Fixing open bracket problem in HTML string using Python

Question

I have the following text:

string = "<i>R</i> subspace  <i>{V.</i> generated by <i>{v<sub>1</sub>,...,v<sub>i</sub></i>, "

A careful reader might notice that there are two brackets missing. I was wondering, how this could be fixed using Python?

The expected output would be:

the <i>R</i>  subspace  <i>{V.}</i> generated by <i>{v<sub>1</sub>,...,v<sub>i</sub>}</i>,

One could:

Check: Is there a bracket after  ?
If yes -> Is there a bracket before  ?

How can I code this?

Edit

I have found this code, that can tell you if the brackets match or not.

What goes inside the brackets? Can "" appear inside brackets (i.e. does "{v" turn into "{v}" or "{v}")? — manveti, Apr 11 '19 at 21:16
@CraigMeier Good point. The second case would be the correct one. — henry, Apr 11 '19 at 21:19
@CraigMeier Would you know how to already code the simpler case (first case) ? — henry, Apr 11 '19 at 21:27
The first case can be handled by a regex (e.g. `re.sub` with a callable `repl` param that ensures there's a final '}' for any chunk with an initial '{'). But the second case falls under https://stackoverflow.com/a/1732454/ — manveti, Apr 11 '19 at 21:34
@CraigMeier Thanks for your input, but I don't know how to code it with `repl`. Sorry to bother you... I see that I can do: re.sub(r', SOMETHING, string) — henry, Apr 11 '19 at 21:45
Another great question. However, as with your previous HTML repair question, is there any significance to the `i` in ``, or should it work on arbitrary tags? — ggorlen, Apr 11 '19 at 22:02
how many times this function has to be called? What generated this string? why cant you fix it manually — aaaaa says reinstate Monica, Apr 11 '19 at 22:02
@ggorlen Thanks for your question: First if would be good to have it working for the `i `but of course, an extinction to arbitrary tags would be very useful for a much wider community ! — henry, Apr 11 '19 at 22:38

score 2 · Accepted Answer · answered Apr 12 '19 at 01:41

How about the following regex solution:

import re

string = "<i>R</i> subspace  <i>{V.</i> generated by <i>{v<sub>1</sub>,...,v<sub>i</sub></i>, "
expected = "<i>R</i> subspace  <i>{V.}</i> generated by <i>{v<sub>1</sub>,...,v<sub>i</sub>}</i>, "

fixed = re.sub(r"<(?P<tag>.*?)>({.*?)</(?P=tag)>", r"<\1>\2}</\1>", string)

print(fixed == expected) # True

The idea is to match a tag followed by a brace, find its closing tag, and plop the companion brace before the closing tag using the capture groups as <\1>\2}</\1>. Breakdown:

< # literal opening bracket
 (?P<tag> # open a named capture group
         .*? # lazily match any characters
            ) # end named capture group
             > # literal closing bracket
              ( # open capture group 2
               { # literal opening brace
                .*? # lazily match any characters
                   ) # end capture group 2
                    < # literal opening bracket
                     / # literal slash
                      (?P=tag) # backreference to the named group
                              > # literal closing bracket

If you just want , you can use re.sub(r"({.*?)", r"\1}", string).

Thank you so much for this perfect answer !! – henry Apr 12 '19 at 15:33 — henry, Apr 12 '19 at 15:33

Fixing open bracket problem in HTML string using Python

Edit

1 Answers1