.select_rows_wanted', #else "select_rows_wanted": None, #end if @@ -181,12 +208,21 @@ #end if } #elif $singtabop.user.mode.value == 'element': - #if str($singtabop.user.elem_val.element_op) != "None": + #if str($singtabop.user.elem_val.element_op) == "None": + "element_op": None, + #elif str($singtabop.user.elem_val.element_op) == "rowcol": + "element_op": "rowcol", + #if $singtabop.user.elem_val.select_rows_wanted + "element_rows": '$singtabop.user.elem_val.select_rows_wanted', + #end if + #if $singtabop.user.elem_val.select_cols_wanted + "element_cols": '$singtabop.user.elem_val.select_cols_wanted', + #end if + "element_inclusive": $singtabop.user.elem_val.inclusive_selection, + #else: "element_op": '$singtabop.user.elem_val.element_op.value', ## Value is string or float, parsed in code later "element_value" : '$singtabop.user.elem_val.element_value', - #else: - "element_op": None, #end if "element_mode": '$singtabop.user.element.mode.value', #if str($singtabop.user.element.mode) == "replace": @@ -252,16 +288,26 @@ #for $i, $s in enumerate($singtabop.tables) { "file": '${s.input}', - #if 'has_col_names' in str($s.col_row_names): + #if $s.adv.header: + "header": int('${s.adv.header}'), + #elif 'has_col_names' in str($s.col_row_names): "header": 0, #else: "header": None, #end if - #if 'has_row_names' in str($s.col_row_names): - "row_names": 0, - #else: - "row_names": False, + "row_names": #if ('has_row_names' in str($s.col_row_names)) then 0 else False#, + "skipfooter": #if $s.adv.skipfooter then int('$s.adv.skipfooter') else 0#, + #if $s.adv.skip_blank_lines: + "skip_blank_lines": $s.adv.skip_blank_lines, + #else + "skip_blank_lines": None, #end if + #if $s.adv.nrows: + "nrows": int('${s.adv.nrows}'), + #else + "nrows": None, + #end if + }, #end for ], @@ -293,12 +339,7 @@ </param> <when value="precision" /> <when value="select"> - <param name="select_cols_wanted" type="text" optional="true" label="List of columns to select" help="Comma separated. (e.g. @COPEN@3:5,99,2:4@CCLOSE@ will select columns @COPEN@3,4,5,99,2,3,4@CCLOSE@). Columns can be duplicated by specifying them multiple times. Leave blank to retain all columns. See Example #1 for an example of using this mode." > - <expand macro="validator_index_ranges" /> - </param> - <param name="select_rows_wanted" type="text" optional="true" label="List of rows to select" help="The same rules apply as above" > - <expand macro="validator_index_ranges" /> - </param> + <expand macro="select_rowcols" /> <param name="select_keepdupe" type="select" display="checkboxes" multiple="true" label="Duplicate Indices" help="Keep duplicates when specifying ranges (e.g. if unset, @COPEN@1:3,2:4@CCLOSE@ will yield @COPEN@1,2,3,4@CCLOSE@ instead of @COPEN@1,2,3,2,3,4@CCLOSE@)" > <option value="select_cols_keepdupe" selected="true" >Keep duplicate columns</option> <option value="select_rows_keepdupe" selected="true" >Keep duplicate rows</option> @@ -447,38 +488,30 @@ <param name="scale_value" type="float" value="0" label="Second operand value" /> </when> <when value="custom"> - <param name="custom_expr" type="text" - label="Custom expression on 'elem'" - help="The parameter name is @COPEN@elem@CCLOSE@, referring to the element being acted on. Most operators and @COPEN@if@CCLOSE@ @COPEN@else@CCLOSE@ statements are supported. See Examples #3 and #4 in the Help section."> + <param name="custom_expr" type="text" label="Custom expression on 'elem'" + help="The parameter name is @COPEN@elem@CCLOSE@, referring to the element being acted on. Most operators and @COPEN@if@CCLOSE@ @COPEN@else@CCLOSE@ statements are supported. See Examples #3 and #4 in the Help section."> <expand macro="validator_functiondef" /> </param> </when> </conditional> <conditional name="elem_val" > <param name="element_op" type="select" label="Operate on elements" - help="Only selected elements will be manipulated. Other elements will retain their original value." > + help="Only selected elements will be manipulated. Other elements will retain their original value." > <expand macro="select_twovaluebooleanops" > <option value="None" selected="true">All</option> + <option value="rowcol" >Specific Rows and/or Columns</option> </expand> </param> <when value="None" /> - <when value="lt"> - <expand macro="elem_val_macro" /> - </when> - <when value="le"> - <expand macro="elem_val_macro" /> - </when> - <when value="gt"> - <expand macro="elem_val_macro" /> - </when> - <when