photolib/photoapp/ingest.py

import argparse
import traceback
from photoapp.image import get_jpg_info, get_hash, get_mtime, special_magic
from itertools import chain
from photoapp.types import Photo, PhotoSet, known_extensions, regular_images, files_raw, files_video, map_extension
import os


"""
Photo sorting rules:

jpeg
    exif date
    file modification date
raw
    group with exif date of jpeg with same name
    file modification date
mov, video, or other
    modification date
"""


def pprogress(done, total=None):
    print("  complete: {}{}\r".format(done, " / {} ".format(total) if total else ''), end='')


def group_by_extension(files):
    byext = {k: [] for k in known_extensions}
    excluded = []

    for item in files:
        if not os.path.isfile(item):  # Not a file
            excluded.append(item)
            continue
        extension = item.split(".")
        if len(extension) < 2:  # no extension
            excluded.append(item)
            continue
        extension = map_extension(extension[-1].lower())
        if extension not in known_extensions:  # an extension we don't support
            excluded.append(item)
            continue
        byext[extension.lower()].append(item)

    return (byext, excluded)


def get_photosets(files):
    byext, skipped = group_by_extension(files)

    photosets = []

    # process regular images first.
    for item in chain(*[byext[ext] for ext in regular_images]):
        photosets.append(get_jpg_info(item))

    # process raws
    for item in chain(*[byext[ext] for ext in files_raw]):
        itemmeta = Photo(hash=get_hash(item), path=item, size=os.path.getsize(item),
                         format=special_magic(item), fname=os.path.basename(item))
        fprefix = os.path.basename(item)[::-1].split(".", 1)[-1][::-1]
        fmatch = "{}.jpg".format(fprefix.lower())  # if we're inspecting "foobar.raw", match it with "foobar.jpg"
                                                   # TODO does this account for extension mappinh like jpeg->jpg?
        foundmatch = False
        for photo in photosets:
            for fmt in photo.files[:]:
                if os.path.basename(fmt.path).lower() == fmatch:
                    foundmatch = True
                    photo.files.append(itemmeta)
                    break
            if foundmatch:
                break
        if not foundmatch:
            mtime = get_mtime(item)
            print("no match found for", itemmeta.path, "but importing anyway")
            photosets.append(PhotoSet(date=mtime, date_real=mtime, files=[itemmeta]))
        # TODO handle any xmp without an associated regular image or cr2

    # process other known formats
    for item in chain(*[byext[ext] for ext in files_video]):
        itemmeta = Photo(hash=get_hash(item), path=item, size=os.path.getsize(item),
                         format=special_magic(item), fname=os.path.basename(item))
        mtime = get_mtime(item)
        photosets.append(PhotoSet(date=mtime, date_real=mtime, files=[itemmeta]))

    return photosets, skipped


def batch_ingest(library, files):
    sets, skipped = get_photosets(files)

    print("\nUpdating database")
    done = 0
    total = len(sets)
    for photoset in sets:
        try:
            library.add_photoset(photoset)
            pprogress(done, total)
            done += 1
        except Exception:
            traceback.print_exc()
            pass
    print("\nUpdate complete")


def main():
    parser = argparse.ArgumentParser(description="Library ingestion tool")
    parser.add_argument("files", nargs="+")
    args = parser.parse_args()

    library = PhotoLibrary("photos.db", "./library", "./cache")

    batch_ingest(library, args.files)


if __name__ == '__main__':
    main()