Post links stored in individual JSON files to Linkding

parse-instapaper-csv.php

<?php

// Instapaper allows you to export your saved links as a CSV file
// And I turn it into my default LINKS json format
// for better or worse

$filename = 'instapaper-export.csv';
$csv = array_map('str_getcsv', file($filename));
$ignore_header_line = true;
$destination_for_link_json = './LINKS/';

// URL,Title,Selection,Folder,Timestamp,Tags

/*
 * // goal is to create json looking like this
 * tags is expected to be [] since I have no tags
 * ignore any urls which do not begin with http ane exit
 *
{
    "id": "sidebarlinks.3",
    "url": "http:\/\/www.adaptivepath.com\/publications\/essays\/archives\/000315.php",
    "tags": [
        "domain:adaptivepath.com"
    ],
    "date_added": "2004-04-07T00:00:00+00:00",
    "title": "Why Content Management Fails",
    "description": "CMSes need managers",
    "private": false,
    "via": "",
    "extra_derived": ""
}
 */

$selection_values = [];
$folder_values = [];

foreach ($csv as $row) {
    if ($row[0] == 'URL') {
        continue;
    }
    $url = $row[0];
    $title = $row[1];
    $selection = $row[2];
    $folder = $row[3] ?? '';
    $timestamp = $row[4] ?? '';
    $tags = []; // I don't have any tags at the moment


    if (strpos($url, 'http') !== 0) {
        print "Ignoring $url because it is not an http(s) url\n";
        continue;
    }

    if (!$timestamp) {
        printf("%s does not have a timestamp\n", $url);
        exit;
    }

    // After looking at the selection values, I'm not keeping them
//    if (in_array($selection, $selection_values)) {
//        // do nothing
//    } else {
//        $selection_values[] = $selection;
//    }
//
//    if (in_array($folder, $folder_values)) {
//        // do nothing
//    } else {
//        $folder_values[] = $folder;
//    }


        if ($folder == 'Starred') {
            $folder = 'instapaper:starred';
        } elseif ($folder == 'Editor’s Picks') {
            $folder = 'instapaper:editors-picks';
        }
        // get the domain of the url and strip any 'www.' from the beginning and add a tag
        // that looks like domain:domain
        $domain = parse_url($url, PHP_URL_HOST);
        $domain = str_replace('www.', '', $domain);
        $tags[] = "domain:$domain";
        $tags[] = 'via:instapaper'; // to commemorate the source of the import

        // save epoch timestamp as an iso 1806 timestamp
        $date_added = date('c', $timestamp);

        $id = 'instapaper' . md5($url);
//    {
//        "id": "sidebarlinks.3",
//    "url": "http:\/\/www.adaptivepath.com\/publications\/essays\/archives\/000315.php",
//    "tags": [
//        "domain:adaptivepath.com"
//    ],
//    "date_added": "2004-04-07T00:00:00+00:00",
//    "title": "Why Content Management Fails",
//    "description": "CMSes need managers",
//    "private": false,
//    "via": "",
//    "extra_derived": ""
//}
        $link_data = [
            'id' => $id,
            'url' => $url,
            'tags' => $tags,
            'date_added' => $date_added,
            'title' => $title,
            'description' => '',
            'private' => false,
            'via' => '',
            'extra_derived' => ''
        ];
        // save it to a json file in the LINKS directory
        $json_filename = $destination_for_link_json . $id . '.json';
        file_put_contents($json_filename, json_encode($link_data, JSON_PRETTY_PRINT));
        print "Saved $json_filename\n";

}
// Folder values were (
//    [0] => Unread
//    [1] => Archive
//    [2] => Starred
//    [3] => Editor’s Picks
//) which I will save as instapaper:starred and instapaper:editors-picks but ignore the others
//print "Selection values\n";
//print_r($selection_values);
//print "Folder values\n";
//print_r($folder_values);

post-links-to-linkding.php

<?php

// https://github.com/sissbruecker/linkding
// https://linkding.link/api/
// Author: Joe Crawford https://artlung.com/
// * Linkding is a bookmark tool I use to store links
// * I exported all my prior links to JSON files
// * Note that Linkding allows for import of a Netscape bookmarks file, which might be easier
// * It looks like submitting date_added values are not respected
// * My other stored metadata I put into thenotes field in linkding

/*
{
    "id": "sidebarlinks.24327",
    "url": "http:\/\/tantek.pbworks.com\/w\/page\/19402872\/CassisProject",
    "tags": [
    "php",
    "javascript",
    "programming",
    "domain:tantek.pbworks.com"
],
    "date_added": "2010-11-20T12:44:00+00:00",
    "title": "tantek \/ CassisProject",
    "description": "javascript and php",
    "private": false,
    "via": null,
    "extra_derived": "hash:9104bf0bf0fdc466dca3d4b692d3f4cf|meta:b2a7bfe14a384bfffde15e86f29e69f4"
}
*/

