Advanced string replacements in python

Question

I have some strings in a python script, they look like

<tag1> thisIsSomeText <otherTag>, <tag1>!

I want to parse these lines, and replace every tag by a string from a dictionary.

Assume my dict looks like:

tag1: Hello
otherTag: Goodbye

Then the output line should look like:

Hello thisIsSomeText Goodbye, Hello!

As one can see, the tags (and its braces) are replaced. Multiple occurences are possible.

In C, I would search for '<', remember its position, search for the '>', do some ugly string manipulation... But i guess, Python has better solutions for this.

Maybe Regex? Well, I hope my task is simple enough that I could be solved without regex. But I have no plan how to start in Python. Any suggestions?

You want a [template engine](http://jinja.pocoo.org/) or [string formatting](https://docs.python.org/2/library/stdtypes.html#str.format). — ThiefMaster, Nov 10 '14 at 13:25

score 4 · Answer 1 · answered Nov 10 '14 at 13:19

4

re.sub accepts not only string, but also a function as a replacement as the second parameter. The function accepts a match object, and the return value of the function is used a replacement string.

>>> import re
>>> mapping = {'tag1': 'Hello', 'otherTag': 'Goodbye'}
>>> re.sub(r'<(\w+)>', lambda m: mapping[m.group(1)],
...        '<tag1> thisIsSomeText <otherTag>, <tag1>!')
'Hello thisIsSomeText Goodbye, Hello!'

answered Nov 10 '14 at 13:19

falsetru

357,413
63
732
636

@falsetrue: I guess your solution is better than the above. However, i couldn't get it working out of the box. So I ended up with the solution from greato, as this was working (after fixing the known issues).... As I'm writing this, I think i found the issue. Reassigning is also necessary, right? – lugge86 Nov 10 '14 at 15:03
@lugge86, Could you explain how my solution does not work? Was there any error? If it had, please share me the traceback. – falsetru Nov 10 '14 at 15:04
@lugge86, Yes, you need reassignment. – falsetru Nov 10 '14 at 15:04
@lugge86, My answer will raise an exception if there's a tag which is not defined in the dictionary. To leave such word as is: use `dict.get` instead of `dict[..]`: `lambda m: mapping.get(m.group(1), m.group(0))` or or `lambda m: mapping.get(m.group(1), m.group(1))` according to your need. – falsetru Nov 10 '14 at 15:06

score 0 · Answer 2 · answered Nov 10 '14 at 13:19

0

for i in dictionary.keys():
    string.replace('<'+i+'>',dictionary[i]]

answered Nov 10 '14 at 13:19

Ricky Han

1,309
1
15
25

1

You don't need to use `keys` method because itearting dictionary yields keys. You need to assign back the replace string (`str.replace` does not replace the string in-place, but returns a new string) – falsetru Nov 10 '14 at 13:21
If dictionary is like `{'a': 'b', 'b': 'a'}`, this could yield unwanted result. – falsetru Nov 10 '14 at 13:23
@falsetru: No, i think thats no problem. Because, after '' has been replaced by 'b', the braces are gone. So, when searching for '' there wont be any result. Correct me pls if I'm wrong. So far, this solution works for me! – lugge86 Nov 10 '14 at 14:18
@lugge86, What if the string is `<>`? And this solution has a syntax error. And as I commented, the return value of the `str.replace` should be assigned back. Otherwise, the string will be remain same. – falsetru Nov 10 '14 at 14:19
@falsetru: Yes, sure, there is a syntax error in this code, and yes, the string must be assigned back. I forgot to mention this in my comment, but i fixed this issues in my script. So, it works fine. Regarding the <> Problem: yes, this might be a problem. However, I'm using this script for generating C source and header files, using a template file and a excel sheet. As the template is maintained by me (or other developers around), the risk of encountering this issue is very low. Maybe I let the script search for << or >> and then yield a syntax error, or something. – lugge86 Nov 10 '14 at 14:49

Advanced string replacements in python

2 Answers2

Linked