Sponsored Content
Top Forums UNIX for Dummies Questions & Answers Search and replace with a sliding window Post 302907340 by bakunin on Friday 27th of June 2014 09:51:01 AM
Old 06-27-2014
Quote:
Originally Posted by Fahmida
(2)Form a 3 letter window and slide 1 letter at a time; check for the patten described above until finish replacing.
To be honest, i have troubles understanding the exact meaning of this. I suppose you want to work on different reading frames.

Because there are only 3 of them (using "T" as any triplett and "B" for any single base there is only "TTT...", "BTTT..." and "BBTTT...") you will not need a sliding window.

Second, because of your first replacements a pattern could match where it hasn't matched before in subsequent passes. Suppose your line is:

Code:
DDGG

Because it will not match "DDG" the triplett 2-4 would not be replaced. Once triplett 1-3 would be replaced, it would match, though:

Code:
DDGG        # replace "DDG" with "DD!"
DD!G        # after first replacement, now prelace "2 not-Gs, then G" again
DD!!        # after second pass

Analogous for your other replacement rules. Somehow i doubt that this is what you really want. Please clarify.

I hope this helps.

bakunin
 

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Perl: Search for string on line then search and replace text

Hi All, I have a file that I need to be able to find a pattern match on a line, search that line for a text pattern, and replace that text. An example of 4 lines in my file is: 1. MatchText_randomNumberOfText moreData ReplaceMe moreData 2. MatchText_randomNumberOfText moreData moreData... (4 Replies)
Discussion started by: Crypto
4 Replies

2. Shell Programming and Scripting

awk - replace number of string length from search and replace for a serialized array

Hello, I really would appreciate some help with a bash script for some string manipulation on an SQL dump: I'd like to be able to rename "sites/WHATEVER/files" to "sites/SOMETHINGELSE/files" within the sql dump. This is quite easy with sed: sed -e... (1 Reply)
Discussion started by: otrotipo
1 Replies

3. Shell Programming and Scripting

Script help: sliding time windows

I have a script like this ... #!/bin/ksh database=$(echo $@ | sed 's/.*-S \(*\).*/\1/') instance=$(grep $database /var/opt/oracle/oratab | awk -F : '{print $1}') command=$(echo $@ | sed "s/$database/$instance/") echo $command if I execute this script in ksh or bash it works fine . ... (3 Replies)
Discussion started by: talashil
3 Replies

4. Shell Programming and Scripting

Sliding window for sequencing data

Hi! I have some sequencing data that I have aligned using maq software Now, I have data that looks like this each line is a 'tag' chr1 10001 chr1 10002 chr1 10005 chr1 10007 chr1 10008 chr1 10008 chr1 10008 chr1 10019 chr1 10019 chr1 10020 What I really want to find out is how... (1 Reply)
Discussion started by: biobio
1 Replies

5. UNIX for Dummies Questions & Answers

Sliding window

Very simple problem I am not able to solve. I have been trying to modify the following code: awk '{t=$1; c = x}{for (i = 1; i <= length; i += wn)print t FS"" substr($2, i, mx) > ("block" ++c)}' mx=100 wn=100 infile.txt What I am tryng to acccomplish, I have a bunch of files where the first... (3 Replies)
Discussion started by: Xterra
3 Replies

6. UNIX for Dummies Questions & Answers

"Sliding window" with variables

I'm doing a little work that involves computing the average completion time of the last 5 of many file decompressions. It's not too tough, but I'm wondering if maybe there's a better way to write it. This is a bash script; here's the current idea: ctime5=$ctime4 ctime4=$ctime3 ctime3=$ctime2... (2 Replies)
Discussion started by: treesloth
2 Replies

7. Shell Programming and Scripting

Sliding window for string manipulation

I have a sting of "0"s and "1"s that I need to analyze. I need to look at each "1" and determine if it is in a neighborhood that is enriched for "1"s which means it is one of at least three "1"s in a 4 character window. My desired output is a count of "1"s in an enriched area. For Example Input... (1 Reply)
Discussion started by: monstrousturtle
1 Replies

8. Shell Programming and Scripting

How do add values in a vector using a sliding window?

Greetings. I have a vector of numbers such as the following: 1 75 79 90 91 92 109 120 167 198 203 204 206 224 230 236 240 (4 Replies)
Discussion started by: Twinklefingers
4 Replies

9. Shell Programming and Scripting

Nested search in a file and replace the inner search

Hi Team, I am new to unix, please help me in this. I have a file named properties. The content of the file is : ##Mobile props east.url=https://qa.east.corp.com/prop/end west.url=https://qa.west.corp.com/prop/end south.url=https://qa.south.corp.com/prop/end... (2 Replies)
Discussion started by: tolearn
2 Replies

