Question

I need to extract in Notepad++ from a package body file the object name (function, procedures, etc) and the description from the object.

I know that is possible by regex but sometimes the comments are compound for more than 3 lines.

CREATE OR REPLACE PACKAGE BODY pac_example AS

/* *********************************************
*  Function f1
*  Description: Search for a data in table1
*  (another comments)
*  ******************************************* */
FUNCTION f1 RETURN NUMBER IS
BEGIN
SELECT * FROM table1;
RETURN 1;
END f1;



/* *********************************************
*  Function f2
*  Description: Search for a data in table2
*  (another comments)
*  (another comments)
*  (another comments)
*  ******************************************* */
FUNCTION f2 RETURN NUMBER IS
BEGIN
  SELECT * FROM table2;
RETURN 1;
END f2;

END pac_example;

And in this case, I need only to replace all in document and get something like:

/* *********************************************
*  Function f1
*  Description: Search for a data in table1
*  (another comments)
*  ******************************************* */
FUNCTION f1 RETURN NUMBER IS
/* *********************************************
*  Function f2
*  Description: Search for a data in table2
*  (another comments)
*  (another comments)
*  (another comments)
*  ******************************************* */
FUNCTION f2 RETURN NUMBER IS

Or (best scenario) this:

FUNCTION f1 Search for a data in table1
FUNCTION f2 Search for a data in table2

Answer 1

Ctrl + H
查找内容：(?:\A.*?|\G)\*\h+((?:Function|Procedure)\h*\w+).*?Description:\h*([^\r\n]+\R)(?:(?!\*\h+(?:Function|Procedure)).)+
替换为：$1 $2
UNcheck区分大小写
检查环绕
检查正则表达式
检查. matches newline
全部替换

说明：

(?:                         # non capture group
    \A                      # beginning of file
    .*?                     # 0 or more any character
  |                         # OR
    \G                      # restart from last match position
)                           # end group
\*                          # an asterisk
\h+                         # 1 or more horizontal spaces
(                           # start group 1
    (?:Function|Procedure)  # literally Function OR Procedure
    \h+                     # 1 or more horizontal spaces
    \w+                     # 1 or more word character
)                           # end group 1
.*?                         # 1 or more any character, not greedy
Description:\h*             # literally Description followed by horizontal spaces
(                           # start group 2
    [^\r\n]+                # 1 or more any character not linebreak
    \R                      # any kind of linebreak
)                           # end group 2
(?:                         # Tempered greedy token
    (?!                     # negative lookahead
        \*                  # an asterisk
        \h+                 # 1 or more horizontal spaces
        (?:Function|Procedure)  # literally Function OR Procedure
    )                       # end lookahead
    .                       # any character
)+                          # end group, appears 1 or more times

替换：

$1          # content of group 1, function or procedure
$2          # content of group 2, description

屏幕截图：

Answer 2

To do this in Notepad++ I would:

Regex for ^([ \t\*]+)(FUNCTION|PROCEDURE)([ \t]+)([A-Z_0-9]+). Choose on the Mark tab and click Mark all.
Now search for ^([ \t\*]+)Description: (.+). Again click Mark all.
Menu Search > Bookmark > Copy Bookmarked Lines
Paste into a new document.

This will give you the basis for best scenario. It's a simple matter to tidy the lines from there using regex replace.

Notes:

The first step selects the procedure name from the comments i.e. * Function f2. You close the capitalisation but at least it's in the right order. Otherwise you'd need to switch the procedure name and description for every line.
This is no good if you need signature details (arguments, etc). Then you'll need to search for ^([ \t]*)(FUNCTION|PROCEDURE)([ \t]+)([A-Z_0-9]+) and re-order the descriptions
I've included tab and space patterns, because it's not clear how your source is indented

Get function and description comment in a pl/sql body package file

2 个答案: