Main dialog box, import and export
The Switchboard
Introduction to the main DeltaAccess dialog box
Importing data
Which DELTA files should be imported?
What happened to the character name during import?
What happened to my units?
Importing subsets (include/exclude items/characters)
Exporting data
Creating natural language item descriptions
Creating item descriptions as HTML forms
Next table of contents: Project management
Previous table of contents 
Main table of contents
DeltaAccess provides a switchboard dialog box (form 'Switchboard'), which can be used instead of the database explorer(1) to open the other forms and dialog boxes. You can always reach the switchboard using the 
-button in the tool bar.
On the left side of the dialog box you can select the task you want to do. On the right side you can optionally choose a project name of an existing descriptor project. Some tasks ignore the project name, e. g., importing DELTA text files into a new project. If the project name is required for a task and you have not selected a descriptor project, you will be prompted to select a project later.
When you select a task, a short description will be displayed at the bottom of the dialog box. Similarly, when you select a project, the project title, project authors, and copyright (see Editing project properties), will be displayed in the same space.
Double-clicking on either a task or a project name will directly start the selected task with the selected project.
Note: DeltaAccess currently does not use any true menus (= the entries immediately below the title bar of the application window). The menus you see are the unchanged Microsoft Access menus. So far I found it too much work to use menus, since Microsoft has changed the programming of menus in every version of Access. I may introduce menus for Microsoft Access97 in a later version of DeltaAccess.
The Switchboard form is necessary if you run DeltaAccess under a run-time version of Microsoft Access. No database explorer is available in the run-time version. You will quit DeltaAccess completely, if you close the switchboard under the run-time version.
The project name identifies your descriptor project in DeltaAccess. You can keep multiple descriptor projects in a single version of DeltaAccess, e.g. import several DELTA data sets. Descriptor projects can be either entirely independent, or they can be
linked to each other. The project name is also used as the prefix for table names and editing forms.A project name can contain 1 to 20 characters, and may not contain the following special characters "-", ".", "!", "[", "]", "`", and " " (= blank).
Introduction to the main DeltaAccess dialog box
|
|
Who should read this: This chapter is relevant only if you open the form named DeltaAccess manually from the database explorer. It does not apply if you normally use the DeltaAccess Switchboard. You will then use the same views of the dialog box, but they will appear as independent dialog boxes. |
The form DeltaAccess contains the import, management, analysis, and export dialog boxes. If you open it manually from the database explorer, you can select on the right hand side of the dialog box whether you want to:
import data from a set of DELTA files into a DeltaAccess descriptor project,
manage projects (create, copy, link, or delete descriptor projects, etc.),
analyze or edit data, or
export data from the tables of a descriptor project to a DELTA text file of other formats.
The left part of the dialog box will display the appropriate controls for each option. Common to all options is the setting of a project name, which identifies your descriptor project in DeltaAccess. The 4 different options of the DeltaAccess dialog box are described in the following sections:
Importing data
Managing projects
Analyzing data
Exporting data
|
|
Who should read this: Users who already worked with DELTA and want to import one or several existing projects using DELTA coded text files as a data interchange format. If you do not have DELTA files to import, please continue with the next major section Manage projects. |
DeltaAccess works with multiple data sets called a project. Each project is equivalent to a DELTA file in Pankey/Pandora, or a set of files (items, specs, chars) in Confor. To import data you must enter a project name. If you have previously used Confor, you probably have separate directories for each DELTA data set, so the name of the directory would be a good choice for the project name.
If you are using Pankey, your data are probably in a single file. Enter the name of this file under Import file name. You can click on the button with the open-folder-symbol (
) to select the files using a dialog box.
If your data set consists of multiple files (e.g. Items, Chars, and Specs used by Dallwitz's Confor) you should select any file (e.g. Chars). The checkbox Import all "Confor" files: Chars, Specs, Items, etc. is automatically activated in this case. This causes DeltaAccess to search for the commonly used files. See the following chapter Which DELTA files should be imported? for more information.
If your import file contains Include/Exclude Character/Items directives, the complete set of data is imported, and a separate subset honoring these directives is created. Consult the chapter Importing subsets for information on how to import subsets separately.
The following options are available for import:
The import process has several limitations: Neither line numbers nor the "data compression" option are allowed. You can use M. Dallwitz's Confor program to remove these features. Empty comments ('<>') are removed, i. e. they will be missing when the data are later exported back to DELTA file format.
Some problems arise with the mixed use of OR (/), AND (&), and TO (-) for character states of the same character, and with the use of multiple numerical entities within a single character. (Multiple statistics like mean, median, s. d., s. e., etc., for one entity are no problem, but multiple minima or maxima are a problem.) See the chapter Limitations for a comprehensive overview of supported directives and problems that may arise during import.
Note that there is no such thing as a unique character name in DELTA. The DELTA character string as used by Confor identifies the character and provides natural language output as well. It may consist of a comment alone or be empty altogether. In contrast, DeltaAccess requires unique character names. It tries to use the information from the input file, and appends random numbers to create unique names if necessary. Consult the chapters What happens to the character name during import? and What happened to my units? for further details.
