Captions for video - Find and Replace across time stamps

MaximillianM

@Alan-Kilborn Thanks! Yes, I like the idea of making it a separate file for ease of updates.

I experimented with the code.

What is the difference between using the ! or $ for the word in the list? They seem to do a similar find/replace in the example, at least in my small test. But I’m probably missing something.

The next step seems to be adding the search over multiple lines code as from your earlier example. How can I do that?

Thanks!

Alan Kilborn

@MaximillianM said in Captions for video - Find and Replace across time stamps:

What is the difference between using the ! or $ for the word in the list? They seem to do a similar find/replace in the example, at least in my small test. But I’m probably missing something.

So the “delimiter” variability is useful if the data itself contains the delimiter.

Say we hardcoded the delimiter to be a colon (:).
Then if you wanted to replace something like a:b with c:d it would be difficult.

The way I defined it, you could just use a different delimiter for this case, e.g. !a:b!c:d.

Alan Kilborn

@MaximillianM said in Captions for video - Find and Replace across time stamps:

The next step seems to be adding the search over multiple lines code as from your earlier example. How can I do that?

This is the point where I’m having trouble envisioning how it would work.
I know you said something about it before, but I didn’t quite understand it.

Would you put a special symbol in the replacement part that you’d want the timestamp to be replaced by?
Maybe a more in-depth walk-through (example(s)) of what is wanted?

I’m certainly willing to do it, or at least help you get started…

MaximillianM

@Alan-Kilborn Thanks again. I see you are helping many other people so I should have put a summary in to help you :-)

The problem
1-Simple Find and replace with a list of words (your most recent code does this)
2-Find and replace multi-word string that goes across a timestamp
Find “like a fire” replace with $2liquefy

0:00:17.680,0:00:20.400
vaporize like a

0:00:19.840,0:00:22.400
fire

I would like to combine the most recent code with this one (minus the manual entry box so it could use the list as in your most recent code) to search across the time stamp.

-- coding: utf-8 --

from Npp import editor, notepad

class T1(object):

def __init__(self):
    search_phrase = 'like a fire'
    while True:
        search_phrase = notepad.prompt('\r\nEnter search phrase and press OK to find next:', '', search_phrase)
        if search_phrase == None or len(search_phrase) == 0: return  # quit
        word_list = search_phrase.strip().split()
        regex = r'(?-is)(?(DEFINE)(\x20|\R|\R*\d{1,2}:.*\R))' + '(?1)'.join(word_list)
        matches = []
        editor.research(regex, lambda m: matches.append(m.span(0)), 0, editor.getCurrentPos(), editor.getLength(), 1)
        if len(matches) == 0:
            notepad.messageBox('No (more) matches', '')
            return
        else:
            (match_start, match_end) = matches[0]
            editor.scrollRange(match_end, match_start)
            editor.setSelection(match_end, match_start)

if name == ‘main’: T1()

Thanks again :-)

Alan Kilborn

@MaximillianM said in Captions for video - Find and Replace across time stamps:

Find “like a fire” replace with $2liquefy
0:00:17.680,0:00:20.400
vaporize like a
0:00:19.840,0:00:22.400
fire

Yes, you used this example before, but I didn’t fully understand it.
I take it the $2 represents the bridged timestamp, and where it would appear in the replacement text.

BTW, why $2?
Is it because a search match could possibly bridge two timestamps?
And $1 might possibly appear in the replace expression as well?
Or $3 etc?

MaximillianM

@Alan-Kilborn Hi, I’m a beginner and was just using the $2 that astrosofista suggested in this post so I don’t fully understand it.

He suggested
Search: (?x-s) \x20like (\x20 | \R | \R* \d{1,2} : .*\R) a ((?1)) fire
Replace: $2liquefy

If I don’t use the $2 before the replacement word then the timestamp is removed in the replacement. I tried $1 and $3 before the replacement word as a test and the time stamp was removed in both cases.

Putting the replacement expression in the second part of the string (second timestamp) is preferable as it is the most likely scenario.

Just one bridged time-stamp is the basic requirement, in the future I might look at across multiple time-stamps.

There a blank line between the timestamp/phrase as in the example below.
$1, $2, $3, would not be present in the text so ok to use in our expression.

