get-check-metadata.ipynb

{
 "cells": [
  {
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "cell_type": "markdown",
   "source": "# Check Image Meta Data (e.g. from EXIF)\n\nThis script pulls the EXIF data from a JPG image. For now, we skip PDF/PNG images although they do contain metadata.\n\nRequires:\n\n>`$ pip install exifread`\n\nSet `mount` to point to the mountpoint.\n\nSet `path_to_checks` to point at the sub-folder with the checks\n\nThe final path is the join of these two - i.e. `mount/path_to_checks`\n\nWe filter mostly for useful EXIF data fields (`cols`). But to grab all EXIF data set `limit_to_cols=False` when calling the `stream_exif` routine."
  },
  {
   "metadata": {
    "trusted": true,
    "deletable": true,
    "collapsed": false,
    "editable": true
   },
   "cell_type": "code",
   "source": "import os\nimport json\nimport sys\nimport exifread\nimport pandas\n\n# Control the checkpoint size - number of images to process and save at a time\ncheck_point = 100\n\n# EXIF tags that we want to extract and use as cols in our csv output\ncols = [\"EXIF ISOSpeedRatings\",\"EXIF Flash\",\"Image YResolution\",\"EXIF FNumber\",\"Image Orientation\",\"Image Model\",\n        \"Image XResolution\",\"Image XResolution\",\"EXIF BrightnessValue\",\"EXIF ExifImageWidth\",\"EXIF ExifImageLength\",\n        \"EXIF LensModel\", \"EXIF SubjectArea\", \"Image Make\"]\n\n# Exif data streamed to here:\ndf = pandas.DataFrame()\n\ndef save(data,csv_file):\n    # Now create a frame for post-processing and/or CSV export\n    df = pandas.DataFrame(data)\n    df.set_index('file', inplace=True)\n    df.to_csv(csv_file)\n\n# Set limit_to_cols to False if you want to collect ALL the EXIF fields (will slow things down)\ndef stream_exif(path,limit_to_cols=True,csv_file='Exif_data_checks.csv'):\n    #path = os.getcwd()  #current directory, if running script from checks folder\n    print(\"Looking for jpg images in {}\".format(path))\n    exif_stream = []\n    try:\n        extensions = [\"jpg\",\"jpeg\"]\n        file_count = 0\n        for subdir, dirs, files in os.walk(folder):\n            filelist = filter(lambda f: f.split('.')[-1].lower() in extensions, files)\n            filelist = sorted(filelist)\n            if len(list(filelist)) == 0:\n                print(\"No jpg images in folder {}\".format(subdir))\n            else:\n                print(\"Found {} jpg images in folder {}\".format(len(list(filelist)),subdir))\n            for file in files:\n                if file.split('.')[-1].lower() in extensions: \n                    with open(os.path.join(subdir,file), 'rb') as f:\n                        tags = exifread.process_file(f, details=False)\n                        if limit_to_cols:\n                            vals = {tag: field for (tag, field) in tags.items() if tag in cols}\n                        else:\n                            vals = {tag: field for (tag, field) in tags.items()}\n                        vals[\"file\"] = file\n                        exif_stream.append(vals)\n                    file_count += 1\n                    if file_count % check_point == 0:\n                        save(exif_stream,csv_file)\n            if len(exif_stream) > 0:\n                save(exif_stream,csv_file) \n        print(\"Checks processed: {}\".format(file_count))\n    except OSError as err:\n        print(\"OS error: {0}\".format(err))\n    except KeyError as err:\n        print(\"KeyError: {} - check that the folder is correct\".format(err))\n    except Exception as err:\n        print(err)\n\n# change this to the folder mount point\nmount = os.getcwd()\n# change this to the checks folder\npath_to_checks = 'sample_checks_prosper'\nfolder = os.path.join(mount,path_to_checks)\n# iterate over the files and stream the exif data\nstream_exif(folder,csv_file='all_checks.csv')\n",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "trusted": true,
    "collapsed": true
   },
   "cell_type": "code",
   "source": "",
   "execution_count": null,
   "outputs": []
  }
 ],
 "metadata": {
  "language_info": {
   "pygments_lexer": "ipython3",
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "nbconvert_exporter": "python",
   "name": "python",
   "version": "3.5.4",
   "file_extension": ".py",
   "mimetype": "text/x-python"
  },
  "kernelspec": {
   "name": "conda-env-ocr-py",
   "display_name": "Python [conda env:ocr]",
   "language": "python"
  },
  "gist_id": "709fc889bd85daa5be88fbf6d2d422da"
 },
 "nbformat": 4,
 "nbformat_minor": 0
}