PK ! ы imserv/__init__.pyfrom flask import Flask
app = Flask(__name__)
from .views import index, get_image
from .api import create_image
from .main import ImServ
PK ! t imserv/__main__.pyimport click
from .main import ImServ
@click.command()
@click.option('--port', default=29635)
def runserver(port):
ImServ(port=port).runserver()
if __name__ == '__main__':
runserver()
PK ! 0[
imserv/api.pyfrom flask import request, jsonify, make_response, abort
from werkzeug.utils import secure_filename
from pathlib import Path
from io import BytesIO
from . import app, db
@app.route('/api/images/create', methods=['POST'])
def create_image():
if 'file' in request.files:
tags = request.form.get('tags')
file = request.files['file']
with BytesIO() as bytes_io:
file.save(bytes_io)
db_image = db.Image.from_bytes_io(bytes_io,
filename=secure_filename(file.filename), tags=tags)
if isinstance(db_image, str):
return abort(make_response(jsonify(message=db_image), 409))
else:
return jsonify({
'filename': str(Path(db_image.filename).name),
'trueFilename': str(db_image.filename)
}), 201
response = make_response()
response.status_code = 304
return response
PK ! x imserv/config.pyimport os
from pathlib import Path
config = {
'database': 'imserv',
'host': 'localhost',
'port': 29635,
'debug': False,
'threaded': False,
'hash_size': 32,
'hash_difference_threshold': 0
}
for k in config.keys():
env_k = 'IMSERV_' + k.upper()
if env_k in os.environ.keys():
v = os.environ[env_k]
if k in {'port', 'hash_size', 'hash_difference_threshold', 'skip_hash'}:
config[k] = int(v)
elif k in {'debug', 'threaded'}:
config[k] = bool(v)
else:
config[k] = v
if config.get('folder', None) is None:
OS_IMG_FOLDER_PATH = Path.home().joinpath('Pictures')
assert OS_IMG_FOLDER_PATH.exists()
IMG_FOLDER_PATH = OS_IMG_FOLDER_PATH.joinpath('imserv')
IMG_FOLDER_PATH.mkdir(exist_ok=True)
config['folder'] = IMG_FOLDER_PATH
else:
config['folder'] = Path(config['folder'])
config['blob_folder'] = config['folder'].joinpath('blob')
config['blob_folder'].mkdir(exist_ok=True)
PK ! imserv/db.pyimport peewee as pv
import imagehash
from uuid import uuid4
from nonrepeat import nonrepeat_filename
from slugify import slugify
import PIL.Image
import os
import sys
from datetime import datetime
from urllib.parse import quote, urlparse
from .config import config
from .util import get_checksum, get_image_hash
__all__ = ('Tag', 'Image', 'ImageTags', 'create_all_tables')
class BaseModel(pv.Model):
class Meta:
database = pv.PostgresqlDatabase(config['database'])
class ImageHashField(pv.TextField):
def db_value(self, value):
if value:
return str(value)
def python_value(self, value):
if value:
return imagehash.hex_to_hash(value)
class Tag(BaseModel):
name = pv.TextField()
def to_json(self):
return {
'name': self.name,
'images': [img.to_json() for img in self.images]
}
class Image(BaseModel):
file_id = pv.IntegerField(null=True)
filename = pv.TextField(null=False)
checksum = pv.TextField(null=False)
image_hash = ImageHashField(null=True)
tags = pv.ManyToManyField(Tag, backref='images')
@classmethod
def from_bytes_io(cls, im_bytes_io, filename=None, tags=None):
"""
:param im_bytes_io:
:param str filename:
:param str|list|tuple tags:
:return:
"""
if tags is None:
tags = list()
if not filename or filename == 'image.png':
filename = slugify('-'.join(tags)) + str(uuid4())[:8] + '.png'
filename = nonrepeat_filename(filename,
root=str(config['blob_folder']))
filename = str(config['blob_folder'].joinpath(filename))
checksum = get_checksum(im_bytes_io)
im = PIL.Image.open(im_bytes_io)
image_hash = get_image_hash(im)
im.save(filename)
db_image = cls.create(
file_id=os.stat(filename).st_ino,
filename=filename,
checksum=checksum,
image_hash=image_hash
)
for tag in tags:
db_image.tags.add(Tag.get_or_create(name=tag)[0])
return db_image
@classmethod
def from_existing(cls, filename, tags=None):
if tags is None:
tags = list()
filename = str(filename)
db_image = cls.create(
file_id=os.stat(filename).st_ino,
filename=filename,
checksum=get_checksum(filename),
image_hash=get_image_hash(filename)
)
for tag in tags:
