[EDOT] Add new object EDOT #682

i506210 · 2025-01-28T09:17:02Z

No description provided.

Markus1812

Thanks for your pull request. This pull request still contains both, EDOI and EDOT (with the .DS_Store). Please remove all the unwanted files here.

Please note, that for the title and description, we use title case and sentence case respectively. I've not marked every occurrence but if you could please adjust that.

"! <p class="shorttext">Interface Version</p>    "<- title (title case)
"! Interface version                             "<- description (sentence case)

Markus1812 · 2025-01-28T14:12:45Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">eDocument Type</p>
+      "! eDoc Type
+      "! $required
+      edocument_type               TYPE ty_edoc_type,


What values can edocument_type have? Is this a fixed set of values?

This is the main object name, created by Developer during design time with new names

So this is the actual object's name also stored in TADIR, right?

If this is the case, please remove it. The object name is stored in the filename like "my_object_name.edot.json"

corrected; removed object name and description from general information

Markus1812 · 2025-01-28T14:13:43Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">Description</p>
+      "! eDocument Type Description
+      "! $required
+      edoc_type_desc               TYPE ty_short_description,


Same question as in EDOI, is this the description of the EDOT object? I see this is only 30 characters, but how does this differ from the description that is contained in ty_main->header?

Do we need to use ty_main->header for description or we can create new field in the structure?

The description can be max of 30 characters as designed by Database table when it was created

The fields for name, description, and original language (as seen in the creation wizard) should not be defined by you in the AFF.
The name of an object (that is entered in the creation wizard and used to uniquely identify your object) is saved in the filename on deserialization. When looking at the server-driven editor, the name is shown in the title section and the repository browser.
The fields description and original language are persisted in the ty_main->header structure, that every AFF needs to have. This means, that you do not have to define these fields yourself.

Maybe, one small addition to Markus' comment. Even if I think 30 character description is very short, it is technically possible to define a header with a character-30-field for the description.

In this case the header must be defined within the object type in a similar way as it is done here:

abap-file-formats/file-formats/sajc/type/zif_aff_sajc_v1.intf.abap

Lines 4 to 21 in 0d3ec0c

TYPES:

"! Header for SAJC Object

"! The header for an application job catalog entry

BEGIN OF ty_header,

"! Description

"! Description of the application job catalog entry

"! $required

description TYPE c LENGTH 120,

"! Original Language

"! Original language of the application job catalog entry

"! $required

original_language TYPE sy-langu,

"! ABAP Language Version

"! ABAP language version

"! $values {@link zif_aff_types_v1.data:co_abap_language_version_cloud}

"! $default {@link zif_aff_types_v1.data:co_abap_language_version_cloud.standard}

abap_language_version TYPE zif_aff_types_v1=>ty_abap_language_version_cloud,

END OF ty_header.

Thank you for the details. I created new header with 30char length description.

Markus1812 · 2025-01-28T14:14:09Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">eDocument Type Created Using Contingency</p>
+      "! Contingency Type
+      contingency_type             TYPE c LENGTH 10,


Does this have a set of fixed values?

No, this will be another eDocument Type value which is created before

file-formats/edot/type/zif_aff_edot_v1.intf.abap

Markus1812 · 2025-01-28T14:19:09Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">File Type</p>
+      "! File Type
+      "! $required
+      file_type               TYPE c LENGTH 10,
+      "! <p class="shorttext">File Structure Type</p>
+      "! File Structure type
+      "! $required
+      file_structure_type     TYPE c LENGTH 30,
+      "! <p class="shorttext">File Description</p>
+      "! File Description type
+      file_description_type   TYPE c LENGTH 60,
+      "! <p class="shorttext">File Cloud Relevancy</p>
+      "! File Cloud Relevancy type
+      not_cloud_relevant_type TYPE abap_bool,


Why are all types ending with _type? Isn't this more like file_type, file_structure, file_description, and not_cloud_relevant?

file_type: Does this have a fixed set of values? Do you have examples for possible file types?

file_structure: Same question, do you have example values?

not_cloud_relevant: Wouldn't it be easier to understand if you flip the bool: is_cloud_relevant? The title is also File Cloud Relevancy.

file_type: Its list of entries maintained in other table EDOFILETYPE but not fixed as entries can be added in Value table. Example INBOUND, REQUEST, RESPONSE, HTML_PREVIEW etc.

file_structure: It represent the ABAP structures to be used to display the file. so that we can enable the read access log when data viewed by user.

changed from not_cloud_relevant to is_cloud_relevant

file_type: Its list of entries maintained in other table EDOFILETYPE but not fixed as entries can be added in Value table. Example INBOUND, REQUEST, RESPONSE, HTML_PREVIEW etc.

General remark for fixed values/enum values: For me, the question is not how this is technically solved (hard coded values, domain, value table, ...) but who maintain the possible values?

If you (as object type owner) maintain the values, I think an enum will bring benefits for UX and validation of the values.

