Index a collection of photographs into a PostgreSQL database.
In order to scratch an itch of mine, I decided to index my digital photograph collection by creating a database with the file pathname, the date the picture it was taken, and the place, obtained from the GPS coordinates. Some of my pictures were sent to me via social networks, which regrettably strip all the metadata from the files and, in this case, only the date the picture was obtained (i.e. the file creation date) is stored. In the future who knows, tags and people's names may be added.
This program depends on the modules psycopg2, to manage the PostgreSQL database, and exif to extract metadata from digital photograph files.
The programs does reverse geolocation using a public API from OpenCage. You will need to obtain an API key from them in order to run this program.
There is no automated install or setup.py program at this time, so the recommended way to use this program for now is:
Create a database on a postgreSQL server to store the data into.
Create a virtual environment and install the dependencies in the directory on which you have git cloned this repo:
$ python3 -m venv photodb $ cd photodb $ . bin/activate (photodb)$ python3 -m pip install -r < requirements.txt
Copy the file photodb.ini.sample into config.ini and edit it with
the details of your OpenCage API key and your database.
Initialise the database:
(photodb)$ python3 photodb.py --config=config.ini --initdb
And at this point you can index your picture collection into the database by running the program:
(photodb)$ python3 photodb.py --config=config.ini --scan-dir=/path/to/pics
You can get help on the different options the program can take by using the --help command line option:
(photodb)$ python3 photodb.py --help
usage: photodb.py [-h] [--version] [--scan-dir DIR] [--save FILE]
[--extract FILE] [--replay FILE] [--picture FILE]
[--config CFG-FILE] [--initdb] [--loglevel LOGLEVEL]
[--errorlog ERRORLOG]
optional arguments:
-h, --help show this help message and exit
--version, -v print version and exit
--scan-dir DIR, -d DIR
recursively scan a directory and add the pictures to
the database
--save FILE, -s FILE save inserted records to FILE instead of default
~/.photodb_records
--extract FILE, -x FILE
extract all records from the database into a json FILE
(opposite of --replay option)
--replay FILE, -r FILE
insert in the database the json records in FILE,
previously specified with the --save option
--picture FILE, -p FILE
add a single picture file to the database
--config CFG-FILE, -c CFG-FILE
read configuration from this file. Default:
/usr/local/etc/photodb.ini
--initdb, -i initialise the database (specified in the
configuration) WARNING: This wipes out an existing
database!
--loglevel LOGLEVEL, -l LOGLEVEL
set logging level. Default INFO
--errorlog ERRORLOG, -e ERRORLOG
error log file. Default ~/photodb_errors.log. Errors
will always get logged into a file. If you don't want
that (bad idea) set this to /dev/null
The program is a bit (too much?) paranoid about not repeating calls to
the OpenCage site, and saves a json copy of the records inserted as a
backup, which can eventually be really big. It can, at a later time,
rebuild the database without re-scanning the files and without querying
OpenCage. That would be the purpose of the --save, --extract,
and --replay command line options.
The configuration is taken by default from the file /usr/local/photodb.ini.
If you don't have the configuration file there, or you want to use a
different configuration file, you can always use the --config command
line option.
A sample configuration file is provided in config.ini.sample which
contains the following:
# these are the parameters for your postgresql database [postgresdb] dbname=testphotodb user=testuser password=hello host=localhost port=5432 # this stores your OpenCage API key for doing inverse geolocation [opencage] apikey = b1ab1ab1ab1ab1ab1ab1ab1ab1ab1ab1
This program is released under the MIT License