baptiste-roullin · May 24, 2024 21:08
diff --git a/raindrop-to-eagle.py b/raindrop-to-eagle.py
 # Crude, offensively non-pythonic script to export Raindrop bookmarks in Eagle.app, as URL objects or media.
 # Needs a CSV export from a Raindrop collection. https://help.raindrop.io/export/
 # The collection needs to be set to public.

 import csv
 from random import choice
 import re
 from sys import argv
 from os import path, sep, mkdir
 from datetime import datetime
 import string

 import requests

 outputEncoding = "utf-8"
 inputEncoding = "utf-8"
 # the path to the IMAGES folder in your Eagle library.
 libraryPath = ""  
 # beginning of the name of the folder of last item added to your Eagle Library,
 # exemple:
 # prefix = "LWJKQGX4"
 # TODO: Automate this part
 prefix = ""

 chars = string.ascii_uppercase + string.digits

 class PasswdDialect(csv.Dialect):
    # field separator
    delimiter = ","
    # string separator
    quotechar = '"'
    escapechar = None
    doublequote = True
    lineterminator = "\n"
    quoting = csv.QUOTE_MINIMAL
    skipinitialspace = True

 def retrieveFilePath(fileSendedByArg):

    if path.isabs(fileSendedByArg):  # is path absolute
        return fileSendedByArg
    else:
        return path.dirname(path.abspath(__file__)) + sep + fileSendedByArg

 def getInputFile():
    if len(argv) > 1:
        csvInputFile = retrieveFilePath(argv[1])
        if not path.isfile(csvInputFile):
            exit("the file " + csvInputFile + " doesn't exist")
        print("Using :" + csvInputFile)
    else:
        print("You must provide path to an input file")
        exit()

    print("######################################")
    print("#            PARSING FILE : " + csvInputFile)
    print("######################################")
    return csvInputFile


 def importItems(reader, count):

    # The tricky, hacky, brittle part.
    # From what I can tell, Eagle generate ids for each items this way
    # [4 chars][4 chars][5 chars]
    # The start seems linked to folders.
    # For instance: LWJKQGX40BGQ4
    # We can add stuff directly to the library folder and it will be recognized IFF:
    # the metadata.json inside the item folder has the proper format
    # AND it finds any .url file in the item folder.
    # AND the first two part of the id is the same as some item recently

    # So we generate as many ids as we need.
    # And a bit more to account for potential duplicates from collisions of ids.
    # Didn't I say it was hacky?
    countWithSpare = count + 20
    ids = [
        prefix + "".join([choice(chars) for i in range(5)])
        for j in range(countWithSpare)
    ]
    # removing duplicates.
    uids = set(ids)
    for row in reader:
        id = uids.pop()
        title = re.sub("[,:'/\\\"\\n]", "", row["title"])
        note = row["note"] or row["excerpt"]
        url = row["url"]
        thumb = row["cover"]
        created = str(row["created"])[0:10]
        tags = str(row["tags"].replace("-", " ").replace(", ", ",").split(",")).replace(
            "'", '"'
        )
        btime = str(datetime.fromisoformat(created).timestamp()).split(".0")[0]
        # creating folder
        folderPath = path.join(libraryPath, id + ".info")
        ext = "url"

        if not path.exists(folderPath):
            mkdir(folderPath)

            if url.startswith("https://api.raindrop"):
                res = requests.get(url, timeout=5)
                contentType = res.headers["Content-Type"]
                if contentType.startswith("image"):
                    ext = contentType.replace("image/", "")
                    filename = title + "." + ext
                    with open(path.join(folderPath, filename), "wb") as handler:
                        handler.write(res.content)
                elif contentType.startswith("video"):
                    ext = contentType.replace("video/", "")
                    filename = "test" + "." + ext
                    with open(path.join(folderPath, filename), "wb") as handler:
                        handler.write(res.content)
            else:
                with open(path.join(folderPath, title + ".url"), "a") as file:
                    file.write(f"[InternetShortcut]\nURL={url}")

            # Not bothering with a real JSON manipulation lib.
            with open(path.join(folderPath, "metadata.json"), "a") as file:
                file.writelines(
                    [
                        "{",
                        f'"id": "{id}",',
                        f'"name": "{title}",',
                        '"size": 56,',
                        f'"btime": {btime},',
                        '"mtime": 1716478593302,',
                        f'"ext": "{ext}",',
                        f'"tags": {tags},',
                        f'"folders": [],',
                        '"isDeleted": false,',
                        f'"url": "{url}",',
                        f'"annotation": "{note}",',
                        '"modificationTime": 1716478593300,',
                        '"height": 408,',
                        '"width": 720,',
                        '"lastModified": 1716479798004,',
                        '"palettes": [],',
                        f'"noThumbnail": false',
                        "}",
                    ]
                )

        else:
            # should not happen hashtag Famous Last Words.
            print("folder already exists")
        continue


 csvInputFile = getInputFile()

 with open(csvInputFile, mode="r", newline="", encoding=inputEncoding) as csvfile:
    counter = csv.DictReader(csvfile, dialect=PasswdDialect())
    count = sum(1 for _ in counter)
    csvfile.seek(0)
    # TODO : read file only once.
    reader = csv.DictReader(csvfile, dialect=PasswdDialect())
    importItems(reader, count)
	# Crude, offensively non-pythonic script to export Raindrop bookmarks in Eagle.app, as URL objects or media.
	# Needs a CSV export from a Raindrop collection. https://help.raindrop.io/export/
	# The collection needs to be set to public.

