Summary of AWK syntax under shell programming_PHP tutorial-PHP Tutorial-php.cn

The AWK utility comes with its own self-contained language and is one of the most powerful data processing engines available in Unix/Linux or any environment. The greatest power of this programming and data manipulation language (which takes its name from the first letters of the surnames of its founders, Alfred Aho, Peter Weinberger, and Brian Kernighan) depends on the knowledge one possesses. It allows you to create short programs that read input files, sort the data, process the data, perform calculations on the input, and generate reports, among countless other functions.
What is AWK? At its simplest, AWK is a programming language tool for working with text. The language of the AWK utility is similar in many ways to the shell programming language, although AWK has a syntax that is entirely its own. When AWK was originally created, it was intended for text processing, and the basis of the language was to execute a sequence of instructions whenever there was a pattern match in the input data. This utility scans each line in a file for patterns that match what is given on the command line. If a match is found, proceed to the next programming step. If no match is found, continue processing the next line. Although the operation can be complex, the syntax of the command is always: awk {pattern + action} where pattern represents what AWK looks for in the data, and action is a sequence of commands that are executed when a match is found. Curly braces ({}) do not need to appear all the time in a program, but they are used to group a sequence of instructions according to a specific pattern. Understanding Fields The utility divides each input row into records and fields. A record is a single line of input, and each record contains several fields. The default field delimiter is space or tab, and the record delimiter is newline. Although both tabs and spaces are treated as field separators by default (multiple spaces still act as one separator), you can change the separator from a space to any other character. For demonstration, view the following employee list file saved as emp_names: 46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL46019 BOGUE ROBERT PHOENIX AZ46021 JUNE MICAH P HOENIX AZ46022 KANE SHERYL UNKNOWN AR46024 WOOD WILLIAM MUNCIE IN46026 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN46029 TUTTLE BOB MUNCIE IN When AWK reads the input content, the entire record is assigned to the variable. Each field is separated by a field delimiter and assigned to variables , , etc. A row can contain essentially an infinite number of fields, each of which is accessed by its field number. Therefore, the print output that the command awk {print ,,,,} names will produce is 46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL46019 BOGUE ROBERT PHOENIX AZ46021 JUNE MICAH PHOENIX AZ46022 KANE SHERYL UNKNOWN AR46024 WOOD WILLIAM MUNCIE IN46026 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN46029 TUTTLE BOB MUNCIE IN One important thing to note is that AWK interprets five fields separated by spaces, but when it prints the display, between each field There is only one space. With the ability to assign a unique number to each field, you can choose to print only specific fields. For example, to print only the names of each record, just select the second and third fields to print: $ awk {print ,} emp_namesDULANEY EVANDURHAM JEFFSTEEN BILLFELDMAN EVANSWIM STEVEBOGUE ROBERTJUNE MICAHKANE SHERYLWOOD WILLIAMFERGUS SARAHBUCK SARAHTUTTLE BOB$ You can also specify the by Print fields in any order, regardless of how they exist in the record. So, just display the name field, and reverse the order, displaying first name first and then last name: $ awk {print ,} emp_namesEVAN DULANEYJEFF DURHAMBILL STEENEVAN FELDMANSTEVE SWIMROBERT BOGUEMICAH JUNESHERYL KANEWILLIAM WOODSARAH FERGUSSARAH BUCKBOB TUTTLE$ Use patterns by including a pattern that must match , you can choose to operate only on specific records instead of all records. The simplest form of pattern matching is a search, where the item to match is enclosed in slashes (/pattern/). For example, to perform the previous operation only for those employees who live in Alabama: $ awk /AL/ {print ,} emp_namesEVAN DULANEYJEFF DURHAMBILL STEENEVAN FELDMANSTEVE SWIM$ If you do not specify the fields to print, the entire matching entry will be printed : $ awk /AL/ emp_names46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL$ Multiple commands for the same data set can be separated by semicolons (;). For example, to print the name on one line and the city and state on another: $ awk /AL/ {print , ; print ,} emp_namesEVAN DULANEYMOBILE ALJEFF DURHAMMOBILE ALBILL STEENMOBILE ALEVAN FELDMANMOBILE ALSTEVE SWIMUNKNOWN AL$ If no semicolon is used (print ,,,) will display everything on the same line.On the other hand, if the two print statements are given separately, it will produce completely different results: $ awk /AL/ {print ,} {print ,} emp_namesEVAN DULANEYMOBILE ALJEFF DURHAMMOBILE ALBILL STEENMOBILE ALEVAN FELDMANMOBILE