data-entry-tool.py

import os, json, csv


with open('new-doc-fields.csv', newline='') as csvfile:
    newfields = list(csv.reader(csvfile, delimiter=','))

os.chdir('./data-dir')

entrycount = 0
docsHiddenCount = 0

validCategories = ["verified", "monitored", "custom", "specialized"]

for root, dirs, files in os.walk("."):
    for file in files:
        change = False
        if file.endswith(".json"):
            try:
                f = open(os.path.join(root, file), "r", encoding="utf-8")

                # Dealing with escapes; doing it wrong
                x = f.read().encode("utf-8").replace(b'\u', b'\\\\u').decode("utf-8")

                j = json.loads(x)

            except json.decoder.JSONDecodeError:
                print("Failed to parse JSON file: " + os.path.join(root, file))

            # Find the "marketing" entry and update it with values from the CSV
            if "marketing" in j and "status" in j and j["status"] != "dead":

                pair = j["marketing"]["path"]
                network = root.split("/")[1]

                for row in newfields:
                    if (row[0] == pair and
                    row[1] == network and
                    row[9].split(" ")[1].lower() in validCategories):

                        new_docs = {
                            "feedCategory": row[9].split(" ")[1].lower(),
                            "feedType": row[4],
                            "assetName": row[2]
                        }

                        if "docs" in j:
                            j["docs"]["feedCategory"] = (new_docs["feedCategory"])
                            j["docs"]["feedType"] = (new_docs["feedType"])
                            j["docs"]["assetName"] = (new_docs["assetName"])
                            entrycount += 1
                            change = True
                        else:
                            j["docs"] = new_docs
                            entrycount += 1
                            change = True
            elif "marketing" in j and "status" in j:
                print("Ignored " + j["marketing"]["path"] + " because status is " + j["status"])

            # If "docsHidden" exists, move it and its current value to "docs"
            if "docsHidden" in j:
                try:
                    j["docs"]["hidden"] = j["docsHidden"]
                    j.pop("docsHidden")
                    docsHiddenCount += 1
                    change = True
                except KeyError:
                    j["docs"] = {}
                    j["docs"]["hidden"] = j["docsHidden"]
                    j.pop("docsHidden")
                    docsHiddenCount += 1
                    change = True

            # Write the file again alphabetically and restore newline
            if change == True:
                with open(os.path.join(root, file), "w") as write_file:
                    out = json.dumps(j, sort_keys=True, indent=2, ensure_ascii=False)

                    # Restore the escapes again
                    x = out.encode("utf-8").replace(b'\\\\u', b'\u').decode("utf-8")

                    write_file.write(x)
                    write_file.write("\n")

                    # Old JSON dump before fixing escapes
                    #json.dump(j, write_file, sort_keys=True, indent=2, ensure_ascii=False)

print("Entries created/updated: " + str(entrycount))
print("docsHidden entries moved: " + str(docsHiddenCount))
print("Done")