db_image.tags.add(Tag.get_or_create(name=tag)[0])
db_image.path = filename
return db_image
@classmethod
def similar_images(cls, im):
image_hash = get_image_hash(im)
if config['hash_difference_threshold']:
for db_image in cls.select():
if db_image.image_hash - image_hash < config['hash_difference_threshold']:
yield db_image
else:
return cls.select().where(cls.image_hash == image_hash)
def get_image(self, max_width=800, max_height=800):
url = self.url
if url:
return f''
def _repr_html_(self):
return self.get_image(800, 800)
def _repr_json_(self):
result = self.to_json()
result['image'] = self.filename
return result
@property
def url(self):
if not urlparse(self.filename).netloc:
return 'http://{}:{}/images?filename={}'.format(
config['host'],
config['port'],
quote(str(self.filename), safe='')
)
else:
return self.filename
def to_json(self):
return dict(
id=self.id,
image=self.get_image(400, 400),
filename=self.filename,
notes=[n.data for n in self.notes],
tags=[t.name for t in self.tags]
)
handsontable_config = {
'renderers': {
'image': 'html'
},
'config': {
'colWidths': {
'image': 400
}
}
}
ImageTags = Image.tags.get_through_model()
class Note(BaseModel):
image = pv.ForeignKeyField(Image, backref='notes')
data = pv.TextField()
def _repr_markdown_(self):
return self.data
def to_json(self):
return {
'file_id': self.file_.to_json(),
'data': self.data
}
def create_all_tables():
for cls in sys.modules[__name__].__dict__.values():
if hasattr(cls, '__bases__') and issubclass(cls, pv.Model):
if cls is not BaseModel:
cls.create_table()
PK ! m6 6 imserv/main.pyimport sys
import re
from threading import Thread
import subprocess
from tqdm import tqdm
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
from pathlib import Path
import random
from .config import config
from .util import open_browser_tab, images_in_path, get_checksum, get_image_hash
from . import app
__all__ = ('ImServ',)
class FileCreationHandler(FileSystemEventHandler):
def __init__(self, expected_total):
self.tqdm = tqdm(
total=expected_total
)
def on_created(self, event):
self.tqdm.update()
class ImServ:
def __init__(self, **kwargs):
config.update(kwargs)
from . import db
self.db = dict(
image=db.Image,
note=db.Note,
tag=db.Tag
)
def __getitem__(self, item):
return self.db[item]
def runserver(self):
"""Run the image server (see README.md)
"""
def _runserver():
app.run(
host=config['host'],
port=config['port'],
debug=config['debug']
)
def _runserver_in_thread():
open_browser_tab('http://{}:{}'.format(
config['host'],
config['port']
))
self.server_thread = Thread(target=_runserver)
self.server_thread.daemon = True
self.server_thread.start()
if config['threaded'] or 'ipykernel' in ' '.join(sys.argv):
_runserver_in_thread()
else:
_runserver()
def search_filename(self, filename_regex, calculate_hash=False):
def _search():
for file_path in images_in_path():
if re.search(filename_regex, str(file_path), flags=re.IGNORECASE):
db_image = self._get_or_create(file_path, calculate_hash)
yield db_image
return sorted(_search(), key=lambda x: x.filename)
def search_database(self, query):
return self.db['image'].select().where(query)
def refresh(self, calculate_hash=False):
for file_path in tqdm(tuple(images_in_path())):
db_image = self._get_or_create(file_path, calculate_hash)
if db_image is not None:
if calculate_hash:
db_image.image_hash = get_image_hash(file_path)
db_image.save()
def _get_or_create(self, file_path, calculate_hash):
checksum = get_checksum(file_path)
image_hash = None
if calculate_hash:
image_hash = get_image_hash(file_path)
db_image = self.db['image'].get_or_none(
file_id=file_path.stat().st_ino
)
if db_image is None:
db_image = self.db['image'].create(
file_id=file_path.stat().st_ino,
checksum=checksum,
image_hash=image_hash,
filename=str(file_path)
)
else:
if image_hash:
db_image.image_hash = image_hash
if checksum != db_image.checksum:
db_image.checksum = checksum
db_image.filename = str(file_path)
db_image.save()
return db_image
def import_pdf(self, pdf_filename, calculate_hash=False):
"""
Import images from a PDF. Poppler (https://poppler.freedesktop.org) will be required.