Find “like a fire” replace (end of expression) with liquefy

0:00:17.680,0:00:20.400
vaporize like a

0:00:19.840,0:00:22.400
fire

0:00:22.400,0:00:24.300
next phrase

Thanks :-)

Alan Kilborn

@MaximillianM said in Captions for video - Find and Replace across time stamps:

I’m a beginner and was just using the $2 that astrosofista suggested in this post so I don’t fully understand it.

I’m not a beginner, but I don’t see how this is going to work in the bigger scheme of things. I mean, well, maybe I see can see it if I squint at it, but I don’t have the desire/time to sort out the regexes needed down to the Nth level so that every situation is covered.

I think finding the matches is one level of difficulty (which has already been conquered), but replacing them introduces a whole new level of complexity to it. Even the single bridged timestamp can be nuancy when you really think about some examples that a generic replace could encounter.

I’m sorry if I misrepresented that I would do the whole solution for your “list” based replacement. My intent was to demo a few things to show what’s possible with scripting, not come up with a full-blown solution for some very specific data.

If someone else (@guy038 loves to do this sort of thing, or maybe @PeterJones since he got the original ball rolling) is willing to do it, I can certainly help put together the final script using the information. What is needed is a find/replace regex pair that would walk through a document doing the replacements desired.

guy038

Hello, @maximillianm, @peterjones, @alan-kilborn, @astrosofista and All,

@maximillianm, I assume that the timestamp always begins lines of your file, without any leading blank characters ! If it’s not the case, just tell me !

Here is a generic regex which searches any range of text, containing one timestamp feature, and replace it with any range of text, still containing the same timestamp

SEARCH (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F)|(?-i)Before_Find_Text((?1))After_Find_Text

REPLACE Before_Replace_Text\2After_Replace_Text

where :

Before_Find_Text represents the text to search, located BEFORE the time-stamp line
After_Find_Text represents the text to search, located AFTER the time-stamp line
Before_Replace_Text represents the text to replace BEFORE the time-stamp line
After_Replace_Text represents the text to replace AFTER the time-stamp line

First example :

Given your initial text :

0:00:17.680,0:00:20.400
vaporize like a

0:00:19.840,0:00:22.400
fire

And the expected result :

0:00:17.680,0:00:20.400
vaporize

0:00:19.840,0:00:22.400
liquefy

The different variable parts of the generic regex S/R are :

Before_Find_Text = vaporize like a
After_Find_Text = fire
Before_Replace_Text = vaporize
After_Replace_Text = liquefy

which gives the functional regex S/R :

SEARCH (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F)|(?-i)vaporize like a((?1))fire

REPLACE vaporize\2liquefy

Second example :

Given this initial example, taken from my previous post :

0:00:17.680,0:00:20.400
The licenses for most software are designed to

0:00:19.840,0:00:22.400
take away your freedom to share and change it.

And the expected result :

0:00:17.680,0:00:20.400
The licenses for most software are generally made to

0:00:19.840,0:00:22.400
always suppress your freedom to share and change it.

The different variable parts of the generic regex S/R are, this time :

Before_Find_Text = designed to
After_Find_Text = take away
Before_Replace_Text = generally made to
After_Replace_Text = always suppress

which gives the functional regex S/R :

SEARCH (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F)|(?-i)designed to((?1))take away

REPLACE generally made to\2always suppress

Third example :

Given, again, this initial example, taken from my previous post :

0:00:17.680,0:00:20.400
The licenses for most software are designed to

0:00:19.840,0:00:22.400
take away your freedom to share and change it.

And the expected result :

0:00:17.680,0:00:20.400
The licenses for most software are

0:00:19.840,0:00:22.400
designed to suppress your freedom.

The different variable parts of the generic regex S/R are, this time :

Before_Find_Text = are designed to
After_Find_Text = take away your freedom to share and change it
Before_Replace_Text = are
After_Replace_Text = designed to suppress your freedom

which gives the functional regex S/R :

SEARCH (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F)|(?-i)are designed to((?1))take away your freedom to share and change it

REPLACE are\2designed to suppress your freedom

Fourth example :

Given this initial example :

0:00:17.680,0:00:20.400
The licenses for most software are designed to

0:00:19.840,0:00:22.400
take away your freedom to share and change it.

And the expected result :

0:00:17.680,0:00:20.400
The licenses for most software

0:00:19.840,0:00:22.400
prevent you from sharing and changing it.

The different variable parts of the generic regex S/R are, this time :

Before_Find_Text = software are designed to
After_Find_Text = take away your freedom to share and change
Before_Replace_Text = software
After_Replace_Text = prevent you from sharing and changing

which gives the functional regex S/R :

SEARCH (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F)|(?-i)software are designed to((?1))take away your freedom to share and change

REPLACE software\2prevent you from sharing and changing

Notes :

The first alternative of this search regex (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R)(*F) is never matched, due to backtracking control verb (*F) which forces a failure of the match attempt.
However, the regex (\R+\d{1,2}:\d\d:\d\d\.\d{3},\d{1,2}:\d\d:\d\d\.\d{3}\R), which would match any range of line-breaks, followed with a complete timestamp line, is stored in group1 for later use, in the second alternative of the search regex
As you can see, the timestamp 0:00:19.840,0:00:22.400 is kept, after replacement because it’s stored in group 2 ( Current timestamp value of the subroutine call (?1), in the regex part ((?1)) ! )
If you prefer an “insensitive to case” search, simply change the part (?-i) by (?i)

Best regards

guy038

Alan Kilborn

@guy038 said:

Before_Find_Text = vaporize like a
After_Find_Text = fire

Before_Find_Text = designed to
After_Find_Text = take away

Before_Find_Text = are designed to
After_Find_Text = take away your freedom to share and change it

Before_Find_Text = software are designed to
After_Find_Text = take away your freedom to share and change

But the OP wants to simply specify the following for each of those searches:

vaporize like a fire
designed to take away
are designed to take away your freedom to share and change it
software are designed to take away your freedom to share and change

In other words, transparency of the timestamp and where it occurs.
Thus, timestamp could occur at any one of the following points (denoted by TS):

vaporizeTSlikeTSaTSfire
designedTStoTStakeTSaway
areTSdesignedTStoTStakeTSawayTSyourTSfreedomTStoTSshareTSandTSchangeTSit
softwareTSareTSdesignedTStoTStakeTSawayTSyourTSfreedomTStoTSshareTSandTSchange

And, this problem has already been solved, by Peter, way above.
What is needed now is a replacement regex that works, for all cases of a generic substitution, to accompany the original regex scheme.

Now, ok, it is fine if the orginal search regex mutates somewhat to meet this need, but the “spirit” of it needs to be retained.

And, it may be simpler than I think it is, truly. But what I don’t want to have happen is the usual – people put a lot of time into it, and then the reality of it is that a different problem was solved than what was wanted.

PeterJones

@Alan-Kilborn said in Captions for video - Find and Replace across time stamps:

What is needed now is a replacement regex that works, for all cases of a generic substitution, to accompany the original regex scheme.

So, the problem with the expression developed by me and @astrosofista , if I’ve understood my catching up with this discussion, is that when you have the “vaporizeTSlikeTSaTSfire”, the last of the timestamps is in group 3 instead of group 2. Or if there are more timestamps, the one you want to keep is always only the final timestamp. (I assume always the final timestamp, because that makes the most sense to me from the way that the OP originally phrased it)

If I understand what you’re doing in your PythonScript, you are just building the regex based on joining the elements of word_list with (?1) (which doesn’t create that final capture group that @astrosofista added to make the replacement work better)

Could you instead join all elements except the last with (?1) and then manually append ((?1)) and word_list[n-1] to the end: '(?1)'.join(word_list[0:n-1]) + '((?1))' + word_list[n-1] – this would then have the group2 always be the final space or timestamp.

That would have worked, except I realized that if our phrase were in the subtitle file as “first twoTSmiddleTSlast two”, so it spans multiple timestamps, but might have multiple words per timestamp, then my put-the-group2-around-the-last-backreference wouldn’t work, because the backreference matches either spaces or the timestamps, so it would throw away the final timestamp and just capture the final space, which isn’t what we want.

At this point, I’d be more tempted to build a regex that captured each of the spaces or TS ('((?1))'.join(word_list)), and use editor.rereplace with a callback rather than a literal replace. That way, in the callback, you can look at each element from m.group and use a space for all the separators except the very last timestamp, which should be kept. I don’t have the time right now to implement that… but I think that’s the direction I’d go at this point.

(Doing a search-and-replace where the match and replacement can go across multiple timestamps is a convoluted mess.)

Alan Kilborn

@PeterJones said in Captions for video - Find and Replace across time stamps:

(Doing a search-and-replace where the match and replacement can go across multiple timestamps is a convoluted mess.)

So, yes, that’s a convoluted mess.

But also there can be the situation where (potentially) all text connected to a timestamp is removed (e.g. a replace-with-nothing situation), and then an orphan timestamp is left sitting there in the file.

This could (logically) happen if the timestamp to be orphaned occurs ahead of the match data, or after it.

Of course, OP said nothing about “replace with nothing”, but in trying to make this a “generic” operation, that’s certainly a possibility with a “replace” algorithm.

I dislike reading too much into an OP’s need, but if this is not done we can end up, after some effort, with the aforementioned “solution is not what I wanted” problem.

I got thinking about this as I was playing around with some of the data originally postulated by Peter, above.

MaximillianM

@Alan-Kilborn @guy038 Thank you so much for all your responses and I see how there are multiple ways to accomplish this and how each solution solves one situation and might mess up another solution. So trying to solve all options get complicated fast.

You both have gone above and beyond and I really appreciate it. :-)

What I think would be most helpful now is creating a script using a list such as Alan suggested a bit back and then parse the list so I could use it in first a simple search and replace (word(s) on the same line) then one of the more complicated ones (words over multiple lines) like Alan or Guy suggested.

The first part was solved by Alan

Alan’s script
the_list = [
‘:findable_you:replaceable_you’,
‘:I can contain spaces:So I see’,
‘:look_for_me:really_want_to_be_you’,
‘!simple!complex’,
‘$fire$liquify’,
]

    editor.beginUndoAction()
    for definition in the_list:
        delim = definition[0]
        (find_what, repl_with) = definition[1:].split(delim, 2)
        editor.replace(find_what, repl_with)
    editor.endUndoAction()

Then a way that
I could parse the list (or a separate list might be easier) so the scripted search/replace could use what is the most common situation

the_list = [
‘:like a fire:vaporize’,

Search: (?x-s) \x20like (\x20 | \R | \R* \d{1,2} : .*\R) a ((?1)) fire
REPLACE vaporize on second line

With that my original problem would be solved and then I could experiment with Guy’s regex for the next scenario.

Thanks again. You really know your stuff and I so appreciate your support!

Alan Kilborn

@MaximillianM

In the spirit of following thru on something I’ve signed up for, I’ll put together something very basic on this, and then it can be evaluated by you and others. After that I’ll have to hand it off, and absolve myself of future development. :-) It may take me a bit of time, not because of the task itself but because of other demands currently biting at me --check back here periodically.

Alan Kilborn

So in attempting to get something to work, I’ve come up with the following regular expression replacement (generated via code):

find: (?(DEFINE)(?<TS>\d{1,3}:\d{2}:\d{2}\.\d{3},\d{1,3}:\d{2}:\d{2}\.\d{3}))like(?<RE>\x20|\R|\R*(?<TSLINE>(?P>TS))\R)a(?P>RE)fire
repl: it so that it will \r\n\r\n$+{TSLINE}\r\nliquify

And running it on this data:

0:00:17.680,0:00:20.400
vaporize like a

0:00:19.840,0:00:22.400
fire

0:00:17.680,0:00:20.400
vaporize like a

0:00:19.840,0:00:22.400
fire

Should produce this result:

0:00:17.680,0:00:20.400
vaporize it so that it will 

0:00:19.840,0:00:22.400
liquify

0:00:17.680,0:00:20.400
vaporize it so that it will 

0:00:19.840,0:00:22.400
liquify

But it is ignoring my use of $+{TSLINE} in the replace expression and produces the following:

0:00:17.680,0:00:20.400
vaporize it so that it will


liquefy

0:00:17.680,0:00:20.400
vaporize it so that it will


liquefy

It is working fine in the Boost emulation mode for RegexBuddy, just not in Notepad++.

Any ideas?

I’m also open to someone changing the regexes I’m using.

Given that this is a general problem with a complicated regex situation between every word in a multiple word search string, I thought it better to switch to named groups instead of numbered groups, but as of yet I don’t feel confident with any approach.