// After submitting, the payload returned from POST /api/bookmarks/ looks like
/*
{
  "id": 941,
  "url": "http://tantek.pbworks.com/w/page/19402872/CassisProject",
  "title": "tantek / CassisProject",
  "description": "javascript and php",
  "notes": "hash:9104bf0bf0fdc466dca3d4b692d3f4cf|meta:b2a7bfe14a384bfffde15e86f29e69f4",
  "web_archive_snapshot_url": "",
  "favicon_url": null,
  "preview_image_url": null,
  "is_archived": false,
  "unread": false,
  "shared": true,
  "tag_names": [
    "domain:tantek.pbworks.com",
    "javascript",
    "php",
    "programming"
  ],
  "date_added": "2024-10-30T17:36:26.689901Z",
  "date_modified": "2024-10-31T04:24:45.961427Z",
  "website_title": null,
  "website_description": null
}
 */


// one directory for json source
// one directory for json results
$source_dir = './LINKS/';
$rest_result_dir = './REST_RESULTS/';

$linkding_instance = 'https://example.com';
$linkding_token = '';

if ($linkding_token === '') {
    printf("No token found, set \$linkding_token\n");
    exit(1);
}

// documentation: https://linkding.link/api/
// Header looks like
// Authorization: Token <Token>
// Methods
// GET /api/bookmarks/
// GET /api/bookmarks/archived/
// GET /api/bookmarks/<id>/
// GET /api/bookmarks/<id>/
// GET /api/bookmarks/check/?url=https%3A%2F%2Fexample.com

// To create a link
// POST /api/bookmarks/
/* payload
{
  "url": "https://example.com",
  "title": "Example title",
  "description": "Example description",
  "notes": "Example notes",
  "is_archived": false,
  "unread": false,
  "shared": false,
  "tag_names": [
    "tag1",
    "tag2"
  ]
}
*/

// To update a link
//PUT /api/bookmarks/<id>/
//PATCH /api/bookmarks/<id>/

// read source directory for only json files

$filenames = scandir($source_dir);

// do one at a time
foreach ($filenames as $filename) {
    $source_file = $source_dir . $filename;
    $rest_result_file = $rest_result_dir . $filename;

    // if filename does not end with json, bail out
    if (!preg_match('/\.json$/', $filename)) {
        continue;
    }

    if (file_exists($source_file) && !file_exists($rest_result_file)) {
        $source = file_get_contents($source_file);
        $json = json_decode($source, true);

        printf("%s has not been submitted to the REST API\n", $filename);

        $payload = [
            'url' => $json['url'],
            'title' => $json['title'],
            'description' => $json['description'],
            'notes' => $json['extra_derived'],
            'is_archived' => false,
            'unread' => false,
            'shared' => !$json['private'],
            'tag_names' => $json['tags'],
            'date_added' => $json['date_added'],
        ];

        $payload_json = json_encode($payload);
        $auth_header = 'Authorization';
        $auth = 'Token ' . $linkding_token;
        $method = 'POST';
        $api_endpoint = $linkding_instance;
        $api_method = '/api/bookmarks/';

        $url = $api_endpoint . $api_method;
        // post to the REST API
        $ch = curl_init($url);
        curl_setopt($ch, CURLOPT_CUSTOMREQUEST, $method);
        curl_setopt($ch, CURLOPT_POSTFIELDS, $payload_json);
        curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
        curl_setopt($ch, CURLOPT_HTTPHEADER, array(
            'Content-Type: application/json',
            'Content-Length: ' . strlen($payload_json),
            $auth_header . ': ' . $auth
        ));

        // save the results to a file in REST_RESULTS
        $result = curl_exec($ch);

        $http_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
        // if it's good, save the resulting json file
        if ($http_code == 200 || $http_code == 201 || $http_code == 204) {
            file_put_contents($rest_result_file, $result);
            printf("Submitted %s to the REST API\n", $filename);
        } else {
            printf("Error submitting %s to the REST API\n", $filename);
            printf("HTTP Code: %d\n", $http_code);
            printf("Result: %s\n", $result);
        }

    } else {
        printf("%s has already been submitted to the REST API\n", $filename);
    }
}

progress.php

<?php

$source_dir = './LINKS/';
$rest_result_dir = './REST_RESULTS/';

$filenames = scandir($source_dir);

// do this no more than 10 times

$count = 0;
foreach ($filenames as $filename) {

    // must end in json
    if (!preg_match('/^\d+.*\.json$/', $filename)) {
        continue;
    }


    $has_been_submitted = false;
    if (file_exists($rest_result_dir . $filename)) {
        continue;
    }

    printf("%s has not been submitted to the REST API\n", $filename);
    $count++;

    if ($count > 30) {
        break;
    }

}

// how many files with .json at the end are in each directory

$count_of_source = 0;
$count_of_rest = 0;

$source_files = glob($source_dir . '*.json');
$rest_files = glob($rest_result_dir . '*.json');

printf("There are %d files in %s\n", count($source_files), $source_dir);
printf("There are %d files in %s\n", count($rest_files), $rest_result_dir);

// how many are left to go?

$source_files = glob($source_dir . '*.json');
$rest_files = glob($rest_result_dir . '*.json');

$source_files = array_map(function ($f) {
    return basename($f);
}, $source_files);

$rest_files = array_map(function ($f) {
    return basename($f);
}, $rest_files);

$diff = array_diff($source_files, $rest_files);

printf("There are %d files left to go\n", count($diff));