10. Programming

Xlib search window by his name

Hello, I just try to get the control of a Window by searching his name. I curretly do that : Window CMD::window_from_name_search(Display *display, Window current, char const *needle) { Window retval, root, parent, *children; unsigned children_count; char *name = NULL; ... (0 Replies)
Discussion started by: Loustak
0 Replies
match(1)                                                        Mail Avenger 0.8.3                                                        match(1)

NAME
match - Match strings against glob paterns SYNOPSIS
match [-gilrqs] [-n <n>] [-c cmd] [-x code] {[-p] pattern | -f <file>} str1 [str2 ...] DESCRIPTION
match checks strings against pattern, which should be a shell-like glob pattern. pattern may contain the following special characters: ? A "?" character in pattern matches any single character in the string, except that the "/" character is only matched if match was given the -s option. * A "*" character in pattern matches zero or more characters in the string. The exception is that it will only match "/" characters if match was given the -s option. [...] A set of characters between square brackets matches any character in the set. In addition, the "-" character can be used to specify a range. For example "[+e0-3]" would match any of the characters "+", "e", 0, 1, 2, or 3 in the input string. To include a hyphen ("-") in the set of characters matched, either include the hyphen first or last, or escape it with a "". [!...] A character class preceded by a "!" matches any character but those specified in the class. The exception is that the negated character class will match a "/" only if match was given the -s option. c The backslash character escapes the next character c. Thus, to match a literal "*", you would use the pattern "*". match prints each string that matches pattern, one per line, and exits 0 if one or more strings matched. If no string matches, match exits with status 67 (or whatever alternate status was specified by the -x flag). If the -n n flag was specified, match prints only the text that matched the nth occurrence of "*" in the patten. OPTIONS -f file Specifies that the pattern should be read from file. match will read each line of the file and consider it as pattern to match against the argument strings. For each argument string, match stops when it hits the first matching line of the file. If file does not exist, match exits 67, or whatever code was specified by -x. -g Normally, the -n option selects text matching particular "*" characters in the patern. -g changes this behavior to use parentheses for grouping. Thus, for instance, the text "foo.c" would match pattern "*(.[ch])", and the output with option -n 1 would be ".c". To include a literal "(" or ")" in the pattern with the -g option, you must precede the character with a "". -i Makes the match case insensitive. str will be considered to match if any variation on its capitalization would match. For example, string "G" would match pattern "[f-h]". -l When a pattern matches the string in more than one way, the -l flag says to assign as much text as possible to the leftmost "*"s in the pattern. For example, pattern "*+*" would match text "a+b+c", and the first "*" would match "a+b". This behavior is the default, thus -l's effect is only to undo a previous -r flag. -n n With this flag, match prints the text that matched the nth "*" in the pattern, as opposed to printing the whole string. The leftmost "*" corresponds to -n 1. Specifying -n 0 causes match to print the whole matching string. Specifying -n -1 or using a value greater than the number of "*"s in the pattern causes match not to print anything, in which case you can still use the exit status to see if there is a match. The default value for n is 0, unless -g has also been specified, in which case the default is 1. -c command When -c is specified, match runs command with the system shell (/bin/sh), giving it as argument $0 the full string that matched, and as arguments $1, $2, etc., the parts of the string that matched any "*"s in pattern. If the command does not exit with status 0, match will exit immediately, before processing further matches, with whatever status command returned. The -c and -n flags are mutually exclusive. -p pattern Specifies the pattern to match against. The -p flag is optional; you can specify pattern as the first argument following the options. However, if you want to try matching the same input string against multiple patterns, then you must specify each pattern with a -p flag. -q This option is synonymous with -n -1; it suppresses output when there is a match. You can still determine whether a match occurred by the exit status. -r When a pattern matches the string in more than one way, the -r flag says to assign as much text as possible to the rightmost "*"s in the pattern. For example, with -r, pattern "*+*" would match text "a+b+c" with the "*" matching "a", and the second matching "b+c". -s Ordinarily, "*", "?", and negated character classes ("[!...]") do not match "/" characters. -s changes this behavior to match slashes. -x code By default, when there is no match, match exits with status 67. With this option, match exits with status code, instead. EXAMPLES
Suppose you have a directory with a bunch of files ending .c and .o. If, for each file named foo.c you want to attempt to delete the file foo.o, you can run the following command: match -p '*.c' -c 'rm -f $1.o' *.c Servers running the mailman list manager often send mail from bounce addresses of the form listname-bounces@host.com. If you subscribe to multiple lists on the same server, the mailman interface makes it easier if you subscribe under the same address. To split the mail into multiple folders based on the bounce address in the environment variable SENDER, you might chose a mailbox with the following shell code: name=`match -n1 "*-bounces@host.com" "$SENDER"` && echo "$HOME/Mail/incoming/host-$name.spool" SEE ALSO
avenger(1), avenger.local(8) The Mail Avenger home page: <http://www.mailavenger.org/>. AUTHOR
David Mazieres Mail Avenger 0.8.3 2012-04-05 match(1)
All times are GMT -4. The time now is 02:33 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy