Synchronzied scrolling with timestamp

Basti Klein

Hi,
I’ve got two log files. Each line has a timestamp, both using the same format. Can I configure the synchronized scrolling to sync with the according timestamp? Is there an Addon?
Example of line:
2023/01/19 10:44:58.362 {STRING}

PeterJones

@Basti-Klein said in Synchronzied scrolling with timestamp:

Can I configure the synchronized scrolling to sync with the according timestamp?

No. Wherever the two views are showing when you select View > Synchronize Vertical Scrolling will become the point at which the two files are synchronized.

Is there an Addon?

There is not a plugin (that I know of) whose primary purpose is synchronizing two files based on timestamp (or based on a regex, or some such). Though depending on your needs, ComparePlus (which has an internal algorithm to try to align partially-matching files) might be able to sync things up the way you want, depending on how the other lines in your data compare. If I were you, that would be my first try.

Alan Kilborn

@Basti-Klein

Not at all an answer to your direct question, but in the spirit of possibly “getting your task done”…

You might try a script conversion on your data that changes the leading timestamp on a line to a number of epoch milliseconds. Then for file “A” append an A to the time value, for file “B” append a B. Then combine all the data into one (temporary) file and sort it.

So as an example of what it might look like after the sorting might be:

1674125098362A foo
1674125098369B bar

etc.

Then I suppose convert the epoch milliseconds value back into a readable time (leaving the A and B in place and you can browse all data in one file, in order of time.

Sounds like a lot of work, but, depending on your need it may be worth it?

Alan Kilborn

As a kickstarter down the solution path from my earlier idea, here’s probably the most difficult scripting piece:

from datetime import datetime as dt

def custom_replace_func(m):
    dt_matched_time = dt.strptime(m.group(0), '%Y/%m/%d %H:%M:%S.%f')
    epoch_diff = dt_matched_time - dt(1970, 1, 1)
    epoch_secs_as_str = str(int(epoch_diff.total_seconds()))
    msecs_as_str = str(epoch_diff.microseconds)[0:3]
    return epoch_secs_as_str + msecs_as_str + 'A'

editor.rereplace(r'^\d{4}/\d\d/\d\d \d\d:\d\d:\d\d\.\d{3}', custom_replace_func)

That script will take text data like this in the active file tab:

2023/01/19 10:44:58.362
2023/01/19 10:44:59.365
2023/01/19 10:45:05.998
2023/01/19 10:46:57.038
2023/01/20 09:35:12.267
2023/01/20 09:44:23.334

And convert it to this:

1674125098362A
1674125099365A
1674125105998A
1674125217380A
1674207312267A
1674207863334A

This has its roots in the FAQ about customized replacement via script, see HERE for more details about that type of thing.

Alan Kilborn

And, for completeness, to go back the other way:

def custom_replace_func2(m):
    secs_as_str = m.group(0)[0:10]
    msecs_as_str = m.group(0)[10:13]
    secs_as_float = float(secs_as_str) + (float(msecs_as_str) / 1000.0)
    dt_from_utc_epoch = dt.utcfromtimestamp(secs_as_float)
    return dt_from_utc_epoch.strftime('%Y/%m/%d %H:%M:%S.%f')[:-3]

editor.rereplace(r'^\d{13}(?=A)', custom_replace_func2)

Will take:

1674125098362A
1674125099365A
1674125105998A
1674125217380A
1674207312267A
1674207863334A

and produce:

2023/01/19 10:44:58.362A
2023/01/19 10:44:59.365A
2023/01/19 10:45:05.998A
2023/01/19 10:46:57.038A
2023/01/20 09:35:12.267A
2023/01/20 09:44:23.334A