This is a script for creating a sequence in Premiere Pro containing all the media in a directory arranged by earliest datetime present in the file metadata. It will search Date modified, Date created, Date taken, Date accessed, and Media created and use the earliest date available from those for sorting (unless otherwise specified) to yield the greatest sorting accuracy possible (at least for the case I'm using it for).
I'm creating this because Premiere Pro itself was for some reason unable to correctly sort the media by date (for me at least). Might've been just something I was doing wrong, but regardless, writing this script is the easiest solution I've found.
This was written on a Windows 10 machine, and almost certainly wouldn't work on Mac or Linux without modification of Part 1 (where the directories are searched and the media is sorted), as that uses win32com.client for getting the metadata of each file. But if anyone wants to attempt that modification, be my guest!
First, Premiere Pro needs to be running with the project you want to add the sequence to already open. You should also have already imported all of the media that you want to be sorted and placed in the sequence, and that imported media should be organized the same way inside Premiere Pro as it is organized in your filesystem. Basically, if you run the script with the search path pointing to a parent directory containing subdirectories each containing photos and videos, then there should be a parent directory (or "bin" as Adobe likes to call it) inside the Premiere Pro project with the same subdirectories/bins as was on the filesystem. This script will not import the files into your project for you. That would just take too long for big projects (or at least the one I'm creating this for).
Next, create a sequence in your Premiere Pro project called "config_sequence", add an example of every media type that the script will encounter to the first video track, and place any motion effects on each one that you want the script to apply to each instance of that media type it encounters. The script will read that sequence and create a new sequence with identical properties, and then add the media in chronological order, deciding what to do with each photo or video based on what was done with the photo or video most similar to it in the config_sequence. It will determine this based on both media type (photo vs video) and the dimensions of the media. It will not consider the file extension (i.e. it will handle a JPG and PNG identically if they have the same dimensions).
Finally, you will need to create a JSON-formatted configuration file following the format of timezone_config.json. That is, each subdirectory should have a dictionary consisting of a "timezones" array containing one or more arrays of length 2, each consisting of a date and time as its first element after which the timezone specified in its second element will be applied to the media. Each timezone should be specified using its name in the tz database. The first timezones array entry should have an empty string "" in place of a date and time to indicate that all media until the second entry's date and time (if applicable) should use the timezone specified in the first entry. Additionally, the dates and times used in each array entry should be represented by the dates and times in the local timezone that is present in the entry's second element. Finally, to override finding the earliest datetime, add a "datefield" entry to the dictionary with the metadata name to always attempt to use instead.
python3 ./create_chronological_prpro_seq <relative search path> <sorted files list json filename> <timezone config json filename> <name of sequence in Premiere>
Example: python3 ./create_chronological_prpro_seq .. sorted_files.json timezone_config.json GENERATED_SEQUENCE
The <sorted files list json filename> will be created if it doesn't already exist, and it will contain a sorted list of all the media. If it already exists, then the script will read from this file rather than going through the sorting process again, as the sorting process can take hours in some cases depending on the media types it is sorting.
A file named bin_dict_pkl.pkl will also be created to store the index of all the media in the Premiere Pro project bins so that the bins don't need to be indexed every time the script is run (this pickling was more useful when initially writing the script though, so it may be removed in the future). However, if you close Premiere and re-open it, you will need to delete bin_dict_pkl.pkl so that the script will regenerate it, otherwise you will get an error.
If there is no sequence in the Premiere Pro project with the name <name of sequence in Premiere>, then a new sequence will created using the same settings as used in the configuration sequence. However, if there is a sequence with that name, then the script will find the penultimate clip in the sequence, and then it will add clips to the sequence beginning from the item subsequent to that penultimate clip in the sorted file list. Essentially, this allows the script to be interrupted and then resumed. This is especially helpful because the longer the script runs, the slower it gets. To speed things up, it can be effective to interrupt the script with Ctrl-z, close and reopen Premiere, delete the bin_dict_pkl.pkl file, then run the script again. The script will resume adding clips to the sequence from where it left off, and will perform much faster for a while than the speed it was performing at before.
I'd like to thank Pymiere for making this possible, and I'd like to ask Adobe why their absurdly expensive software-as-a-service can't seem to provide this service itself.