In Mac OSX, `brew install poppler`.
In Linux, `yum install poppler-utils` or `apt-get install poppler-utils`.
Arguments:
pdf_filename {str, pathlib.Path} -- Path to PDF file.
"""
def _extract_pdf():
filename_initials = ''.join(c for c in str(Path(pdf_filename).name) if c.isupper())
if not filename_initials:
filename_initials = pdf_filename[0]
number_of_images = len(subprocess.check_output([
'pdfimages',
'-list',
str(pdf_filename)
]).split(b'\n')) - 2
observer = Observer()
event_handler = FileCreationHandler(expected_total=number_of_images)
observer.schedule(event_handler, str(dst_folder_path), recursive=False)
observer.setDaemon(True)
observer.start()
try:
subprocess.call([
'pdfimages',
'-p',
'-png',
str(pdf_filename),
str(dst_folder_path.joinpath(filename_initials))
])
observer.stop()
except KeyboardInterrupt:
observer.stop()
observer.join()
event_handler.tqdm.close()
dst_folder_path = config['folder'].joinpath('pdf').joinpath(Path(pdf_filename).stem)
if not dst_folder_path.exists():
dst_folder_path.mkdir(parents=True)
_extract_pdf()
for file_path in tqdm(tuple(images_in_path(dst_folder_path))):
self._get_or_create(file_path, calculate_hash=calculate_hash)
def get_pdf_image(self, filename_regex, page_start, page_end,
calculate_hash=True, randomize=False):
"""Search images corresponding to PDF in config['folder']
Arguments:
filename_regex {str} -- Regex matching the PDF filename
page_start {int} -- First page to search
page_end {int} -- Last page to search
Yields:
db.Image object corresponding to the criteria
"""
def _get_image():
for file_path in images_in_path():
match_obj = re.search(rf'{filename_regex}.*(?:[^\d])(\d+)-\d+\.png', str(file_path), flags=re.IGNORECASE)
if match_obj is not None:
page_number = int(match_obj.group(1))
if page_number in range(page_start, page_end + 1):
db_image = self._get_or_create(file_path, calculate_hash)
yield page_number, db_image
if randomize:
images = [db_image for i, db_image in _get_image()]
random.shuffle(images)
return images
else:
return [db_image for i, db_image in sorted(_get_image(), key=lambda x: x[0])]
def last_created(self):
return self.db['image'].select().order_by(self.db['image'].id.desc())
PK ! " imserv/static/index.cssinput {
font-size: 24px;
margin: 0 auto;
margin-top: 20px;
}
#tags-area, #input-bar {
width: 500px;
}
#tags-bar {
width: 100%;
}
img {
margin-top: 20px;
}
PK ! \ imserv/static/index.jsconst tagsBar = document.getElementById('tags-bar');
const inputBar = document.getElementById('input-bar');
const imageCanvas = document.getElementById('image-canvas');
let tags = JSON.parse(localStorage.getItem('tags') || '[]');
let imagePath = '';
tagsBar.value = joinTags();
inputBar.value = imagePath;
inputBar.onpaste = ()=>{
const items = (event.clipboardData || event.originalEvent.clipboardData).items;
// console.log(items); // will give you the mime types
for (index in items) {
const item = items[index];
if (item.kind === 'file') {
const file = item.getAsFile();
let reader = new FileReader();
reader.onload = function(event) {
const extension = file.type.match(/\/([a-z0-9]+)/i)[1].toLowerCase();
let formData = new FormData();
formData.append('file', file, file.name);
formData.append('extension', extension);
formData.append('mimetype', file.type);
formData.append('submission-type', 'paste');
// formData.append('imagePath', imagePath);
formData.append('tags', tags);
localStorage.setItem('tags', JSON.stringify(tags));
fetch('/api/images/create', {
method: 'POST',
body: formData
}).then(response=>response.json())
.then(responseJson=>{
if(responseJson.filename){
inputBar.value = responseJson.filename;
imageCanvas.src = '/images?filename=' + encodeURIComponent(responseJson.trueFilename);
} else {
alert(responseJson.message);
}
})
.catch(error => {
console.error(error);
});
};
reader.readAsBinaryString(file);
}
}
}
tagsBar.addEventListener("keyup", function(event) {
function purge(tag){
tag = tag.trim();
if(tag){
tags.push(tag);
}
}
tags = [];
let purgable = true;
let currentTag = ''
tagsBar.value.split('').forEach((character, index)=>{
if(character === ',' && purgable){
if(purgable){
purge(currentTag);
currentTag = '';
} else {
currentTag += character;
}
} else if (character === '\"'){
purgable = !purgable;
} else {
currentTag += character;
}
});
purge(currentTag);
});
function joinTags(){
let result = []
tags.forEach((tag, index)=>{
if(tag.indexOf(',') !== -1){
result.push('\"' + tag + '\"');
} else {
result.push(tag);
}
});
return result.join(', ')
}PK ! imserv/templates/index.html
PK ! tB
imserv/util.pyfrom threading import Thread
from time import sleep
import webbrowser
from PIL import Image, ImageChops
from pathlib import Path
import imagehash
from send2trash import send2trash
import hashlib
from io import BytesIO
import logging
from .config import config
def open_browser_tab(url):
def _open_tab():
sleep(1)
webbrowser.open_new_tab(url)
thread = Thread(target=_open_tab)
thread.daemon = True
thread.start()
def trim_image(im):
bg = Image.new(im.mode, im.size, im.getpixel((0,0)))
diff = ImageChops.difference(im, bg)
diff = ImageChops.add(diff, diff, 2.0, -100)
bbox = diff.getbbox()
if bbox:
im = im.crop(bbox)
return im
def shrink_image(im, max_width=800):
width, height = im.size
if width > max_width:
im.thumbnail((max_width, height * max_width / width))
return im
def remove_duplicate(file_path=config['folder']):
hashes = set()
for p in images_in_path(file_path):
h = get_image_hash(Image.open(p), trim=True)
if h in hashes:
print('Deleting {}'.format(p))
send2trash(p)
else:
hashes.add(h)
def remove_non_images(file_path=config['folder']):
images = set(images_in_path(file_path))
for p in Path(file_path).glob('**/*.*'):
if not p.is_dir() and p not in images:
send2trash(str(p))
def images_in_path(file_path=config['folder']):
for p in Path(file_path).glob('**/*.*'):
if not p.is_dir() and not p.name.startswith('.') \
and p.suffix.lower() in {'.png', '.jpg', '.jp2', '.jpeg', '.gif'}:
yield p
def complete_path_split(path, relative_to=config['folder']):
components = []
path = Path(path)
if relative_to:
path = path.relative_to(relative_to)
while path.name:
components.append(path.name)
path = path.parent
return components
def get_image_hash(im, trim=True, **kwargs):
if isinstance(im, (str, Path)):
try:
im = Image.open(im)
except OSError:
return None
if trim:
im = trim_image(im)
return imagehash.whash(
im,
hash_size=config['hash_size'],
**kwargs
)
def get_checksum(fp):
if isinstance(fp, BytesIO):
checksum = hashlib.md5(fp.getvalue()).hexdigest()
elif isinstance(fp, (str, Path)):
checksum = hashlib.md5(Path(fp).read_bytes()).hexdigest()
else:
logging.error('Cannot generate checksum')
return None
return checksum
PK ! R~/ imserv/views.pyfrom flask import render_template, send_file, request
from urllib.parse import unquote
from pathlib import Path
from .config import config
from . import app
@app.route('/')
def index():
return render_template('index.html')
@app.route('/images')
def get_image():
file_path = Path(unquote(request.args.get('filename')))
if not file_path.exists():
file_path = config['folder'].joinpath(file_path)
print('Serving: {}'.format(file_path))
return send_file(str(file_path))
PK !H=/ 4 * imserv-0.1.11.3.dist-info/entry_points.txtN+I/N.,()-N-*PzyVEy ".. PK ! : : ! imserv-0.1.11.3.dist-info/LICENSEMIT License
Copyright (c) 2018 Pacharapol Withayasakpunt
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
PK !H W X imserv-0.1.11.3.dist-info/WHEEL
A
н#Z."jm)Afb~ڠO68oF04UhoAf
f4=4h0k::wXPK !H9N %
" imserv-0.1.11.3.dist-info/METADATAVKs6Wl'H3+؞vbc; ! r)"&
{С|݅ЉD8ZE0G(0YX4O0 QVE!*R*bh4=NgR 8e9$@x+"G(sc(
>rTe.c(Z\% s`.aAI9Q|9hҼ1*K]pRL"<$cVbok-^I& FyA8 ~MDJ!jĈDݡvׂRZ,Q8F#; mJ%"(nD^2d/ vr1ϥ!cak62]bZTv}O
xpqggf2 mld#?V%ÿ܀c*9
ϥ8:IɏV.#kas.9_' -J :֗Yё}$HKZ쩊`6y:dNJm Ć
(҃ mI^;JIt`*cWnN
8L! {1_૿R>Ư05DangG$ 'Fh5V/<zGm?)ai)h;a`?HwʁYLEسVHA-O hVӢMYn~ JIMԯ]4s.V
sg rmf Rtmö夎mJ5e<ټۺ ƨ? ФIg(AmҮIʂ幉{YI㔇4qE?hMQ ci^ E^`}Kk/^$ ,pA_*Vbר=O4 j2'RDp9j.
p[H& N-#;]9]\()txgHVkPxS6~1;`17QמA(Ed$1Vrޢ"T//bBSXvaFL0}՜4ѓӡ":uV#9<焤m
_kXS
iԨö5艖
l oډ:ތVmgOE| -P@1{A2.:G|uSlvKW==&P֝/
?ɮ"Iť[o3yvPu,oEhY盳 ?d"o Lݦ3lv
M$ü>d)(mJ1 PK ! ы imserv/__init__.pyPK ! t imserv/__main__.pyPK ! 0[
imserv/api.pyPK ! x imserv/config.pyPK ! imserv/db.pyPK ! m6 6 imserv/main.pyPK ! " 6 imserv/static/index.cssPK ! \ 6 imserv/static/index.jsPK ! @ imserv/templates/index.htmlPK ! tB
IC imserv/util.pyPK ! R~/ {M imserv/views.pyPK !H=/ 4 * O imserv-0.1.11.3.dist-info/entry_points.txtPK ! : : ! P imserv-0.1.11.3.dist-info/LICENSEPK !H W X T imserv-0.1.11.3.dist-info/WHEELPK !H9N %
" !U imserv-0.1.11.3.dist-info/METADATAPK !H_! Y imserv-0.1.11.3.dist-info/RECORDPK L \