If the values are maintained using extension mechanisms (e.g., consumers of your object type can extend the set values on their own), I agree, enums don't make sense at all.

file-formats/edot/type/zif_aff_edot_v1.intf.abap

schneidermic0

I have also added some comments and questions

schneidermic0 · 2025-01-31T11:05:04Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">File Cloud Relevancy</p>
+      "! File Cloud Relevancy type
+      not_cloud_relevant_type TYPE abap_bool,


I see for a lot of fields a mismatch in the title ("File Cloud Relevancy"; and its description) and the field name (not_cloud_relevant_type). It would be great if this can be harmonized in general.

However, in this case it seems to me just wrong:

File Cloud Relevancy = "Yes" vs. not_cloud_relevant_type = true

schneidermic0 · 2025-01-31T11:06:37Z

file-formats/edoi/README.md

As mentioned by @Markus1812, all EDOI files should not be part of this PR

EDOI files removed

schneidermic0 · 2025-01-31T11:07:16Z

file-formats/edot/.DS_Store

This file should be deleted

somehow i coudn't see this hidden file but anyhow i replaced whole folder

schneidermic0 · 2025-01-31T11:07:47Z

file-formats/edot/README.md

Example is missing

I think having an example would also help me for the review

Somehow I am unable to generate example, I created class and Transformation but its not clear to me in the report what object to be mentioned as input.

schneidermic0 · 2025-01-31T11:08:54Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">Header</p>
+      "! Header
+      "! $required
+      header              TYPE zif_aff_types_v1=>ty_header_60,


What ABAP language versions does your object type support? "standard", "keyUser", "cloudDevelopment"

Standard & cloudDevelopment

schneidermic0 · 2025-01-31T11:13:22Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">Additional Selection Fields</p>
+      "! Additional selection fields of validation report
+      edocument_sral_configuration TYPE ty_sral_configurations,
+      "! <p class="shorttext">eDocument Type Specific Additional Tables</p>
+      "! eDocument Type Specific Additional Tables
+      edoc_spec_additional_table   TYPE ty_additional_tables,


I am wondering whether it make sense to move these two fields (representing tables/arrays) on top level into ty_main. Maybe, you want to check this also during UX review.

For "general information" it seems to me too detailed.

schneidermic0 · 2025-01-31T11:15:28Z

file-formats/edot/type/zif_aff_edot_v1.intf.abap

+      "! <p class="shorttext">eDocument Type</p>
+      "! eDoc Type
+      "! $required
+      edocument_type               TYPE ty_edoc_type,


So this is the actual object's name also stored in TADIR, right?

If this is the case, please remove it. The object name is stored in the filename like "my_object_name.edot.json"

i506210 added 6 commits January 27, 2025 13:15

Add files for EDOI & EDOT

a8ec8b7

React to abaplint

499d83a

React to abaplint

e1e96b2

Merge branch 'main' into feature/edoi

ecdefa0

React to comments from Michael Schneider

3a2e9b9

[EDOT] Add files for EDOT

a1ec907

i506210 changed the title ~~[EDOTFeature/edot~~ [EDOT] Add new object EDOT Jan 28, 2025

Markus1812 reviewed Jan 28, 2025

View reviewed changes

schneidermic0 requested changes Jan 31, 2025

View reviewed changes

i506210 added 4 commits February 4, 2025 10:49

reaction to comments

6ed5d4c

delete other object files in PR

4c96d92

correction

a7025f3

corrections

bdd762a

i506210 requested a review from schneidermic0 February 6, 2025 09:35

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[EDOT] Add new object EDOT #682

[EDOT] Add new object EDOT #682

i506210 commented Jan 28, 2025

Markus1812 left a comment •

edited

Loading

Markus1812 Jan 28, 2025

i506210 Jan 28, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 5, 2025

Markus1812 Jan 28, 2025

i506210 Jan 28, 2025

Markus1812 Jan 31, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 5, 2025

Markus1812 Jan 28, 2025

i506210 Jan 28, 2025

Markus1812 Jan 28, 2025

i506210 Jan 28, 2025

schneidermic0 Jan 31, 2025

schneidermic0 left a comment

schneidermic0 Jan 31, 2025

i506210 Feb 5, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 5, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 6, 2025

schneidermic0 Jan 31, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 6, 2025

schneidermic0 Jan 31, 2025

i506210 Feb 5, 2025

schneidermic0 Jan 31, 2025

schneidermic0 Jan 31, 2025

	TYPES:
	"! <p class="shorttext">Header for SAJC Object</p>
	"! The header for an application job catalog entry
	BEGIN OF ty_header,
	"! <p class="shorttext">Description</p>
	"! Description of the application job catalog entry
	"! $required
	description TYPE c LENGTH 120,
	"! <p class="shorttext">Original Language</p>
	"! Original language of the application job catalog entry
	"! $required
	original_language TYPE sy-langu,
	"! <p class="shorttext">ABAP Language Version</p>
	"! ABAP language version
	"! $values {@link zif_aff_types_v1.data:co_abap_language_version_cloud}
	"! $default {@link zif_aff_types_v1.data:co_abap_language_version_cloud.standard}
	abap_language_version TYPE zif_aff_types_v1=>ty_abap_language_version_cloud,
	END OF ty_header.

[EDOT] Add new object EDOT #682

Are you sure you want to change the base?

[EDOT] Add new object EDOT #682

Conversation

i506210 commented Jan 28, 2025

Markus1812 left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

schneidermic0 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Markus1812 left a comment •

edited

Loading