Python regexp conundrum

Question

I am trying to break a huge SQL file into little sql file and I am using python to achieve this, but the code that I'm using doesn't match and from what I've seen on google it should.

Here is the code:

    import sys, re
    p = [0]
    f = open('/root/testsql/data.sql', 'r')
    tables =["tabel1", "table2"]
    contor = 0;
    con = 0;

    for line in f:
        for table in tables:
            stri = "root/testsql/" + str(con)
            con = con + 1
            stri2 = ".*" + table + ".*"
            if re.match(stri2,line):
                    print table
                    f2 = open(stri,"w")
                    f2.write(line)
                    f2.close()

If anybody has an idea why re.match doesn't work, it would be much appreciated.

The sql file is very long (73595 lines)and contains lines like:

insert into table ...
insert into table

What is the output when you write f2? Edit: or is that not happening at all? — danseery
– danseery, Commented Jun 22, 2012 at 15:43
I have no output, that is the problem. I have the print table in that if just to verify that i enter the if but there is no output, so it doesn't enter the loop — primero
– primero, Commented Jun 22, 2012 at 15:51
Copying the text you show for your lines and tables and then doing an re.match did lead to matchs. Are you sure the text you show is what you're getting from the file? — GreenMatt
– GreenMatt, Commented Jun 22, 2012 at 16:07
Serves me right for working over time, I can't even think straight — primero
– primero, Commented Jun 22, 2012 at 16:17

Tim Pietzcker · Accepted Answer · 2012-06-22 16:02:16Z

3

You're only looking for verbatim strings. In that case, regex is overkill. Instead, use in:

for line in f:
    for table in tables:
        # snip...
        if table in line:
            # ...

answered Jun 22, 2012 at 16:02

Tim Pietzcker

337k59 gold badges520 silver badges572 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

primero Over a year ago

I think the matching was not the problem. I used your suggestion, but still no result. No files were created and nothing was shown to the console output. The thing is there are 73595 lines and some of them are very long and maybe that is why it doesn't work.

primero Over a year ago

I tested from the python console and it worked with your suggestion, but still from the script it doesn't work.

Tim Pietzcker Over a year ago

@primero: You're right, my approach wouldn't have changed the results. But the length of the lines is certainly not the problem. I think you should insert some more print statements and check whether the program flow really is as expected. And maybe check your SQL file in an editor and do some searches there to make sure.

Tim Pietzcker Over a year ago

@primero: Hmm, grasping at straws here. What character encoding does your SQL file use? Not UTF-16 by any chance? What do you get when you insert a print line after for line in f:?

primero Over a year ago

Well I get the INSERT statements one by one separated by a new line

|

msanders · Accepted Answer · 2012-06-22 15:54:41Z

2

I think

stri2 = ".*" + table + ".*"

should be:

stri2 = ".*?" + table + ".*"

The .* is greedy and will match the whole line.

answered Jun 22, 2012 at 15:54

msanders

5,9491 gold badge33 silver badges31 bronze badges

4 Comments

Aprillion Over a year ago

it should be faster this way, but .* should backtrack in the first place

primero Over a year ago

Still not entering the if statement. I tried on the python console my match statement and it seemed to work, but i can't seem to enter the in the if from my script.

primero Over a year ago

And as you can see from the code in the if statement i create some files. None of them get created. I'm new to python an I don't know how to debug this.

Jirka Hanika Over a year ago

@primero - the info you provided in comments belongs to the body of the question.

Jirka Hanika · Accepted Answer · 2012-06-22 17:37:21Z

1

You should use re.search instead of re.match instead of wrapping the regex in .*.

The reason why you see no matches is that the the inputs end with a newline, and the dot metacharacter does not match newlines.

edited Jun 22, 2012 at 17:37

answered Jun 22, 2012 at 15:58

Jirka Hanika

13.7k3 gold badges52 silver badges80 bronze badges

5 Comments

primero Over a year ago

My first draft was with re.search, and i used re.compile to create my pattern but i don't know how to test in the if statement.

Aprillion Over a year ago

f2 is not a file, but a lot of files, 1 in each cycle, so w is ok in this case

primero Over a year ago

But now that i think of it it might be the same like in my example

Tim Pietzcker Over a year ago

Ad 1: True, but the regex should still have matched. Ad 2: He's not appending anything. He's writing lots of one-line files. Or trying to.

Jirka Hanika Over a year ago

@primero - re.search is used in exactly the same way as re.match except that it will fix your code. It returns the same kind of a match object.

Greg Smethells · Accepted Answer · 2014-08-19 20:35:44Z

0

I would use a raw string instead of a plain string in any regular expression so that you don't end up fooling yourself when a char get interpreted.

r'.*' + table + r'.*'

edited Aug 19, 2014 at 20:35

answered Jun 22, 2012 at 16:20

Greg Smethells

4506 silver badges20 bronze badges

Collectives™ on Stack Overflow

Python regexp conundrum

4 Answers 4

6 Comments

4 Comments

5 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

6 Comments

4 Comments

5 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related