Utility ~ ACC2GED (CSV to GEDCOM) Readme Release Notes

Original Release Notes

ACC2GED v1.0 - 13th August, 1998.

Copyright © D.J.Cooke, 1998. All rights reserved.

Despite its name, this program produces a GEDCOM file not directly from an Access database, but from delimited text files, in which format data may be exported from Access and other database programs. It was written originally for my own use, and its design reflects the quirks of my own database. It is, however, configurable, to allow some flexibility in the format of the input data, and I make it available in case anyone else might find it useful.

Where data cannot be directly matched to the program's configuration options, it should be possible to produce data in an appropriate format by running queries. In the case of Access, the results of a query can be exported to an ascii text file using the TransferText macro action (for brief information, see http://office.microsoft.com/en-us/access-help/transfertext-macro-action-HA001226310.aspx). However, a crude but simple method for Access is to use a query to make a new table, exporting this directly to a new, empty and temporary database created for the purpose, from which the table can be exported as a delimited text file.

This version of the program does not offer an altogether comprehensive implementation, in particular its handling of notes fields is limited, but it is free for personal use.

This package is provided under the following conditions:

  • The acc2ged zip package may be freely copied and distributed provided that the package is complete and unmodified, and subject to acceptance of these conditions. The package comprises the following files:
          acc2ged.exe       34,476 bytes
          readme.txt        12,377 bytes
          config.one         1,518 bytes
          config.two         1,414 bytes
          births.one         2,809 bytes
          baptisms.one       5,899 bytes
          missing.one        3,859 bytes
          mar1.one           3,908 bytes
          mar2.one             964 bytes
          deaths.one         2,578 bytes
          burials.one          400 bytes
          individu.two      23,348 bytes
          family.two         1,329 bytes
          children.two       1,646 bytes
  • The package may not be sold. Any charge made for distributing the package on diskette, CD-ROM or other media must be in respect of materials/handling only.
  • The program is free for personal use only. Use of the program for commercial purposes, or its distribution with commercial software, is prohibited.
  • The user accepts sole responsibility for determining the fitness for use of the program. The author accepts no liability for any failure of the program to perform as expected, or for any loss or damages of any kind which might result from its use.


The only essential files are the executable file acc2ged.exe and the configuration file, and these may simply be copied into any suitable directory. The configuration file must be named acc2ged.cfg, and must be in the working directory when the program is run.

Acc2ged is an msdos program, which must be run by typing "acc2ged" (without the quotation marks) at the dos prompt. There are some command-line options, and information about these can be obtained by typing "acc2ged ?". For the most part, however, these have been superseded by the configuration file.

The two files config.one and config.two are copies of the configuration file set up to process the sample data comprised in the two sets of files with extensions .one and .two. These are delimited text files, and are provided merely to demonstrate the program's use.

If the file config.two is renamed acc2ged.cfg, and the program is run with the .two files in the same directory, it will produce a GEDCOM file from the data they contain. If, alternatively, the file config.one is renamed acc2ged.cfg, and the .one files are in the working directory, an identical GEDCOM file can be produced from the .one files by typing "acc2ged -y" at the dos prompt (the '-y' switch simply inserts 'ABT' before year-only dates).


The program can be configured to take input from up to 10 source files, allowing some flexibility in the format of the data which can be processed. Note that the format of the configuration file must not be altered, and that the program cannot itself produce a new copy of this file.

The first 10 sections of the configuration file specify all the fieldnames for the input data, and the files in which they are to be found. All filenames and fieldnames must be entered in quotation marks ("). Note that 'FID' is used to represent 'Family ID'. It is not necessary, and may not be desirable, that a filename be specified in every section, provided that every source file is specified in the appropriate place. A filename must be specified in at least one of the [personal], [births] and [baptisms] sections, and in at least one of the sections [marriages1] and [marriages2]. Fieldnames must, however, be given in every relevant section, and repeated if necessary.

If a filename is specified in the [personal] data section, only individuals who appear in this file will appear in the GEDCOM file. In this case, if the filenames in any of the sections [births], [baptisms], [deaths] and [burials] are left blank, the program will look for the data in the [personal] file.

If the filename in the [personal] section is left blank. the program will look for the data in the [births], [baptisms] and [missing] files.

If the [FAMC] filename is left blank, the data will be looked for in the [personal] file, or if this is blank, in the [births], [baptisms] and [missing] files.