value="ge"> - <expand macro="elem_val_macro" /> - </when> - <when value="eq"> - <expand macro="elem_val_macro" /> - </when> - <when value="ne"> - <expand macro="elem_val_macro" /> + <expand macro="elem_val_macro" when="lt" /> + <expand macro="elem_val_macro" when="le" /> + <expand macro="elem_val_macro" when="gt" /> + <expand macro="elem_val_macro" when="ge" /> + <expand macro="elem_val_macro" when="eq" /> + <expand macro="elem_val_macro" when="ne" /> + <when value="rowcol" > + <expand macro="select_rowcols" /> + <param name="inclusive_selection" type="boolean" truevalue="True" falsevalue="False" checked="true" label="Inclusive Selection" help="If enabled, all elements that match ANY row or column given above will be acted on. Otherwise, only the elements that match BOTH a row or column given above will be acted on." /> </when> </conditional> </when> @@ -1039,6 +1072,205 @@ </assert_contents> </output> </test> + <test expect_num_outputs="1" > + <!-- Test 33: Subtracting table means #1 --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.5.tsv" /> + <param name="col_row_names" value="has_col_names,has_row_names" /> + <conditional name="user" > + <param name="mode" value="fulltable" /> + <conditional name="general" > + <param name="use" value="custom" /> + <param name="fulltable_custom_expr" value="table - table.mean(1)/table.std(1)" /> + </conditional> + </conditional> + </conditional> + <output name="table" > + <assert_contents> + <has_n_columns n="8" /> + <!-- should give nonsense empty output --> + <has_line_matching expression="^\sc1\sc2\sc3\sg1\sg2\sg3\sg4$" /> + <has_line_matching expression="^g4\s+$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 34: Subtracting table means #2 --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.5.tsv" /> + <param name="col_row_names" value="has_col_names,has_row_names" /> + <conditional name="user" > + <param name="mode" value="fulltable" /> + <conditional name="general" > + <param name="use" value="custom" /> + <param name="fulltable_custom_expr" value="table.sub(table.mean(1), 0).div(table.std(1),0)" /> + </conditional> + </conditional> + </conditional> + <output name="table" > + <assert_contents> + <has_n_columns n="4" /> + <has_line_matching expression="^g1\s+-1\.\d+\s+0\.\d+\s+1\.\d+\s*$" /> + <has_line_matching expression="^g4\s+1\.1\d+\s+-0\.5\d+\s+-0\.5\d+\s*$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 35: Subtracting table means #3 --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.5.tsv" /> + <param name="col_row_names" value="has_col_names,has_row_names" /> + <conditional name="user" > + <param name="mode" value="fulltable" /> + <conditional name="general" > + <param name="use" value="custom" /> + <param name="fulltable_custom_expr" value="(table.T - table.mean(1)).T" /> + </conditional> + </conditional> + </conditional> + <output name="table" > + <assert_contents> + <has_n_columns n="4" /> + <has_line_matching expression="^g1\s+-10\.\d+\s+0\.\d+\s+10\.\d+\s*$" /> + <has_line_matching expression="^g4\s+47\.3\d+\s+-23\.6\d+\s+-23\.6\d+\s*$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 36: Negative range test #1 --> + <expand macro="test_inputs_ranges" > + <param name="select_cols_wanted" value="-2:2,1,-1" /> + <param name="select_rows_wanted" value="1,3:-3,1" /> + </expand> + <output name="table" > + <assert_contents> + <has_n_columns n="7" /> + <has_line_matching expression="^\s*c2\s+c3\s+c1\s+c2\s+c1\s+c3$" /> + <has_line_matching expression="^g1\s+20\s+30\s+10\s+20\s+10\s+30$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 37: Negative range test #2 --> + <expand macro="test_inputs_ranges" > + <param name="select_cols_wanted" value="-1,-3,3:-3,2" /> + <param name="select_rows_wanted" value="3:-1" /> + <param name="select_keepdupe" value="" /> + </expand> + <output name="table" > + <assert_contents> + <has_n_columns n="7" /> + <has_line_matching expression="^\s*c3\s+c1\s+c3\s+c2\s+c1\s+c2$" /> + <has_line_matching expression="^g4\s+3\s+81\s+3\s+6\s+81\s+6$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1"> + <!-- Test 38: Skip first 3 lines --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="skiplines.tsv" /> + <section name="adv" > + <param name="header" value="2" /> + <param name="nrows" value="4" /> + <param name="skip_blank_lines" value="true" /> + </section> + <conditional name="user" > + <param name="mode" value="precision" /> + </conditional> + </conditional> + <param name="precision" value="2" /> + <output name="table" > + <assert_contents> + <has_n_columns n="4" /> + <has_line_matching expression="^\s+c1\s+c2\s+c3$" /> + <has_line_matching expression="^g4\s+81\s+6\s+3$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1"> + <!