ALSTEVE SWIMUNKNOWN ALPHOENIX AZPHOENIX AZUNKNOWN ARMUNCIE INMUNCIE INMUNCIE INMUNCIE IN$ will only give fields three and two if AL is found in the list. However, fields four and five are unconditional and they are always printed. Only the command in the first set of curly braces has an effect on the command immediately preceding it (/AL/). The result is very unreadable and could be made slightly clearer. First, insert a space and comma between the city and state. Then, place a blank line after every two lines displayed: $ awk /AL/ {print , ; print ", ""n"} emp_namesEVAN DULANEYMOBILE, ALJEFF DURHAMMOBILE, ALBILL STEENMOBILE, ALEVAN FELDMANMOBILE, ALSTEVE SWIMUNKNOWN, AL$ on the fourth and the fifth field, add a comma and a space (between the quotes), and after the fifth field, print a newline character (n). All the special characters that can be used in the echo command can also be used in the AWK print statement, including: n (newline) t (tab) b (backspace) f (feed) r (carriage return) Therefore, to read To take all five fields initially separated by tabs and print them using tabs too, you can program as follows $ awk {print "t""t""t""t"} emp_names46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL46019 BOGUE ROBERT PHOENIX AZ46021 JUNE MICAH PHOENIX AZ46022 KANE SHERYL UNKNOWN AR46024 WOOD WILLIAM MUNCIE IN460 26 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN46029 TUTTLE BOB MUNCIE IN$ by setting multiple items consecutively Standard and separated by pipe (|) symbols, you can search for multiple pattern matches at once: $ awk /AL|IN/ emp_names46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL46024 WOOD WILLIAM MUNCIE IN46026 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN46029 TUTTLE BOB MUNCIE IN$ This will find matching records for every resident of Alabama and Indiana. But while trying to find out who lives in Arizona, a problem arises: $ awk /AR/ emp_names46019 BOGUE ROBERT PHOENIX AZ46021 JUNE MICAH PHOENIX AZ46022 KANE SHERYL UNKNOWN AZ46026 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN$Employees 46026 and 4 6027 No Live in Arizona; but their names contain the sequence of characters being searched for. Keep in mind that when doing pattern matching in AWK, such as grep, sed, or most other Linux/Unix commands, a match will be found anywhere in the record (line) unless otherwise specified. To solve this problem, the search must be tied to a specific field. This is accomplished by utilizing a tilde (?) along with a description of a specific field, as shown in the following example: $ awk ? /AR/ emp_names46019 BOGUE ROBERT PHOENIX AZ46021 JUNE MICAH PHOENIX AZ46022 KANE SHERYL UNKNOWN AZ$ tilde (indicates a match ) is a tilde (!?) preceded by an exclamation point. These characters tell the program to find all rows that match the search sequence if it does not appear in the specified field: $ awk !? /AR/ names46012 DULANEY EVAN MOBILE AL46013 DURHAM JEFF MOBILE AL46015 STEEN BILL MOBILE AL46017 FELDMAN EVAN MOBILE AL46018 SWIM STEVE UNKNOWN AL46024 WOOD WILLIAM MUNCIE IN46026 FERGUS SARAH MUNCIE IN46027 BUCK SARAH MUNCIE IN46029 TUTTLE BOB MUNCIE IN$ In this case, all rows without an AR in the fifth field will be displayed—including the two Sarah entries, both of which The entry does contain an AR, but in the third field instead of the fifth. Braces and field delimiters The brace characters play an important role in AWK commands. Actions that appear between parentheses indicate what is going to happen and when. When using only one pair of brackets: {print,} all operations between the brackets occur simultaneously. When using more than one pair of parentheses: {print }{print } executes the first set of commands, and after that command completes, executes the second set of commands. Note the difference between the following two lists: $ awk {print ,} namesEVAN DULANEYJEFF DURHAMBILL STEENEVAN FELDMANSTEVE SWIMROBERT BOGUEMICAH JUNESHERYL KANEWILLIAM WOODSARAH FERGUSSARAH BUCKBOB TUTTLE$$ awk {print }{print } namesEVANDULANEYJEFFDURHAMBILLSTEENEVANFELDMANSTEVESWIMROBER TBOGUEMICAHJUNESHERYLKANEWILLIAMWOODSARAHFERGUSSARAHBUCKBOBTUTTLE$ To use multiple sets of brackets to perform repeated searches, execute the first The commands in the group are processed until completion; then the second group of commands is processed. If there is a third set of commands, it is executed after the second set of commands completes, and so on. In the resulting printout, there are two separate print commands, so the first command is executed first, followed by the second command, causing each entry to appear on two lines instead of one. The field separator that distinguishes two fields does not always have to be a space; it can be any recognized character.For demonstration purposes, assume that the emp_names file uses colons instead of tabs to separate fields: $ cat emp_names46012:DULANEY:EVAN:MOBILE:AL46013:DURHAM:JEFF:MOBILE:AL46015:STEEN:BILL:MOBILE:AL46017:FELDMAN:EVAN: MOBILE:AL46018:SWIM:STEEVE:UNKNOWN:AL46019:BOGUE:ROBERT:PHOENIX:AZ46021:JUNE:MICAH:PHOENIX:AZ46022:KANE:SHERYL:UNKNOWN:AR46024:WOOD:WILLIAM:MUNCIE:IN4602