If the [FAMS] filename is left blank, then if the ID_male and ID_female fieldnames are different, the program will look in the [marriages1] and [marriages2] files. If these 2 fieldnames are the same, it will look, as for FAMC, in the [personal] file, of if this is left blank, in [births] etc. In this case, the program will look for additional FAMS references in the notes fields. These must take the form xFAMSxxx, where x is a digit, and the initial digit specifies the number of the marriage, first, second, third etc. e.g. 2FAMS34.

Actually, there is no notes field specified in the [personal] section. The program always looks for a notes field as specified in the [births] or [baptisms] sections.

If the logic behind all this appears elusive, the two sample configuration files provide guidance. The most important point to note is that specifying a filename in a data section may cause the program to look exclusively in that file for the corresponding data, whilst leaving the filename blank may allow that data to be split between two files or more.

In each section, the fieldnames establish the relation between the data elements in a particular file, without any implication that different sections necessarily pertain to different files; and all relevant fieldnames must be shown. Thus, for example, if the FAMC data appears in the [personal] file, the ID fieldname specified in the [personal] section must be specified again in the [FAMC] section, along with the FAMC fieldname.

The 10 data sections are followed by the [settings] section, which specifies output filenames, some options, and the values of some parameters. The first 5 entries require a string to be specified, which must be enclosed in quotation marks ("), the next 3 require yes or no to be specified, without quotation marks, and the last four require an integer to be specified.


selection This specifies a fieldname in the [personal], [births],[baptisms] and [missing] files, which allows the program to be selective of the records it processes. The field should be a single character in width.
select_charThis specifies, in the form of a string, the characters which may appear in the selection field if the program is to process the data. If, in a given data record, the single character appearing in the selection field does not match any of the characters in this string, the data in that record will be ignored.
baptismtagEither CHR (the default), or BAPM.
outputfileThe name and path of the .ged file to be produced.
logfileThe name and path of the log file. The program produces a log file and there is no option to dispense with it. If none is specified, the default is "family.log".
show_progresEither yes or no: if yes, the program displays the number first of individuals then of families processed, as the program runs.
idscanEither yes or no: if yes, as each individual is processed, a list is scanned of all the ids so far processed, to avoid duplication.
show_sourcedata Either yes or no: if yes, everything in the configuration file after the [sourcedata] heading, blank lines excepted, is printed in the GEDCOM file after the 0 HEAD tag.
max_fieldsSpecifies the maximum number of data fields in any file,default 30.
max_marrSpecifies the maximum number of marriages allowed for any individual, default 4.
max_indiThe maximum number of individuals who can be processed and appear in the .ged file, default 2000.
max_famThe maximum number of families which can be processed,default 400.

If either of these last two parameters is found too small when the program is running, it will terminate with advice to increase one or other. The maximum number of children in any family is also limited, to 20, and there is no option to increase it.

The [sourcedata] section, in contrast to all the preceding sections, may be edited freely, provided that the heading is preserved.


The source file must present the data in the form of delimited text files. Fields must be separated by a comma (,), text must be delimited by quotation marks ("), and the fieldnames must be stored in the first line. If dates are detected in the form xx/yy/zzzz, this will be assumed to represent DD/MM/YYYY, and will be converted by the program. Leading zeros must be present, along with four-digit years: e.g. 24/08/1883. If dates are not in this form, they will be preserved unaltered. ID references must be numerical: if additional characters are present, only digits, and specifically the first block of digits, will be extracted. Individuals will appear in the output .ged file only if an associated family reference, FAMS or FAMC, can be identified.

The program cannot handle a separate file for placenames. If, therefore, placenames are held in a separate table in the source database, it will be necessary to run queries before exporting the data.


The program produces an obligatory log file. This lists firstly data records representing individuals who do not appear in the GEDCOM file because no FAMC or FAMS references were found, secondly families for which no marriage data is recorded, and thirdly families for which one or other spouse is absent. Finally it records the number of FAMC and of CHIL references, and in the event of any discrepancy, identifies the aberrant ones.

This program is provided "as is", without any warranty, express or implied, and without support. If any bugs are reported to me, I will probably fix them, but I give no undertaking to do so, nor to enter into correspondence of any kind regarding this program. However, let me know if you find it useful!

D.J.Cooke, 113730.150@compuserve.com