Shulou(Shulou.com)06/02 Report--

What is the search command under Linux? I believe many inexperienced people are at a loss about it. Therefore, this article summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

Search is what we do every day, wake up in the morning to find the phone, check the bus before going out, sit down and check the data, analyze the data and check the pattern.

Finding files, finding information, and finding errors are some of the more specific tasks that can be applied, while the Linux command behavior provides a lot of quick and powerful ways to find.

Command / executable program lookup

Whereis program_name: look for binaries, source code, and files that contain given query keywords in the system's default installation directory (usually the software installed by default when you have root permissions). (the default directory has paths such as / bin, / sbin, / usr/bin, / usr/lib, / usr/local/man, etc.)

Which program_name: the path of all programs in the environment variable is given to make it easy to know where the program is running and to modify it. (PATH and path, silly and confused)

For example, vim which sp_pheatmap.sh can directly modify the heat map drawing script, cpwhich sp_pheatmap.sh. You can copy the source code directly to the current directory, saving the trouble of writing the full path.

If you run which bwa and the system returns / usr/bin/which: no bwa in (/ home/usr/bin:/bin), it means that bwa is not placed in the environment variable and cannot be called by name directly.

Quick location of locate for ordinary files

Locate is a good way to find location files quickly, but it depends on the index built by updatedb. Updatedb is usually run once a day, so new files for the day cannot be indexed.

If you have root privileges, you can manually run updatedb to do an update, and then locate bwa.

(individual users can also build their own updatedb and use locate to find it in the local environment.)

General file multi-conditional search find

Find /-name bwa can search all files named bwa in the root directory

Running the above command will output a lot of Permission denied, because as an ordinary user, you do not have permission to access some directories, so you will be prompted for output. You can use find /-name bwa 2 > / dev/null to redirect the standard error to the empty device, and the error message will be thrown away without affecting the normal output.

Search by time

The online drawing website (www.ehbio.com/ImageGP) we developed needs to check the log regularly every day in order to track what problems users encounter when using it every day.

This command find. -name * .log-mmine-60 can view log files modified within one hour under the current directory (including all subdirectories). With head, you can view the contents of each log file to see what errors occurred during use, and how to add prompts or modify the drawing program.

It is with this sharp weapon that there is such a sentence in the error prompt at the front desk. If you check the data and parameters, please try again one day later. If there is a problem with the program, we usually fix it within 1 day.

Of course, the background data is stored with a timestamp, and if no errors are reported, the data will be deleted directly, and only those who report errors will keep the log and will not disclose user information.

Now the drawing website is becoming more and more stable, there are fewer and fewer problems, and the front desk tips are becoming more and more perfect. I hope you can read more tips when you use them, and the frequency of viewing logs is also less, so use find. -name * .log-mtime-1 check the log 24 hours from now.

There is also a problem with this. The time of each check may be inconsistent, omitted or overlapped, so after viewing the log, use touch check to create a new empty file in the current directory. When you check the log file later, just use find. -name * .log-newer check can get all the new logs since they were last viewed. It is much more convenient to make a bookmark after each check.

Slowly discover that log files are available and use find. -name * .log-newer check-size + 0 is filtered out, leaving only files of size greater than 0. In this way, under the smart and diligent maintenance of our friends, our drawing website has provided nearly 300000 services for you in the past 2 years. (drawing manual | ImageGP: have you "plot" today?

Find by type and size

What if I want to get all the png and jpg photos in the current directory?

Use find. \ (- name "* .png"-o-name "* .jpg"\) | less

Or find. -regex ". *\ (\ .png\ |\ .jpg\) $"

Find. -type f-size + 100g can get files larger than 100G.

Limit the depth of search

Only look at the files in the current directory layer 2 subdirectory find. -maxdepth 2-name * .log.

View the file find. Exe that is not at the end of log. -not-name * .log. There are more combination operations, as detailed in the find documentation (I threw a question about finding in the group, and as a result... ).

Find grep by file content

Can find find a file that contains a sentence? Let's talk about our journal, find. -name * .log-exec grep-l 'Error' {}\; can return all file names that contain the word Error.

Find. -name * .log | xargs grep-l 'Error' is also possible.

Grep-rl 'Error' * is also fine, and matching rows can be returned without adding-l.

Match the front and back lines of a line

Grep-A 5-B 1 'Bioinfo' ehbio.log can view the first 1 line (B, before) and the last 5 lines (A, after) of the matching row.

Matching times

Grep-c 'Bioinfo' ehbio.log can count the number of rows containing Bioinfo

Grep-ci 'Bioinfo' ehbio.log ignores case when matching.

Count the number of sequences grep'^ > 'ehbio.fa in FASTA sequences

Count the number of sequences grep'^ + $'ehbio.fq in FASTQ sequences. (^ indicates what to start with and $means what to end with).

Get unmatched rows

Grep-v 'Bioinfo' ehbio.log, read the manual (man grep), you can see more parameters used.

Sequence extraction

Suppose you have a gene list file (ID) and a single-line sequence FASTA file (ehbio.fa). Run the command grep-A1-Fw-f id ehbio.fa | grep-v--'-'to extract the sequence in batches.

-f id means to treat each line in the id file as a matching pattern. -F means to match the pattern as a raw string rather than a regular expression, in case special characters are parsed. -w means to match as a word, that is, if there is Sox2 in id, it will match Sox2, and it will also match Sox21;. If you add-w, it will not match Sox21.

For better batch extraction of sequences, see the use of awk.

Pattern matching

The powerful feature of grep is to support regular matching, using basic regular expressions by default,-E using extended regular expressions, and-P using regular expressions in perl format.

For example, you want to remove all the blank lines in the file grep-v'^ $'ehbio.fa > ehbio.clean.fa.

Search for grep 'writ.*' *. Md related to article writing from official account articles (can match write, writing, etc.)

What is Linux system Linux is a free-to-use and free-spread UNIX-like operating system, is a POSIX-based multi-user, multi-task, multi-threaded and multi-CPU operating system, using Linux can run major Unix tools, applications and network protocols.

After reading the above, have you mastered what the search command under Linux is? If you want to learn more skills or want to know more about it, you are welcome to follow the industry information channel, thank you for reading!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.