-- Test 39: Skip first 3 lines --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="skiplines.tsv" /> + <section name="adv" > + <param name="header" value="2" /> + <param name="skipfooter" value="2" /> + <param name="skip_blank_lines" value="false" /> + </section> + <conditional name="user" > + <param name="mode" value="precision" /> + </conditional> + </conditional> + <param name="precision" value="2" /> + <param name="out_opts" value="" /> + <output name="table" > + <assert_contents> + <has_n_columns n="3" /> + <has_line_matching expression="^10.00\s+20.00\s+30.00$" /> + <has_line_matching expression="^\s+$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 40: Row Col custom op #1 --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.1.tsv" /> + <param name="col_row_names" value="has_col_names,has_row_names" /> + <conditional name="user" > + <param name="mode" value="element" /> + <conditional name="element" > + <param name="mode" value="custom" /> + <param name="custom_expr" value=""chr%.f" % elem" /> + </conditional> + <conditional name="elem_val" > + <param name="element_op" value="rowcol" /> + <param name="select_cols_wanted" value="2" /> + <param name="select_rows_wanted" value="2,4" /> + <param name="inclusive_selection" value="True" /> + </conditional> + </conditional> + </conditional> + <output name="table" > + <assert_contents> + <has_n_columns n="4" /> + <has_line_matching expression="^g2\s+chr3\s+chr6\s+chr9$" /> + <has_line_matching expression="^g4\s+chr81\s+chr6\s+chr3$" /> + </assert_contents> + </output> + </test> + <test expect_num_outputs="1" > + <!-- Test 41: Row Col custom op #2 --> + <conditional name="singtabop" > + <param name="use_type" value="single" /> + <param name="input" value="examples.1.tsv" /> + <param name="col_row_names" value="has_col_names,has_row_names" /> + <conditional name="user" > + <param name="mode" value="element" /> + <conditional name="element" > + <param name="mode" value="custom" /> + <param name="custom_expr" value=""chr%.f" % elem" /> + </conditional> + <conditional name="elem_val" > + <param name="element_op" value="rowcol" /> + <param name="select_cols_wanted" value="2" /> + <param name="select_rows_wanted" value="2,4" /> + <param name="inclusive_selection" value="False" /> + </conditional> + </conditional> + </conditional> + <output name="table" > + <assert_contents> + <has_n_columns n="4" /> + <has_line_matching expression="^g2\s+3\s+chr6\s+9$" /> + <has_line_matching expression="^g4\s+81\s+chr6\s+3$" /> + </assert_contents> + </output> + </test> </tests> <help><![CDATA[ This tool computes table expressions on the element, row, and column basis. It can sub-select, @@ -1436,6 +1668,65 @@ This splits the matrix using "foo" and "bar" using only the values from "baz". Header values may contain extra information. + +Example 9: Replacing text in specific rows or columns +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +We have the following table + + === === === === + . c1 c2 c3 + === === === === + g1 10 20 30 + g2 3 3 9 + g3 4 8 12 + g4 81 6 3 + === === === === + +and we want to add "chr" to the elements in column 2 AND rows 2 and 4: + + === === ==== === + . c1 c2 c3 + === === ==== === + g1 10 20 30 + g2 3 chr3 9 + g3 4 8 12 + g4 81 chr6 3 + === === ==== === + +In Galaxy we would select the following: + + * *Input Single or Multiple Tables* → **Single Table** + * *Column names on first row?* → **Yes** + * *Row names on first column?* → **Yes** + + * *Type of table operation* → **Manipulate selected table elements** + + * *Operation to perform* → **Custom** + + * *Custom Expression* → :: + + "chr%.f" % elem + + * *Operate on elements* → **Specific Rows and/or Columns** + * *List of columns to select* → "2" + * *List of rows to select* → "2,4" + * *Inclusive Selection* → "No" + + +If we wanted to instead add "chr" to the ALL elements in column 2 and rows 2 and 4, we would repeat the steps above but set the *Inclusive Selection* to "Yes", to give: + + === ===== ===== ===== + . c1 c2 c3 + === ===== ===== ===== + g1 10 chr20 30 + g2 chr3 chr3 chr9 + g3 4 8 12 + g4 chr81 chr6 chr3 + === ===== ===== ===== + + + ]]></help> <citations></citations> </tool>