As a result of these limitations you might receive several warning messages. A typical warning message is "A random number was appended to the following duplicate values to achieve unique names in the field 'CharName': blades structure (CID=36.); scape (CID=60.)". DELTA does not require characters to have a unique name, but DeltaAccess does. The warning message means that two characters had the name "blades structure" and two other characters had the name "scape". The character names of the characters with number 36 and 60 were automatically modified so that all characters have unique names.
You can later check the warning messages again in the _PROPERTY table property OriginalImportWarnings. Other information stored as properties are:
See Editing project properties for further information about how to view and edit project properties.
You can not import character definition or item data separately, e. g., to overwrite the item data, but preserve the character and item definition. For a method to accomplish this task nevertheless, see the overwrite action of the management dialog box.
Introduction to the DeltaAccess dialog box Main table of contents
Which DELTA files should be imported?
If you are using Richard Pankhurst's Pankey all your data are probably in a single file. Import this file.
If you are using Mike Dallwitz's Confor or Eric Gouda's TAXASOFT, you are probably using multiple files. The main files are usually called Items, Chars, and Specs. Please note that some information important to DeltaAccess is usually not present in the 3 main files, but in additional 'batch' files, e. g. ToInt, ToKey, ToNat, or Intkey.ini. DELTA, as used by the Confor program, does not distinguish between data directives and program command directives. Some directives, which could well be viewed as formatting commands, are treated as data by DeltaAccess. For example, the Character Reliabilities and Key States directives might be available only in the ToKey command batch file, and the New Paragraphs, Item Subheadings, and Link Characters directives only in the ToNat command batch file. See the List of supported directives for information about which directives can be imported by the current version of DeltaAccess.
DeltaAccess will automatically search for these files if you activate the checkbox Import all "Confor" files: Chars, Specs, Items, etc. The following files will automatically be imported: 'Chars', 'Specs', 'Items', 'Extra', 'CNotes', 'CImages', 'TImages', 'ToInt', 'ToNat', 'ToKey', 'ToDis', and 'Intkey.ini'. 'Extra' is an additional file used by DeltaAccess for its own directives and is present only if the data set was previously exported by DeltaAccess. The only information imported from intkey.ini is currently the Define characters directive. If you use additional files, DeltaAccess will automatically include them as well, as long as they are referenced by an INPUT FILE directives in one or several of the files listed above.
DeltaAccess is not sensitive to multiple appearances of the same directive. For most directives it will simply import the first version of a directive.
You can import one set of Include/Exclude Character/Items directives together with your data. Further sets can be imported in separate files, see the chapter Importing subsets.
Importing data Main table of contents
What happened to the character name during import?
|
|
Who should read this: Users who are experienced with DELTA and want to understand how DeltaAccess modifies the data during the import process. |
Because DELTA mixes the functions of character name (unique identifier for a character, suitable for data entry, retrieval and interactive identification) and character phrase (optimized for natural language output), some extra processing is necessary during import. First, if a unit-string is present, it is removed and written to the separate Unit attribute.
The resulting string is written directly to CharPhrase. This is the only place, where the DELTA comment brackets "<>" are preserved in DeltaAccess. You might want to use them for some special formatting in other DELTA programs like Pankey, TAXASOFT,or Confor.
The CharName attribute is constructed from CharPhrase, by appending the notes without brackets. DeltaAccess requires the name to be unique. If the same name occurs multiple times, a random number must be appended to the second and further occurrences of the same name. Usually it will be advisable to manually edit the CharName attribute after import, to create short and precise character names for data entry and identification.
If CharName, CharPhrase, or Unit are longer than 255 characters, they will be truncated to 255 character. You will be warned about this in the message displayed at the end of the import process.
Importing data Main table of contents
|
|
Who should read this: Users who are experienced with DELTA and want to understand how DeltaAccess modifies the data during the import process. |
After the character name in a DELTA character list, an optional measurement Unit can be placed, like "mm" in the following example:
#6. length/ mm/
Units for numerical characters are fully supported by DeltaAccess (please inform me if you are missing them in some report or output, I may have forgotten to include them there).
Units for text characters are not defined in DELTA and make no sense. If DeltaAccess detects them in an import file, they are preserved, but never used.
Problematic are units after categorical characters (UM/OM). They are not officially allowed in DELTA version 3, but are frequently used, esp. by users of TAXASOFT. They are used for repeating phrases after the character states. This is considerably more difficult to handle in a database, because the descriptive string for a character state must be assembled each time it is used, e. g., during data entry or identification. I see no reason why the text entered under Unit should not be at the end of each character state, even if this means pasting some text into several states. I also consider this a better solution for multi-language data sets (i.e., several languages concurrently available in a single descriptor project) because languages will differ in their ability to use a common end phrase for a set of character states.
The answer to the question "What happened to my units?" in the case of categorical characters is therefore: They have been appended to the end of the character state name during the import procedure.
