case-preserving LDIF reader/parser/stream

Question

I have to read an LDIF file, split the entries, modify them and then write the result as LDIF files.

I found an LdifReader at the Apache Directory LDAP API (org.apache.directory.api:api-ldap-client-api), so I tried that one like this:

Stream<LdifEntry> stream = StreamSupport.stream(reader.spliterator(), false);
Predicate<LdifEntry> isEnabled = entry -> entry.get("pwdAccountLockedTime") == null;
Map<Boolean, List<LdifEntry>> parts = stream.collect(Collectors.partitioningBy(isEnabled));
List<LdifEntry> enabledAccounts = parts.get(true);
List<LdifEntry> disabledAccounts = parts.get(false);

Works well. However, for some reason all the attribute names/ids are lowercased (“pwdAccountLockedTime” becomes “pwdaccountlockedtime”, etc.) but I need them as they are (preserving the case) – in order to keep the same human readability as before.

Any idea how to do that? If needed I will use a different library.

Note: I would like to improve my question, because I got some downvotes. Please tell me, what’s wrong with it or what’s missing.

Answer 1

I was able to solve my problem by replacing the library with org.springframework.ldap:spring-ldap-ldif-core and writing a little helper.

public class LdifUtils {
    /**
     * Reads an LDIF file and returns its entries as a collection 
     * of <code>LdapAttributes</code> (LDAP entries).
     * <br>
     * Note: This method is not for huge files, 
     * as the content is loaded completely into memory.
     *
     * @param pathToLdifFile the <code>Path</code> to the LDAP Data Interchange Format file
     * @return a <code>Collection</code> of <code>LdapAttributes</code> (LDAP entries)
     * @throws IOException if reading the file fails
     */
    public static Collection<LdapAttributes> read(final Path pathToLdifFile) throws IOException {
        final LdifParser ldifParser = new LdifParser(pathToLdifFile.toFile());
        ldifParser.open();
        final Collection<LdapAttributes> c = new LinkedList<>();
        while (ldifParser.hasMoreRecords()){
            c.add(ldifParser.getRecord());
        }
        ldifParser.close();
        return c;
    }
}

usage like before ...

final Stream<LdapAttributes> stream = LdifUtils.read(path).stream();
final Predicate<LdapAttributes> isEnabled = entry -> entry.get("pwdAccountLockedTime") == null;
final Map<Boolean, List<LdapAttributes>> parts = stream.collect(Collectors.partitioningBy(isEnabled));
final List<LdapAttributes> enabledAccounts = parts.get(true);
final List<LdapAttributes> disabledAccounts = parts.get(false);
logger.info("enabled accounts: " + enabledAccounts.size());
logger.info("disabled accounts: " + disabledAccounts.size());

Note: I would like to improve my answer, because I got some downvotes. Please tell me, what’s wrong with it or what’s missing.