	import csv
	from random import choice
	import re
	from sys import argv
	from os import path, sep, mkdir
	from datetime import datetime
	import string

	import requests

	outputEncoding = "utf-8"
	inputEncoding = "utf-8"
	# the path to the IMAGES folder in your Eagle library.
	libraryPath = ""
	# beginning of the name of the folder of last item added to your Eagle Library,
	# exemple:
	# prefix = "LWJKQGX4"
	# TODO: Automate this part
	prefix = ""

	chars = string.ascii_uppercase + string.digits

	class PasswdDialect(csv.Dialect):
	# field separator
	delimiter = ","
	# string separator
	quotechar = '"'
	escapechar = None
	doublequote = True
	lineterminator = "\n"
	quoting = csv.QUOTE_MINIMAL
	skipinitialspace = True

	def retrieveFilePath(fileSendedByArg):

	if path.isabs(fileSendedByArg): # is path absolute
	return fileSendedByArg
	else:
	return path.dirname(path.abspath(__file__)) + sep + fileSendedByArg

	def getInputFile():
	if len(argv) > 1:
	csvInputFile = retrieveFilePath(argv[1])
	if not path.isfile(csvInputFile):
	exit("the file " + csvInputFile + " doesn't exist")
	print("Using :" + csvInputFile)
	else:
	print("You must provide path to an input file")
	exit()

	print("######################################")
	print("# PARSING FILE : " + csvInputFile)
	print("######################################")
	return csvInputFile


	def importItems(reader, count):

	# The tricky, hacky, brittle part.
	# From what I can tell, Eagle generate ids for each items this way
	# [4 chars][4 chars][5 chars]
	# The start seems linked to folders.
	# For instance: LWJKQGX40BGQ4
	# We can add stuff directly to the library folder and it will be recognized IFF:
	# the metadata.json inside the item folder has the proper format
	# AND it finds any .url file in the item folder.
	# AND the first two part of the id is the same as some item recently

	# So we generate as many ids as we need.
	# And a bit more to account for potential duplicates from collisions of ids.
	# Didn't I say it was hacky?
	countWithSpare = count + 20
	ids = [
	prefix + "".join([choice(chars) for i in range(5)])
	for j in range(countWithSpare)
	]
	# removing duplicates.
	uids = set(ids)
	for row in reader:
	id = uids.pop()
	title = re.sub("[,:'/\\\"\\n]", "", row["title"])
	note = row["note"] or row["excerpt"]
	url = row["url"]
	thumb = row["cover"]
	created = str(row["created"])[0:10]
	tags = str(row["tags"].replace("-", " ").replace(", ", ",").split(",")).replace(
	"'", '"'
	)
	btime = str(datetime.fromisoformat(created).timestamp()).split(".0")[0]
	# creating folder
	folderPath = path.join(libraryPath, id + ".info")
	ext = "url"

	if not path.exists(folderPath):
	mkdir(folderPath)

	if url.startswith("https://api.raindrop"):
	res = requests.get(url, timeout=5)
	contentType = res.headers["Content-Type"]
	if contentType.startswith("image"):
	ext = contentType.replace("image/", "")
	filename = title + "." + ext
	with open(path.join(folderPath, filename), "wb") as handler:
	handler.write(res.content)
	elif contentType.startswith("video"):
	ext = contentType.replace("video/", "")
	filename = "test" + "." + ext
	with open(path.join(folderPath, filename), "wb") as handler:
	handler.write(res.content)
	else:
	with open(path.join(folderPath, title + ".url"), "a") as file:
	file.write(f"[InternetShortcut]\nURL={url}")

	# Not bothering with a real JSON manipulation lib.
	with open(path.join(folderPath, "metadata.json"), "a") as file:
	file.writelines(
	[
	"{",
	f'"id": "{id}",',
	f'"name": "{title}",',
	'"size": 56,',
	f'"btime": {btime},',
	'"mtime": 1716478593302,',
	f'"ext": "{ext}",',
	f'"tags": {tags},',
	f'"folders": [],',
	'"isDeleted": false,',
	f'"url": "{url}",',
	f'"annotation": "{note}",',
	'"modificationTime": 1716478593300,',
	'"height": 408,',
	'"width": 720,',
	'"lastModified": 1716479798004,',
	'"palettes": [],',
	f'"noThumbnail": false',
	"}",
	]
	)

	else:
	# should not happen hashtag Famous Last Words.
	print("folder already exists")
	continue


	csvInputFile = getInputFile()

	with open(csvInputFile, mode="r", newline="", encoding=inputEncoding) as csvfile:
	counter = csv.DictReader(csvfile, dialect=PasswdDialect())
	count = sum(1 for _ in counter)
	csvfile.seek(0)
	# TODO : read file only once.
	reader = csv.DictReader(csvfile, dialect=PasswdDialect())
	importItems(reader, count)