Importing data Main table of contents
Importing subsets (include/exclude items/characters)
When the import process detects one of the following directives:
Include Items,
Exclude Items,
Include Characters, or
Exclude Characters
it will create an Item subset, a Character subset, or both. If the DELTA file contains data (i.e. at least the character list and item description directives) as well as these directives, the subset is automatically based on the imported data. The name entered under project name is used for the main project, and a name for the additional subset is requested.
If the imported DELTA text file contains no data directives, you will be asked upon which existing project the subset should be based. The name entered under project name is now used for the subset project. Thus, to import several predefined subsets, and link them to existing projects, you should import these directives alone. A file to import an additional subset definition could, for example, contain just the following lines:
*EXCLUDE CHARACTERS 1 4-7 71-96
*INCLUDE ITEMS 1-120 310-420
Avoid importing subsets using batch files, which contain references to the Items, Chars, and Specs files in addition to the include/exclude directives. Importing such files will import the data independently for each subset. Although you will have successfully imported the subset for analysis functions, you will not be able to edit the underlying data for all subsets simultaneously, because multiple copies will exist. Create copies with the include/exclude directives alone, and import these files into DeltaAccess.
Importing data Main table of contents
After you have selected the main option Export to file from the pick list, you must select the project name, the name of the file into which the data shall be exported, and an export format. Available export formats are:
All formats use the period (".") as decimal separator in numbers, to be internationally compatible. If your national settings are different and you experience problems importing, e. g., a CSV file into MS Excel, you should temporarily change the settings in the Windows control panel, applet International, to a period and change it back after import.
If you choose the ASCII character set, ANSI characters are converted into their (IBM/DOS) equivalents where possible. This option is applicable only to some formats.
If you select DELTA text file, you can choose whether to Export into a single file (preferred by DeltaAccess and Pankey) or whether to use separate multiple files for ITEMS, CHARS, and SPECS. If you enter ITEMS as a character name, the other files will be named CHARS and SPECS, otherwise the extensions '.CHARS' and '.SPECS' are appended to the filename you have entered. A fourth file named 'EXTRA' is used to export the additional directives defined by DeltaAccess, see also the following chapter Variants of the DELTA format.
Related topics: To export a formatted report of the current character definition, use the tabular character definition report and export it to Word (see Office Links).
Introduction to the DeltaAccess dialog box Main table of contents
Since there are already many variant DELTA dialects, files exported by DeltaAccess will generally differ from the import files. With few exceptions, the data will be identical, but the formatting may be more difficult to read, e.g. because no range notation ("1-3") is used in the output. Also, you may have to adjust the wording of the directive, if your preferred tool does not support the directive name I use for output. I am open to criticism about which directive names should be preferred.
In many ways, DeltaAccess provides features that are not supported in standard DELTA. To improve the usability of the DELTA format for export and reimport into DeltaAccess, additional DeltaAccess-directives have been defined. Unfortunately, M. Dallwitz's Confor aborts whenever it encounters a directive which has not been defined by Mike Dallwitz and his coworkers. This is a violation of the DELTA standard, which explicitly states that programs should skip any unknown directives which they encounter.
To solve this problem, two solutions are provided: - 1. If you select the multiple files option, the special DeltaAccess directives will be written to a separate file named EXTRA. Since Confor normally does not read this file, the data remain compatible with Confor. - 2. If you want your data in a single file, you can select the export option Strict DELTA. DELTA coded files created using this option include only directives defined by Confor itself.
Currently the export of additional features of DeltaAccess back to DELTA files is still under development, so that not all DeltaAccess features will be preserved when a descriptor project is exported to DELTA and re-imported into DeltaAccess. For this reason, the preferred format for backup copies of a project should be the internal backup in MDB format, see the chapter Backup and restore for further details.
Notes on specific problems: The DELTA character "name" is imported into the CharPhrase attribute, and if a CharPhrase is present it is preferred during export over the CharName attribute. The Notes originally extracted from ItemName comments are exported as a separate directive. The LitRef/CollUnit attributes are not yet exported. Modifiers are added back as item description notes, but slight changes compared with the original description may occur.
DeltaAccess allows the character numbers (CID) to be non-continuous (e.g. 1, 2, 4, 5). Many other DELTA compatible programs including Confor will not import such a data set. The CID numbers need to be renumbered, e.g. using the auto-resequence utility of the Reorganize form. Since this would permanently change the project that is exported, it is not done automatically, but an appropriate warning will be issued upon trying to export a descriptor project with non-continuous character numbers to DELTA.
TAXASOFT style comments (before and after the character state) are imported using the pipe symbol '|'. This is resolved during export to natural language by DeltaAccess, but not when exporting to DELTA files, since the TAXASOFT style is incompatible with other DELTA programs. Instead you will find entries like "8,1<bright yellow- | and often white-cretaceous> " " in your export file.
Compare also the chapter Limitations and the table of recognized DELTA directives.
Introduction to the DeltaAccess dialog box Main table of contents 
Next
